業績情報

• Response Time Analysis With Cause-Effect Chain Considering DAG Structure and High-Load Tasks.　　　　　　　　　　　　　　　
  Tomoya Kobayashi; Ryo Okamura; Takuya Azumi
  IEEE Access, 巻：13, 開始ページ：21557, 終了ページ：21568, 2025年01月, ［査読有り］, ［最終著者］
  英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.1109/ACCESS.2025.3536013
  DOI ID：10.1109/ACCESS.2025.3536013, DBLP ID：journals/access/KobayashiOA25
• Work-in-Progress: Multi-Deadline DAG Scheduling Model for Autonomous Driving Systems.　　　　　　　　　　　　　　　
  Atsushi Yano; Takuya Azumi
  RTSS, 開始ページ：451, 終了ページ：454, 2024年12月, ［査読有り］, ［最終著者］
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/RTSS62706.2024.00049
  DOI ID：10.1109/RTSS62706.2024.00049, DBLP ID：conf/rtss/YanoA24
• ROS-lite2: Autonomous-driving Software Platform for Clustered Many-core Processor.　　　　　　　　　　　　　　　
  Yuta Tajima; Shuhei Tsunoda; Takuya Azumi
  IROS, 開始ページ：4578, 終了ページ：4585, 2024年10月, ［査読有り］, ［最終著者］
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/IROS58592.2024.10802569
  DOI ID：10.1109/IROS58592.2024.10802569, DBLP ID：conf/iros/TajimaTA24
• Model-Based Development for Autonomous Driving Software Considering Parallelization.　　　　　　　　　　　　　　　
  Kenshin Obi; Takumi Onozawa; Hiroshi Fujimoto; Takuya Azumi
  29th IEEE International Conference on Emerging Technologies and Factory Automation(ETFA), 開始ページ：1, 終了ページ：8, 2024年09月, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ETFA61755.2024.10711120
  DOI ID：10.1109/ETFA61755.2024.10711120, DBLP ID：conf/etfa/ObiOFA24
• HRMP3+TECS v2: Component Framework for Reliable Multiprocessor Real-time Operating System.　　　　　　　　　　　　　　　
  Yoshitada Takaso; Nao Yoshimura; Hiroshi Oyama 0002; Hiroaki Takada; Takuya Azumi
  Journal of Information Processing, 巻：32, 開始ページ：818, 終了ページ：828, 2024年09月, ［査読有り］, ［最終著者］
  英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.2197/ipsjjip.32.818
  DOI ID：10.2197/ipsjjip.32.818, DBLP ID：journals/jip/TakasoYOTA24
• Preliminary Approach to Parallelizing Autonomous Driving Applications Using High-Performance Many-core Processor.　　　　　　　　　　　　　　　
  Xuankeng He; Takuya Azumi
  30th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications(RTCSA), 開始ページ：134, 終了ページ：135, 2024年08月, ［査読有り］, ［最終著者］
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/RTCSA62462.2024.00030
  DOI ID：10.1109/RTCSA62462.2024.00030, DBLP ID：conf/rtcsa/HeA24
• Parallelized Code Generation from Simulink Models for Event-driven and Timer-driven ROS 2 Nodes.　　　　　　　　　　　　　　　
  Kenshin Obi; Ryo Yoshinaka; Hiroshi Fujimoto; Takuya Azumi
  SEAA, 開始ページ：48, 終了ページ：55, 2024年08月, ［査読有り］
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/SEAA64295.2024.00017
  DOI ID：10.1109/SEAA64295.2024.00017, DBLP ID：conf/euromicro/ObiYFA24
• AUTOSAR AP and ROS 2 Collaboration Framework.　　　　　　　　　　　　　　　
  Ryudai Iwakami; Bo Peng; Hiroyuki Hanyu; Tasuku Ishigooka; Takuya Azumi
  27th Euromicro Conference on Digital System Design(DSD), 開始ページ：319, 終了ページ：326, 2024年08月, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/DSD64264.2024.00050
  DOI ID：10.1109/DSD64264.2024.00050, DBLP ID：conf/dsd/IwakamiPHIA24
• Deadline Miss Early Detection Method for DAG Tasks Considering Variable Execution Time.　　　　　　　　　　　　　　　
  Hayate Toba; Takuya Azumi
  36th Euromicro Conference on Real-Time Systems(ECRTS), 開始ページ：8, 終了ページ：21, 2024年07月, ［査読有り］, ［最終著者］
  Schloss Dagstuhl - Leibniz-Zentrum für Informatik, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.4230/LIPIcs.ECRTS.2024.8
  DOI ID：10.4230/LIPIcs.ECRTS.2024.8, DBLP ID：conf/ecrts/TobaA24
• Point Cloud Automatic Annotation Framework for Autonomous Driving.　　　　　　　　　　　　　　　
  Chaoran Zhao; Bo Peng; Takuya Azumi
  IEEE Intelligent Vehicles Symposium(IV), 開始ページ：3063, 終了ページ：3070, 2024年06月, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/IV55156.2024.10588577
  DOI ID：10.1109/IV55156.2024.10588577, DBLP ID：conf/ivs/ZhaoPA24
• Energy Consumption Prediction Framework in Model-based Development for Edge Devices
  Yue Hou; Takuya Azumi
  2024 IEEE 3rd Real-Time and Intelligent Edge Computing Workshop (RAGE), 開始ページ：1, 終了ページ：6, 2024年05月, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/rage62451.2024.00012
  DOI ID：10.1109/rage62451.2024.00012
• Federated Learning Platform on Embedded Many-core Processor with Flower
  Masahiro Hasumi; Takuya Azumi
  2024 IEEE 3rd Real-Time and Intelligent Edge Computing Workshop (RAGE), 開始ページ：1, 終了ページ：6, 2024年05月, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/rage62451.2024.00015
  DOI ID：10.1109/rage62451.2024.00015
• 3D Point Cloud Object Detection on Edge Devices for Split Computing
  Taisuke Noguchi; Takuya Azumi
  2024 IEEE 3rd Real-Time and Intelligent Edge Computing Workshop (RAGE), 開始ページ：1, 終了ページ：6, 2024年05月, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/rage62451.2024.00009
  DOI ID：10.1109/rage62451.2024.00009
• TECS/Rust: Memory-safe Component Framework for Embedded Systems.　　　　　　　　　　　　　　　
  Nao Yoshimura; Hiroshi Oyama 0002; Takuya Azumi
  27th IEEE International Symposium on Real-Time Distributed Computing(ISORC), 開始ページ：1, 終了ページ：11, 2024年05月, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ISORC61049.2024.10551370
  DOI ID：10.1109/ISORC61049.2024.10551370, DBLP ID：conf/isorc/Yoshimura0A24
• Work-in-Progress: Federated and Bundled-Based DAG Scheduling.　　　　　　　　　　　　　　　
  Tomoya Kobayashi; Takuya Azumi
  IEEE Real-Time Systems Symposium(RTSS), 開始ページ：443, 終了ページ：446, 2023年12月, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/RTSS59052.2023.00048
  DOI ID：10.1109/RTSS59052.2023.00048, DBLP ID：conf/rtss/KobayashiA23
• Model-based Development for ROS 2-based Autonomous-driving Software.　　　　　　　　　　　　　　　
  Takumi Onozawa; Hiroshi Fujimoto; Takuya Azumi
  22nd IEEE International Conference on Trust, Security and Privacy in Computing and Communications(TrustCom), 開始ページ：2511, 終了ページ：2519, 2023年11月, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/TrustCom60117.2023.00352
  DOI ID：10.1109/TrustCom60117.2023.00352, DBLP ID：conf/trustcom/OnozawaFA23
• DAG Scheduling for Clustered Many-Core Processor Considering Execution-Time-Reduction Effectiveness.　　　　　　　　　　　　　　　
  Yutaro Nozaki; Takuya Azumi
  27th IEEE/ACM International Symposium on Distributed Simulation and Real Time Applications(DS-RT), 開始ページ：101, 終了ページ：104, 2023年10月, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/DS-RT58998.2023.00022
  DOI ID：10.1109/DS-RT58998.2023.00022, DBLP ID：conf/dsrt/NozakiA23
• Estimation of Deadline Miss Rate for DAG Mixed Timer-Driven and Event-Driven Nodes.　　　　　　　　　　　　　　　
  Daichi Yamazaki; Takuya Azumi
  27th IEEE/ACM International Symposium on Distributed Simulation and Real Time Applications(DS-RT), 開始ページ：20, 終了ページ：29, 2023年10月, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/DS-RT58998.2023.00012
  DOI ID：10.1109/DS-RT58998.2023.00012, DBLP ID：conf/dsrt/YamazakiA23, 共同研究・競争的資金等ID：44548618
• TILDE: Topic-Tracking Infrastructure for Dynamic Message Latency and Deadline Evaluator for ROS 2 Application.　　　　　　　　　　　　　　　
  Xuankeng He; Hiromi Sato; Yoshikazu Okumura; Takuya Azumi
  27th IEEE/ACM International Symposium on Distributed Simulation and Real Time Applications(DS-RT), 開始ページ：1, 終了ページ：9, 2023年10月, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/DS-RT58998.2023.00010
  DOI ID：10.1109/DS-RT58998.2023.00010, DBLP ID：conf/dsrt/HeSOA23
• Automated Testing Framework for Embedded Component Systems.　　　　　　　　　　　　　　　
  Hinata Tomimori; Hiroshi Oyama 0002; Takuya Azumi
  26th IEEE International Symposium on Real-Time Distributed Computing(ISORC), 開始ページ：176, 終了ページ：183, 2023年05月, ［査読有り］, ［最終著者］
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ISORC58943.2023.00032
  DOI ID：10.1109/ISORC58943.2023.00032, DBLP ID：conf/isorc/TomimoriOA23
• HRMP3+TECS: Component Framework for Multiprocessor Real-time Operating System with Memory Protection.　　　　　　　　　　　　　　　
  Yoshitada Takaso; Hiroshi Oyama 0002; Hiroaki Takada; Takuya Azumi
  26th IEEE International Symposium on Real-Time Distributed Computing(ISORC), 開始ページ：86, 終了ページ：96, 2023年05月, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ISORC58943.2023.00021
  DOI ID：10.1109/ISORC58943.2023.00021, DBLP ID：conf/isorc/TakasoOTA23
• RD-Gen: Random DAG Generator Considering Multi-rate Applications for Reproducible Scheduling Evaluation.　　　　　　　　　　　　　　　
  Atsushi Yano; Takuya Azumi
  26th IEEE International Symposium on Real-Time Distributed Computing(ISORC), 開始ページ：21, 終了ページ：31, 2023年05月, ［査読有り］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ISORC58943.2023.00015
  DOI ID：10.1109/ISORC58943.2023.00015, DBLP ID：conf/isorc/YanoA23
• Deadline Miss Early Detection Method for Mixed Timer-Driven and Event-Driven DAG Tasks.　　　　　　　　　　　　　　　
  Atsushi Yano; Takuya Azumi
  IEEE Access, 巻：11, 開始ページ：22187, 終了ページ：22200, 2023年03月, ［最終著者］
  英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.1109/ACCESS.2023.3251359
  DOI ID：10.1109/ACCESS.2023.3251359, DBLP ID：journals/access/YanoA23
• Performance Evaluation Framework for Arbitrary Nodes of Autonomous-driving Systems.　　　　　　　　　　　　　　　
  Yuta Tajima; Tatsuya Miki; Takuya Azumi
  22nd IEEE International Conference on Trust, Security and Privacy in Computing and Communications(TrustCom), 開始ページ：2520, 終了ページ：2529, 2023年, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/TrustCom60117.2023.00353
  DOI ID：10.1109/TrustCom60117.2023.00353, DBLP ID：conf/trustcom/TajimaMA23
• Scheduling Performance Evaluation Framework for ROS 2 Applications.　　　　　　　　　　　　　　　
  Bo Peng; Atsushi Hasegawa; Takuya Azumi
  HPCC/DSS/SmartCity/DependSys, 開始ページ：2031, 終了ページ：2038, 2022年12月, ［最終著者］
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/HPCC-DSS-SmartCity-DependSys57074.2022.00302
  DOI ID：10.1109/HPCC-DSS-SmartCity-DependSys57074.2022.00302, DBLP ID：conf/hpcc/PengHA22
• Component Framework for Multiprocessor Real-Time Operating Systems.　　　　　　　　　　　　　　　
  Yoshitada Takaso; Hiroshi Oyama 0002; Takuya Azumi
  EUC, 開始ページ：112, 終了ページ：119, 2022年12月, ［最終著者］
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/EUC57774.2022.00026
  DOI ID：10.1109/EUC57774.2022.00026, DBLP ID：conf/euc/TakasoOA22
• CARET: Chain-Aware ROS 2 Evaluation Tool.　　　　　　　　　　　　　　　
  Takahisa Kuboichi; Atsushi Hasegawa; Bo Peng; Keita Miura; Kenji Funaoka; Shinpei Kato; Takuya Azumi
  EUC, 開始ページ：1, 終了ページ：8, 2022年12月, ［最終著者］
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/EUC57774.2022.00010
  DOI ID：10.1109/EUC57774.2022.00010, DBLP ID：conf/euc/KuboichiHPMFKA22
• New Trends in Intelligent Software Methodologies, Tools and Techniques - Proceedings of the 21st International Conference on New Trends in Intelligent Software Methodologies, Tools and Techniques, SoMeT 2022, Kitakyushu, Japan, 20-22 September, 2022.　　　　　　　　　　　　　　　
  SoMeT, 巻：355, 2022年09月, ［査読有り］, ［最終著者］
  IOS Press, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.3233/FAIA355
  DOI ID：10.3233/FAIA355, DBLP ID：conf/somet/2022
• Communication Overhead Schema Independent of Libraries for Software/Hardware Interface.　　　　　　　　　　　　　　　
  Yutaro Kobayashi; Hiroshi Fujimoto; Takuya Azumi
  SoMeT, 開始ページ：30, 終了ページ：42, 2022年09月, ［査読有り］, ［最終著者］
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.3233/FAIA220236
  DOI ID：10.3233/FAIA220236, DBLP ID：conf/somet/KobayashiFA22
• Self-Driving Software Benchmark for Model-Based Development.　　　　　　　　　　　　　　　
  Takumi Onozawa; Takuya Azumi
  SoMeT, 開始ページ：3, 終了ページ：17, 2022年09月, ［最終著者］
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.3233/FAIA220234
  DOI ID：10.3233/FAIA220234, DBLP ID：conf/somet/OnozawaA22
• DAG Scheduling Considering Parallel Execution for High-Load Processing on Clustered Many-core Processors.　　　　　　　　　　　　　　　
  Ryo Okamura; Takuya Azumi
  DS-RT, 開始ページ：67, 終了ページ：76, 2022年09月, ［最終著者］
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/DS-RT55542.2022.9932084
  DOI ID：10.1109/DS-RT55542.2022.9932084, DBLP ID：conf/dsrt/OkamuraA22
• CQGA-HEFT: Q-learning-based DAG Scheduling Algorithm Using Genetic Algorithm in Clustered Many-core Platform.　　　　　　　　　　　　　　　
  Atsushi Yano; Takuya Azumi
  Journal of Information Processing, 巻：30, 開始ページ：659, 終了ページ：668, 2022年09月, ［最終著者］
  英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.2197/ipsjjip.30.659
  DOI ID：10.2197/ipsjjip.30.659, DBLP ID：journals/jip/YanoA22
• LET Paradigm Scheduling Algorithm Considering Parallel Processing on Clustered Many-core Processor.　　　　　　　　　　　　　　　
  Atsushi Yano; Shingo Igarashi; Takuya Azumi
  Journal of Information Processing, 巻：30, 開始ページ：646, 終了ページ：658, 2022年09月, ［最終著者］
  英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.2197/ipsjjip.30.646
  DOI ID：10.2197/ipsjjip.30.646, DBLP ID：journals/jip/YanoIA22
• Exploring the Performance of Deep Neural Networks on Embedded Many-Core Processors.　　　　　　　　　　　　　　　
  Takuma Yabe; Takuya Azumi
  ICCPS, 開始ページ：193, 終了ページ：202, 2022年05月, ［査読有り］, ［最終著者］
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ICCPS54341.2022.00024
  DOI ID：10.1109/ICCPS54341.2022.00024, DBLP ID：conf/iccps/YabeA22
• Autoware_Perf: A tracing and performance analysis framework for ROS 2 applications.　　　　　　　　　　　　　　　
  Zihang Li; Atsushi Hasegawa; Takuya Azumi
  Journal of Systems Architecture, 巻：123, 開始ページ：102341, 終了ページ：102341, 2022年02月, ［最終著者］
  英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.1016/j.sysarc.2021.102341
  DOI ID：10.1016/j.sysarc.2021.102341, DBLP ID：journals/jsa/LiHA22
• Mapping Method Usable with Clustered Many-core Platforms for Simulink Model
  Yutaro Kobayashi; Kentaro Honda; Sasuga Kojima; Hiroshi Fujimoto; Masato Edahiro; Takuya Azumi
  Journal of Information Processing, 巻：30, 開始ページ：141, 終了ページ：150, 2022年02月, ［査読有り］, ［最終著者］
  Information Processing Society of Japan, 英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.2197/ipsjjip.30.141
  DOI ID：10.2197/ipsjjip.30.141, eISSN：1882-6652
• Dynamically Interchangeable Framework for Component Behavior of Embedded Component Systems.　　　　　　　　　　　　　　　
  Ryota Shimomura; Hiroshi Oyama; Takuya Azumi
  19th IEEE International Conference on Embedded and Ubiquitous Computing(EUC), 開始ページ：15, 終了ページ：22, 2021年10月, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/EUC53437.2021.00012
  DOI ID：10.1109/EUC53437.2021.00012, DBLP ID：conf/euc/ShimomuraOA21
• Contention-Free Scheduling Algorithm Using LET Paradigm for Clustered Many-core Processor　　　　　　　　　　　　　　　
  Atsushi Yano; Shingo Igarashi; Takuya Azumi
  Proceedings of the 2021 IEEE/ACM 25th International Symposium on Distributed Simulation and Real Time Applications, DS-RT 2021, 2021年09月
  Self-driving systems require multi-/many-core platforms with high computing power and low power consumption. However, for hard real-time applications, multiple demands on shared resources can impede real-time performance. Therefore, making the timing of memory access deterministic is important. The logical execution time (LET) paradigm has gained attention as a means to achieve this purpose. However, this approach lacks scalability owing to the overhead caused because the LET paradigm is set longer than the actual execution time of the task. This paper proposes a theoretical scheduling method for a model applying the LET paradigm to the directed acyclic graph (DAG) nodes for a multi-/many-core platform. The proposed method considers communication timing and generates a schedule that does not cause communication contentions. In addition, the proposed method performs a parallel calculation of tasks to deal with the overhead caused by adopting the LET paradigm.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/DS-RT52167.2021.9576147
  DOI ID：10.1109/DS-RT52167.2021.9576147, SCOPUS ID：85119097396
• Federated Scheduling in Clustered Many-core Processors　　　　　　　　　　　　　　　
  Ryotaro Koike; Takuya Azumi
  Proceedings of the 2021 IEEE/ACM 25th International Symposium on Distributed Simulation and Real Time Applications, DS-RT 2021, 2021年09月, ［査読有り］, ［最終著者］
  High-performance embedded systems, such as self-driving systems require platforms that reduce power consumption and perform high-performance processing. As satisfying both requirements, multi-tmany-core processors are attracting attention. This paper focuses on a clustered many-core processors represented by Kalray MPPA. Clustered many-core processors regard multiple cores as a single cluster, with each cluster having private memory. A bus within each cluster performs communication. A network-on-chip (NoC) is used between clusters. These two communication speeds are different. However, it is inappropriate to always assume that the worst calculation is the slow NoC communication speed. This paper discusses a method to improve worst-case communication time estimation when applying federated scheduling on clustered many-core processors with routes at a different speed. Furthermore, this paper investigates an approach to assign dedicated cores in multiple clusters to reduce the number of required cores.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/DS-RT52167.2021.9576150
  DOI ID：10.1109/DS-RT52167.2021.9576150, SCOPUS ID：85119092208
• CoSAM: Co-Simulation Framework for ROS-based Self-driving Systems and MATLAB/Simulink
  Keita Miura; Shota Tokunaga; Yuki Horita; Yasuhiro Oda; Takuya Azumi
  Journal of Information Processing, 巻：29, 開始ページ：227, 終了ページ：235, 2021年03月, ［最終著者］
  Information Processing Society of Japan, 研究論文（学術雑誌）
  DOI：https://doi.org/10.2197/ipsjjip.29.227
  DOI ID：10.2197/ipsjjip.29.227, eISSN：1882-6652
• Work-in-Progress: Reinforcement Learning-Based DAG Scheduling Algorithm in Clustered Many-Core Platform.　　　　　　　　　　　　　　　
  Atsushi Yano; Takuya Azumi
  42nd IEEE Real-Time Systems Symposium(RTSS), 開始ページ：548, 終了ページ：551, 2021年
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/RTSS52674.2021.00062
  DOI ID：10.1109/RTSS52674.2021.00062, DBLP ID：conf/rtss/YanoA21
• Accurate Contention-aware Scheduling Method on Clustered Many-core Platform
  Shingo Igarashi; Takuro Fukunaga; Takuya Azumi
  Journal of Information Processing, 巻：29, 開始ページ：216, 終了ページ：226, 2021年, ［最終著者］
  Information Processing Society of Japan, 研究論文（学術雑誌）
  DOI：https://doi.org/10.2197/ipsjjip.29.216
  DOI ID：10.2197/ipsjjip.29.216, eISSN：1882-6652
• Contention-Free Scheduling for Clustered Many-Core Platform　　　　　　　　　　　　　　　
  Ryotaro Koike; Takuro Fukunaga; Shingo Igarashi; Takuya Azumi
  2020 IEEE International Conference on Embedded Software and Systems (ICESS), 開始ページ：1, 終了ページ：8, 2020年12月, ［最終著者］
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/icess49830.2020.9301600
  DOI ID：10.1109/icess49830.2020.9301600, DBLP ID：conf/icess/KoikeFIA20
• Exploring Real-Time Executor on ROS 2
  Yuqing Yang; Takuya Azumi
  2020 IEEE International Conference on Embedded Software and Systems (ICESS), 2020年12月
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/icess49830.2020.9301530
  DOI ID：10.1109/icess49830.2020.9301530
• Estimation Method Considering OS Overheads for Embedded Many-Core Platform
  Kentaro Honda; Hiroshi Fujimoto; Takuya Azumi
  2020 IEEE 18th International Conference on Embedded and Ubiquitous Computing (EUC), 2020年12月, ［最終著者］
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/euc50751.2020.00011
  DOI ID：10.1109/euc50751.2020.00011
• Unit Testing Framework for Embedded Component Systems
  Shuichiro Morisaki; Seito Shirata; Hiroshi Oyama; Takuya Azumi
  2020 IEEE 18th International Conference on Embedded and Ubiquitous Computing (EUC), 2020年12月, ［最終著者］
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/euc50751.2020.00013
  DOI ID：10.1109/euc50751.2020.00013
• Converting Driving Scenario Framework for Testing Self-Driving Systems
  Keita Miura; Takuya Azumi
  2020 IEEE 18th International Conference on Embedded and Ubiquitous Computing (EUC), 2020年12月, ［最終著者］
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/euc50751.2020.00015
  DOI ID：10.1109/euc50751.2020.00015
• Case Study: ECHONET Lite Applications based on Embedded Component Systems　　　　　　　　　　　　　　　
  Jieying Jiang; Hiroshi Oyama; Hiroaki Nagashima; Takuya Azumi
  Asia Pacific Conference on Robot IoT System Development and Platform (APRIS2020), 2020年11月, ［招待有り］
  英語, 研究論文（国際会議プロシーディングス）
• Dynamic Linking Method for an Embedded Component System　　　　　　　　　　　　　　　
  Katsuya Yamauchi; Tomoaki Kawada; Hiroshi Oyama; Takuya Azumi; Hiroaki Takada
  Asia Pacific Conference on Robot IoT System Development and Platform (APRIS2020), 2020年11月, ［査読有り］
  研究論文（国際会議プロシーディングス）
• ROS-lite: ROS Framework for NoC-Based Embedded Many-Core Platform
  Takuya Azumi; Yuya Maruyama; Shinpei Kato
  2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020年10月, ［査読有り］, ［最終著者］
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/iros45743.2020.9340977
  DOI ID：10.1109/iros45743.2020.9340977
• Model-Based Development Considering Self-Driving Systems for Many-Core Processors
  Ryo Yoshinaka; Takuya Azumi
  2020 25th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA), 2020年09月, ［査読有り］, ［最終著者］
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/etfa46521.2020.9212016
  DOI ID：10.1109/etfa46521.2020.9212016
• Heuristic Contention-Free Scheduling Algorithm for Multi-core Processor using LET Model
  Shingo Igarashi; Tasuku Ishigooka; Tatsuya Horiguchi; Ryotaro Koike; Takuya Azumi
  2020 IEEE/ACM 24th International Symposium on Distributed Simulation and Real Time Applications (DS-RT), 2020年09月, ［査読有り］, ［最終著者］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ds-rt50469.2020.9213582
  DOI ID：10.1109/ds-rt50469.2020.9213582
• Accurate Contention Estimate Scheduling Method Using Multiple Clusters of Many-core Platform for Applications Requiring Parallel Computations　　　　　　　　　　　　　　　
  Shingo Igarashi; Yuto Kitagawa; Takuro Fukunaga; Takuya Azumi
  Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), 開始ページ：67, 終了ページ：71, 2020年03月, ［査読有り］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/PDP50117.2020.00017
  DOI ID：10.1109/PDP50117.2020.00017, DBLP ID：conf/pdp/IgarashiKFA20
• Mapping Method of MATLAB/Simulink Model for Embedded Many-Core Platform　　　　　　　　　　　　　　　
  Kentaro Honda; Sasuga Kojima; Hiroshi Fujimoto; Masato Edahiro; Takuya Azumi
  Proceedings - 2020 28th Euromicro International Conference on Parallel, Distributed and Network-Based Processing, PDP 2020, 開始ページ：182, 終了ページ：186, 2020年03月, ［査読有り］, ［最終著者］
  Multi-/many-core processors are being increasingly used to reduce power consumption and improve performance. In addition, the use of Model-Based Development for embedded systems has been increasing. Relative to these trends, Model-Based Parallelizer (MBP) has an essential role in parallelizing applications (i.e., Simulink blocks) at the model level. MBP maps Simulink blocks to cores using various types of information such as block characteristics, a C code, and the multi-/many-core hardware implementation. However, MBP does not consider many-core hardware with cluster structures. This paper proposes an algorithm that decides on core allocations by considering cluster structures. The proposed algorithm combines two other algorithms: one algorithm uses the core allocation of MBP and path analysis at the cluster-level and considers the influence of communication contention to decide on cluster allocations, and the other algorithm uses the results of MBP and remaps cluster allocations. The proposed algorithm produces better results than its component algorithms could separately. Evaluations demonstrate that the proposed algorithm obtained the better results than the existing method in terms of execution time on random and real models.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/PDP50117.2020.00034
  DOI ID：10.1109/PDP50117.2020.00034, SCOPUS ID：85085489282
• Work in progress: Considering heuristic scheduling for NoC-Based clustered many-core processor using LET model　　　　　　　　　　　　　　　
  Shingo Igarashi; Takuya Azumi
  Proceedings - Real-Time Systems Symposium, 巻：2019-, 開始ページ：516, 終了ページ：519, 2019年12月
  Embedded systems such as self-driving systems and advanced driver-assistance systems (ADAS) require a computing platform with high computing power and low power consumption. Many-core platforms satisfy exactly these requirements. However, for hard real-time applications, multiple demands on shared resources can hinder real-time performance. Memory is among the resources that can most dramatically impair desired performance. In particular, it is important that the timing for accessing the memory is deterministic. As a method for realizing this, the Logical Execution Time (LET) paradigm is currently attracting attention. Research on the LET model is mainly conducted for control systems such as engines. However, scalability is required for application to autonomous driving systems, and high-speed task scheduling and mapping to the core must be considered. This paper summarizes the scheduling issues of hard real-time applications on a many-core processor using the LET model and discusses possible approaches.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/RTSS46320.2019.00053
  DOI ID：10.1109/RTSS46320.2019.00053, ISSN：1052-8725, SCOPUS ID：85083215228
• Autoware toolbox: MATLAB/Simulink benchmark suite for ROS-based self-driving software platform　　　　　　　　　　　　　　　
  Keita Miura; Shota Tokunaga; Noriyuki Ota; Yoshiharu Tange; Takuya Azumi
  Proceedings of the International Workshop on Rapid System Prototyping, 開始ページ：8, 終了ページ：14, 2019年10月
  This paper describes a MATLAB/Simulink benchmark suite for an open-source self-driving system based on Robot Operating System (ROS). In recent years, self-driving systems have been developed worldwide, and the technology has been making remarkable progress. One approach to the development of the self-driving systems is the utilization of ROS which is an open-source middleware framework used for developing robot applications. On the other hand, the popular approach in the automotive industry is the utilization of MATLAB/Simulink which is the software for modeling, simulating, and analyzing.MATLAB/Simulink provides an interface between ROS and MATLAB/Simulink that enables to create functionalities of ROS-based robots in MATLAB/Simulink. However, it has not been fully utilized in the development of the self-driving systems yet because there are not enough open-source models for self-driving. Thus the co-development is difficult. Therefore, we provide a MATLAB/Simulink benchmark suite for a ROS-based self-driving system called Autoware. Autoware is popular opensource software that provides a complete set of self-driving modules. The provided benchmark contains MATLAB/Simulink models. They help to design the ROS-based self-driving systems using MATLAB/ Simulink. Moreover, we investigated other benchmarks for self-driving and indicated that this is the first work aiming to the ROS-based self-driving systems with MATLAB/Simulink.
  IEEE Computer Society, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1145/3339985.3358494
  DOI ID：10.1145/3339985.3358494, ISSN：1074-6005, SCOPUS ID：85077209879
• Multi-rate DAG Scheduling Considering Communication Contention for NoC-based Embedded Many-core Processor　　　　　　　　　　　　　　　
  Shingo Igarashi; Yuto Kitagawa; Tasuku Ishigooka; Tatsuya Horiguchi; Takuya Azumi
  Proceedings - 2019 IEEE/ACM 23rd International Symposium on Distributed Simulation and Real Time Applications, DS-RT 2019, 2019年10月
  Computing platforms for embedded systems are increasingly being transformed into multi/many-core platforms because embedded systems have become extensive, complex, and automated. In the case of an autonomous driving system, various applications are simultaneously running, and low power consumption and large-scale calculation are required. Many-core processors with a multiple instruction, multiple data (MIMD) architecture can meet these requirements. This paper proposes a scheduling algorithm for an automotive driving system expressed in a directed acyclic graph (DAG) and we use Kalray MPPA-256 as the target many-core processor. On the basis of the architecture of Kalray MPPA-256, task processing that requires large-scale calculation and intercore communication is performed while avoiding communication contention by using a proposed grouping computational resource. In addition, we propose a scheduling method for a multi-rate DAG which is a DAG with multiple periods. This method generates a DAG task in a hyperperiod and schedules the DAG with dependency on tasks that have been released closely. The formulas for prioritization and processor selection are proposed for various generated tasks in a hyperperiod. Evaluation results show that the proposed algorithm is superior to existing DAG scheduling algorithms with regard to schedulability and deadline miss ratio.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/DS-RT47707.2019.8958696
  DOI ID：10.1109/DS-RT47707.2019.8958696, SCOPUS ID：85079075931
• Performance Estimation for Many-core Processor in Model-Based Development　　　　　　　　　　　　　　　
  Kentaro Honda; Takuya Azumi
  2019 8th Mediterranean Conference on Embedded Computing, MECO 2019 - Proceedings, 2019年06月
  Recently, embedded systems such as automotive applications, have become large and more complicated. In addition, the use of Model-Based Development (MBD) such as MATLAB/Simulink for embedded systems becomes increasingly common. Various processors used in embedded systems have evolved, and many-core processors have appeared. Many-core processors are high performance and low consumption, and is beginning to be introduced in embedded systems that require sufficient computational resources such as self-driving systems. Nowadays, software developers should select actual optimum processors that meet the target performance. However, examining and considering a large number of actual processors are not efficient. Relative to these trends, a tool that can easily convey hardware information to the software is now available and is expected to be used in parallelizing compilers, parallelizing tools, and others. In this paper, we utilized such tools, and estimated the software performance at the instruction level. We also evaluated the performance result, such as the instruction set of the many-core processors and influence of contention.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/MECO.2019.8760061
  DOI ID：10.1109/MECO.2019.8760061, SCOPUS ID：85073887708
• MATLAB/simulink benchmark suite for ros-based self-driving software platform　　　　　　　　　　　　　　　
  Shota Tokunaga; Keita Miura; Takuya Azumi
  Proceedings - 2019 IEEE 22nd International Symposium on Real-Time Distributed Computing, ISORC 2019, 開始ページ：83, 終了ページ：84, 2019年05月
  In recent years, self-driving systems have been developed worldwide, and the technology has been making remarkable progress. One approach to the development of the autonomous vehicle is using ROS which is an open-source middleware framework used for developing robot applications. On the other hand, the popular approach in the automotive industry is using MATLAB/Simulink which is the software for modeling, simulating, and analyzing. MATLAB/Simulink has an interface connecting ROS and MATLAB/Simulink. However, it is not used much in the development of self-driving systems because there are not enough samples for the self-driving systems. Therefore, we provide a MATLAB/Simulink benchmark suite for a ROS-based self-driving system called Autoware. Autoware provides an abundant set of self-driving modules and enables to simulate and operate the autonomous vehicle. The provided benchmark is a set of MATLAB code and Simulink model samples. They assist to design the self-driving systems using MATLAB/Simulink.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ISORC.2019.00023
  DOI ID：10.1109/ISORC.2019.00023, SCOPUS ID：85070412326
• Demo Abstract: MATLAB/Simulink Benchmark suite for ROS-based self-driving system　　　　　　　　　　　　　　　
  Shota Tokunaga; Noriyuki Ota; Keita Miura; Takuya Azumi
  ICCPS 2019 - Proceedings of the 2019 ACM/IEEE International Conference on Cyber-Physical Systems, 開始ページ：308, 終了ページ：309, 2019年04月
  This paper proposes a MATLAB/Simulink benchmark suite for an open-source self-driving system based on Robot Operating System (ROS). In recent years, self-driving systems have been developed around the world. One approach to the development of self-driving systems is the utilization of ROS which is an open-source middle-ware framework used in the development of robot applications. On the other hand, the popular approach in the automotive industry is the utilization of MATLAB/Simulink which is software for modeling, simulating, and analyzing. MATLAB/Simulink provides an interface between ROS and MATLAB/Simulink that enables to create functionalities of ROS-based robots in MATLAB/Simulink. However, it is not been fully utilized in the development of self-driving systems yet because there are not enough samples for self-driving, and it is difficult for developers to adopt co-development. Therefore, we provide a MATLAB/Simulink benchmark suite for a ROS-based self-driving system called Autoware. Autoware is popular open-source software that provides a complete set of self-driving modules. The provided benchmark contains MATLAB/Simulink samples available in Autoware. They help to design ROS-based self-driving systems using MATLAB/Simulink.
  Association for Computing Machinery, Inc, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1145/3302509.3313315
  DOI ID：10.1145/3302509.3313315, SCOPUS ID：85066607395
• IDF-Autoware: Integrated development framework for ROS-based self-driving systems using MATLAB/Simulink　　　　　　　　　　　　　　　
  Shota Tokunaga; Yuki Horita; Yasuhiro Oda; Takuya Azumi
  OpenAccess Series in Informatics, 巻：68, 2019年03月
  This paper proposes an integrated development framework that enables co-simulation and operation of a Robot Operating System (ROS)-based self-driving system using MATLAB/Simulink (IDF-Autoware). The management of self-driving systems is becoming more complex as the development of self-driving technology progresses. One approach to the development of self-driving systems is the use of ROS
  however, the system used in the automotive industry is typically designed using MATLAB/Simulink, which can simulate and evaluate the models used for self-driving. These models are incompatible with ROS-based systems. To allow the two to be used in tandem, it is necessary to rewrite the C++ code and incorporate them into the ROS-based system, which makes development inefficient. Therefore, the proposed framework allows models created using MATLAB/Simulink to be used in a ROS-based self-driving system, thereby improving development efficiency. Furthermore, our evaluations of the proposed framework demonstrated its practical potential.
  Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.4230/OASIcs.ASD.2019.3
  DOI ID：10.4230/OASIcs.ASD.2019.3, ISSN：2190-6807, SCOPUS ID：85064130563
• ROSCH: Real-time scheduling framework for ROS　　　　　　　　　　　　　　　
  Yukihiro Saito; Futoshi Sato; Takuya Azumi; Shinpei Kato; Nobuhiko Nishio
  Proceedings - 2018 IEEE 24th International Conference on Embedded and Real-Time Computing Systems and Applications, RTCSA 2018, 開始ページ：52, 終了ページ：58, 2019年01月
  This paper presents a real-time scheduling framework for the robot operating system (ROS) called ROSCH. ROS is an open-source software platform and a meta-operating system designed for robots. ROS provides extensive development libraries and has been widely used for autonomous driving systems. However, ROS does not guarantee real-time performance
  hence, a ROS-based autonomous driving car could cause a traffic accident. Therefore, ROSCH comprises three functionalities that do not exist in the ROS to guarantee real-time performance: (1) a synchronization system, (2) fixed-priority based directed acyclic graph (DAG) scheduling framework, and (3) fail-safe function. In particular, the synchronization system guarantees that the timestamp gaps between sensor measurements will be less than or equal to the calculated value. The fixed-priority based DAG scheduling framework guarantees that end-to-end latency is less than or equal to an estimated value. Operating both mechanisms simultaneously guarantees the final output topic frequency. The fail-safe functionality provides a danger avoidance action at a deadline miss. In addition, these functionalities are designed to be compatible with legacy software. No previous method that guarantees real-time operation provides a comparable range of features while maintaining the compatibility with the existing software. Experimental results showed that the time differences between sensor timestamps remained below a calculated value. Moreover, correctly scheduled tasks incur an end-to-end latency that is within an estimated value. In addition, the throughput of the end node is achieved. Finally, the fail-safe functionality overhead is 45 μsec on an average, which is 0.004% of the average execution time for one node.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/RTCSA.2018.00015
  DOI ID：10.1109/RTCSA.2018.00015, SCOPUS ID：85061776076
• Towards Heterogeneous Computing Platforms for Autonomous Driving.　　　　　　　　　　　　　　　
  Hiroyuki Chishiro; Kazutoshi Suito; Tsutomu Ito; Seiya Maeda; Takuya Azumi; Kenji Funaoka; Shinpei Kato
  15th IEEE International Conference on Embedded Software and Systems(ICESS), 開始ページ：1, 終了ページ：8, 2019年
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ICESS.2019.8782446
  DOI ID：10.1109/ICESS.2019.8782446, DBLP ID：conf/icess/ChishiroSIMAFK19
• Resource Manager for Scalable Performance in ROS Distributed Environments.　　　　　　　　　　　　　　　
  Daisuke Fukutomi; Takuya Azumi; Shinpei Kato; Nobuhiko Nishio
  Design(DATE), 開始ページ：1088, 終了ページ：1093, 2019年
  IEEE, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.23919/DATE.2019.8715030
  DOI ID：10.23919/DATE.2019.8715030, DBLP ID：conf/date/FukutomiAKN19
• Anomaly Prediction Based on Machine Learning for Memory-Constrained Devices　　　　　　　　　　　　　　　
  Yuto Kitagawa; Tasuku Ishigooka; Takuya Azumi
  IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, 巻：E102-D, 号：9, 開始ページ：1797, 終了ページ：1807, 2019年, ［査読有り］
  英語, 研究論文（学術雑誌）
• DAG Scheduling Algorithm for a Cluster-Based Many-Core Architecture　　　　　　　　　　　　　　　
  Yuto Kitagawa; Tasuku Ishigooka; Takuya Azumi
  Proceedings - 16th International Conference on Embedded and Ubiquitous Computing, EUC 2018, 開始ページ：150, 終了ページ：157, 2018年12月
  This paper proposes a directed acyclic graph (DAG) scheduling algorithm for cluster-based many-core architecture. Most of DAG scheduling methods that consider multiple processors and communication delays use a heuristic approach because it is difficult to shorten a schedule length (i.e.,makespan). Unfortunately, existing heuristic algorithms do not consider tasks that require a large number of computational resources. Such tasks typically need to be offloaded to a large-scale computational resource such as a many-core system or graphics processing unit (GPU). Therefore, we propose a DAG scheduling algorithm that uses a cluster-based many-core architecture to offload such computation
  here, we use Kalray MPPA-256 as the many-core architecture. Our algorithm divides many-core computational resources for such tasks. Comparing our results with existing algorithms, we succeeded in shortening makespan with many types of DAGs while also considering Amdahl's law. We also compared the deadline miss ratio and execution time of our algorithm to existing algorithms, and show that the proposed algorithm is superior to the existing algorithms for any evaluation metrics.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/EUC.2018.00030
  DOI ID：10.1109/EUC.2018.00030, SCOPUS ID：85061449253
• Runtime Component Information on Embedded Component Systems　　　　　　　　　　　　　　　
  Seito Shirata; Hiroshi Oyama; Takuya Azumi
  Proceedings - 16th International Conference on Embedded and Ubiquitous Computing, EUC 2018, 開始ページ：166, 終了ページ：173, 2018年12月
  To efficiently build large embedded software systems, dividing the embedded software into components and subsystems and enhance reusability by converting these into distinct parts are essential. Component-Based Development (CBD) is used to reduce costs and improve development efficiency and can be applied to reusable software development. However, CBD systems generally do not support obtaining runtime component information about interfaces and variables generated at runtime, which makes testing for, verifying, and fixing bugs difficult. To deal with this issue, we propose the component framework for obtaining information about component-based systems. This makes it possible to obtain static information on generated components, including interfaces, and runtime component information, including component states and component variables. This makes it an invaluable tool for discovering mistakes and bugs. In addition, the proposed framework was designed with flexibility and efficiency in mind.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/EUC.2018.00032
  DOI ID：10.1109/EUC.2018.00032, SCOPUS ID：85061447059
• GPUhd: Augmenting YARN with GPU Resource Management　　　　　　　　　　　　　　　
  Daisuke Fukutomi; Yuki Iida; Takuya Azumi; Shinpei Kato; Nobuhiko Nishio
  ACM International Conference Proceeding Series, 開始ページ：127, 終了ページ：136, 2018年01月, ［査読有り］
  This paper presents GPUhd, a graphics processing unit (GPU) resource management approach that combines Hadoop and a GPU to obtain scale-out and scale-up functionality. There are several researches that combine Hadoop and GPU. However, there are no researches that can schedule tasks in consideration of GPU resource on Hadoop. Moreover, these researches cannot use multiple distributed frameworks. GPUhd extends the Yet Another Resource Negotiator (YARN) management mechanism and distributed processing frameworks for the coordinated use of GPU resources in Hadoop. We extend the YARN scheduling algorithm to consider GPU resources and incorporate a resources monitoring function. GPU resources can be managed on the basis of existing development methods because GPUhd simply handles GPU resources as host memory and CPU resources. In addition, GPUhd achieves high-speed processing, e.g., the computational time required to calculate 2048 × 2048 matrix multiplication is approximately 25 times less than that required when using only a CPU with Hadoop. GPUhd achieves high scalability and excellent response times in a heterogeneous distributed environment.
  Association for Computing Machinery, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1145/3149457.3155313
  DOI ID：10.1145/3149457.3155313, SCOPUS ID：85044373093
• Anomaly prediction based on k-means clustering for memory-constrained embedded devices　　　　　　　　　　　　　　　
  Yuto Kitagawa; Tasuku Ishigoka; Takuya Azumi
  Proceedings - 16th IEEE International Conference on Machine Learning and Applications, ICMLA 2017, 巻：2018-, 開始ページ：26, 終了ページ：33, 2018年01月, ［査読有り］
  This paper proposes an anomaly prediction method based on k-means clustering that assumes embedded devices with memory constraints to predict control system anomalies. With this method, by checking control system behavior, it is possible to predict anomalies. However, continuing clustering is difficult because data accumulate in memory similar to existing k-means clustering method, which is problematic for embedded devices with low memory capacity. Therefore, we also propose k-means clustering to continue clustering for infinite stream data. The proposed k-means clustering method is based on online k-means clustering of sequential processing. The proposed k-means clustering method only stores data required for anomaly prediction and releases other data from memory. Experimental results show that anomalies can be predicted by k-means clustering, and the proposed method can predict anomalies similar to standard k-means clustering while reducing memory consumption. Moreover, the proposed k-means clustering demonstrates better results of anomaly prediction than existing online k-means clustering.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ICMLA.2017.0-182
  DOI ID：10.1109/ICMLA.2017.0-182, SCOPUS ID：85048496597
• Component-Based mruby Platform for IoT Devices　　　　　　　　　　　　　　　
  Takuro Yamamoto; Takuma Hara; Takuya Ishikawa; Hiroshi Oyama; Hiroaki Takada; Takuya Azumi
  Journal of Information Processing, 巻：26, 開始ページ：549, 終了ページ：561, 2018年, ［査読有り］
  英語, 研究論文（学術雑誌）
• Remapping method to minimize makespan of simulink model for embedded multi-core systems　　　　　　　　　　　　　　　
  Sasuga Kojima; Masato Edahiro; Takuya Azumi
  Proceedings of the 33rd International Conference on Computers and Their Applications, CATA 2018, 巻：2018-, 2018年
  Multi-core processors have been increasingly used to reduce power consumption and improve performance in embedded systems. In addition, a use of Model-Based Development has been increasing. Relative to these trends, Model-Based Parallelizer (MBP) has an essential role in parallelizing applications (i.e., Simulink blocks). MBP maps Simulink blocks to cores at the model level using various information. Generally, mapping blocks to cores is considered an NP-hard graph optimization problem
  thus, MBP uses a heuristic method to solve it. A heuristic method can solve problems at high speed, but it is often not accurate. Therefore, this paper proposes two methods to improve results of MBP. The first method remaps blocks to cores based on the results of MBP. The second method determines an execution order that can finish the entire process faster than existing methods. Evaluations demonstrate that the proposed methods can obtain better results than existing methods in terms of Speedup and Load-Balancing.
  The International Society for Computers and Their Applications (ISCA), 英語, 研究論文（国際会議プロシーディングス）
  SCOPUS ID：85048427343
• Real-Time ROS Extension on Transparent CPU/GPU Coordination Mechanism.　　　　　　　　　　　　　　　
  Yuhei Suzuki; Takuya Azumi; Shinpei Kato; Nobuhiko Nishio
  21st IEEE International Symposium on Real-Time Distributed Computing(ISORC), 開始ページ：184, 終了ページ：192, 2018年
  IEEE Computer Society, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ISORC.2018.00035
  DOI ID：10.1109/ISORC.2018.00035, DBLP ID：conf/isorc/SuzukiAKN18
• Autoware on board: enabling autonomous vehicles with embedded systems.　　　　　　　　　　　　　　　
  Shinpei Kato; Shota Tokunaga; Yuya Maruyama; Seiya Maeda; Manato Hirabayashi; Yuki Kitsukawa; Abraham Monrroy; Tomohito Ando; Yusuke Fujii; Takuya Azumi
  Proceedings of the 9th ACM/IEEE International Conference on Cyber-Physical Systems(ICCPS), 開始ページ：287, 終了ページ：296, 2018年
  IEEE Computer Society / ACM, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ICCPS.2018.00035
  DOI ID：10.1109/ICCPS.2018.00035, DBLP ID：conf/iccps/KatoTMMHKMAFA18
• ROS-based support system for supervision of multiple UAVs by a single operator.　　　　　　　　　　　　　　　
  Hiroki Hayakawa; Takuya Azumi; Akinori Sakaguchi; Toshimitsu Ushio
  Proceedings of the 9th ACM/IEEE International Conference on Cyber-Physical Systems, ICCPS 2018, Porto, Portugal, April 11-13, 2018, 開始ページ：341, 終了ページ：342, 2018年, ［査読有り］
  IEEE Computer Society / ACM, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ICCPS.2018.00044
  DOI ID：10.1109/ICCPS.2018.00044, DBLP ID：conf/iccps/HayakawaASU18
• Exploring scalable data allocation and parallel computing on NoC-based embedded many cores　　　　　　　　　　　　　　　
  Yuya Maruyama; Shinpei Kato; Takuya Azumi
  Proceedings - 35th IEEE International Conference on Computer Design, ICCD 2017, 開始ページ：225, 終了ページ：228, 2017年11月, ［査読有り］
  In embedded systems, high processing requirements and low power consumption need heterogeneous computing platforms. Considering embedded requirements, applications need to be designed based on scalable data allocation and parallel computing with non-uniform memory access (NUMA) many cores. In this paper, we use one of the embedded commercial off-the-shelf (COTS) multi/many-core components, the Massively Parallel Processor Arrays (MPPA) 256 developed by Kalray, and conduct evaluations of data transfer and parallelization of a practical application. We investigate currently achievable data transfer latencies between distributed memories on network-on-chip (NoC), memory access characteristics, and parallelization potential with many cores. Subsequently, we run a practical application, the core of the autonomous driving system, on many-core processors and acceleration by parallelization indicates practicality of many cores. By highlighting many-core computing capabilities, we explore the scalable data allocation and parallel computing on NoC-based embedded many cores.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ICCD.2017.41
  DOI ID：10.1109/ICCD.2017.41, SCOPUS ID：85041685170
• Scheduling parallel and distributed processing for automotive data stream management system　　　　　　　　　　　　　　　
  Jaeyong Rho; Takuya Azumi; Mayo Nakagawa; Kenya Sato; Nobuhiko Nishio
  JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 巻：109, 開始ページ：286, 終了ページ：300, 2017年11月, ［査読有り］
  In this paper, to analyze end-to-end timing behavior in heterogeneous processor and network environments accurately, we adopt and modify a heterogeneous selection value on communication contention (HSV_CC) algorithm, which can synchronize tasks and messages simultaneously, for stream processing distribution. In order to adapt the concepts of a static algorithm like HSV_CC to automotive data stream management system (DSMSs), one must first address three issues: (i) previous task and message schedules might lead to less efficient resource usages in this scenario; (ii) the conventional method to determine the task scheduling order may not be best suited to deal with stream processing graphs, and; (iii) there is a need to be able to schedule tasks with time-varying computational requirements efficiently. To address (i), we propose the heterogeneous value with load balancing and communication contention (HVLB_CC) (A) algorithm, which considers load balancing in addition to the parameters considered by the HSV_CC algorithm. We propose HVLB_CC (B) to address issue (ii). HVLB_CC (B) can deal with stream processing task graphs and more various directed acyclic graphs to prevent assigning a higher priority to successor tasks. In addition, to address issue (iii), we propose HVLB_CC_IC. To schedule tasks more efficiently with various computation times, HVLB_CC_IC utilizes schedule holes left in processors. These idle time slots can be used for the execution of an optional part to generate more precise data results by applying imprecise computation models. Experimental results demonstrate that the proposed algorithms improve minimum schedule length, accuracy, and load balancing significantly compared to the HSV_CC algorithm. In addition, the proposed HVLB_CC (B) algorithm can schedule more varied task graphs without reducing performance, and, using imprecise computation models, HVLB_CC_IC yields higher precision data than HVLB_CC without imprecise computation models. (C) 2017 Elsevier Inc. All rights reserved.
  ACADEMIC PRESS INC ELSEVIER SCIENCE, 英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.1016/j.jpdc.2017.06.012
  DOI ID：10.1016/j.jpdc.2017.06.012, ISSN：0743-7315, eISSN：1096-0848, DBLP ID：journals/corr/RhoANSN17, Web of Science ID：WOS:000408298400021
• Scheduling parallel and distributed processing for automotive data stream management system　　　　　　　　　　　　　　　
  Jaeyong Rho; Takuya Azumi; Mayo Nakagawa; Kenya Sato; Nobuhiko Nishio
  JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 巻：109, 開始ページ：286, 終了ページ：300, 2017年11月, ［査読有り］
  In this paper, to analyze end-to-end timing behavior in heterogeneous processor and network environments accurately, we adopt and modify a heterogeneous selection value on communication contention (HSV_CC) algorithm, which can synchronize tasks and messages simultaneously, for stream processing distribution. In order to adapt the concepts of a static algorithm like HSV_CC to automotive data stream management system (DSMSs), one must first address three issues: (i) previous task and message schedules might lead to less efficient resource usages in this scenario; (ii) the conventional method to determine the task scheduling order may not be best suited to deal with stream processing graphs, and; (iii) there is a need to be able to schedule tasks with time-varying computational requirements efficiently. To address (i), we propose the heterogeneous value with load balancing and communication contention (HVLB_CC) (A) algorithm, which considers load balancing in addition to the parameters considered by the HSV_CC algorithm. We propose HVLB_CC (B) to address issue (ii). HVLB_CC (B) can deal with stream processing task graphs and more various directed acyclic graphs to prevent assigning a higher priority to successor tasks. In addition, to address issue (iii), we propose HVLB_CC_IC. To schedule tasks more efficiently with various computation times, HVLB_CC_IC utilizes schedule holes left in processors. These idle time slots can be used for the execution of an optional part to generate more precise data results by applying imprecise computation models. Experimental results demonstrate that the proposed algorithms improve minimum schedule length, accuracy, and load balancing significantly compared to the HSV_CC algorithm. In addition, the proposed HVLB_CC (B) algorithm can schedule more varied task graphs without reducing performance, and, using imprecise computation models, HVLB_CC_IC yields higher precision data than HVLB_CC without imprecise computation models. (C) 2017 Elsevier Inc. All rights reserved.
  ACADEMIC PRESS INC ELSEVIER SCIENCE, 英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.1016/j.jpdc.2017.06.012
  DOI ID：10.1016/j.jpdc.2017.06.012, ISSN：0743-7315, eISSN：1096-0848, DBLP ID：journals/jpdc/RhoANSN17, Web of Science ID：WOS:000408298400021
• Real-Time GPU Resource Management with Loadable Kernel Modules　　　　　　　　　　　　　　　
  Yuhei Suzuki; Yusuke Fujii; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato
  IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 巻：28, 号：6, 開始ページ：1715, 終了ページ：1727, 2017年06月, ［査読有り］
  Graphics processing unit (GPU) programming environments have matured for general-purpose computing on GPUs. Significant challenges for GPUs include system software support for bounded response times and guaranteed throughput. In recent years, GPU technologies have been applied to real-time systems by extending the operating system modules to support real-time GPU resource management. Unfortunately, such a system extension makes it difficult to maintain the system with version updates because the OS kernel and device drivers must be modified at the source-code level, thereby preventing continuous research and development of GPU technologies for real-time systems. A loadable kernel module (LKM) framework, called Linux Real-Time eXtention with GPUs (Linux-RTXG), for managing real-time GPU resources with Linux without modifying the OS kernel and device drivers is proposed and evaluated experimentally. Linux-RTXG provides mechanisms for interrupt interception and independent synchronization to achieve real-time scheduling and resource reservation capabilities for GPU applications on top of existing device drivers and runtime libraries. Experimental results demonstrate that the overhead incurred by introducing the proposed Linux-RTXG is comparable to that of introducing existing kernel-dependent approaches. In addition, the results demonstrate that multiple GPU applications can be scheduled successfully by Linux-RTXG to meet their priority and quality-of-service requirements in real time.
  IEEE COMPUTER SOC, 英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.1109/TPDS.2016.2630697
  DOI ID：10.1109/TPDS.2016.2630697, ISSN：1045-9219, eISSN：1558-2183, DBLP ID：journals/tpds/SuzukiFANK17, Web of Science ID：WOS:000401365300014
• Component-Based Framework of Lightweight Ruby for Efficient Embedded Software Development　　　　　　　　　　　　　　　
  Takuro Yamamoto; Hiroshi Oyama; Takuya Azumi
  JSSST Journal on Computer Software, 巻：34, 号：4, 開始ページ：3, 終了ページ：16, 2017年, ［査読有り］
  日本ソフトウェア科学会 ; 1984-, 英語
  ISSN：0289-6540, CiNii Articles ID：40021383596, CiNii Books ID：AN10075819
• GPUrpc: Exploring Transparent Access to Remote GPUs　　　　　　　　　　　　　　　
  Yuki Iida; Yusuke Fujii; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato
  ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS, 巻：16, 号：1, 開始ページ：17:1-17:25, 2016年11月, ［査読有り］
  Graphics processing units (GPUs) are increasingly used for high-performance computing. Programming frameworks for general-purpose computing on GPUs (GPGPU), such as CUDA and OpenCL, are also maturing. Driving this trend is the recent proliferation of mobile devices such as smartphones and wearable computers. These devices are increasingly incorporating computationally intensive applications that involve some form of environmental recognition such as augmented reality (AR) or voice recognition. However, devices with low computational power cannot satisfy such demanding computing requirements. The CPU load of these devices could be reduced by offloading computation onto GPUs on the cloud. This paper presents GPUrpc, a remote procedure call (RPC) extension to Gdev, which is a rich set of runtime libraries and device drivers for achieving first-class GPU resource management. GPUrpc allows developers to use CUDA for GPGPU development work. Existing research uses RPCs based on the CUDA application programming interfaces (APIs); hence, all CUDA APIs require communication. To reduce communication overhead, we use an RPC based on a low-level API than CUDA API and reduced API that does not require communication. Our evaluation conducted on Linux and NVIDIA GPUs shows that the basic performance of our prototype implementation is reliable in comparison with the existing method. Evaluation using the Rodinia benchmark suite designed for research in heterogeneous parallel computing showed that GPUrpc is effective for applications such as image processing and data mining. GPUrpc also can improve power consumption to approximately 1/6 that of CPU processing for performing 512 x 512 matrix multiplication.
  ASSOC COMPUTING MACHINERY, 英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.1145/2950056
  DOI ID：10.1145/2950056, ISSN：1539-9087, eISSN：1558-3465, DBLP ID：journals/tecs/IidaFANK16, Web of Science ID：WOS:000391441900017
• Exploring the Performance of ROS2　　　　　　　　　　　　　　　
  Yuya Maruyama; Shinpei Kato; Takuya Azumi
  2016 PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON EMBEDDED SOFTWARE (EMSOFT), 2016年, ［査読有り］
  Middleware for robotics development must meet demanding requirements in real-time distributed embedded systems. The Robot Operating System (ROS), open-source middleware, has been widely used for robotics applications. However, the ROS is not suitable for real-time embedded systems because it does not satisfy real-time requirements and only runs on a few OSs. To address this problem, ROS1 will undergo a significant upgrade to ROS2 by utilizing the Data Distribution Service (DDS). DDS is suitable for realtime distributed embedded systems due to its various transport configurations (e.g., deadline and fault-tolerance) and scalability. ROS2 must convert data for DDS and abstract DDS from its users; however, this incurs additional overhead, which is examined in this study. Transport latencies between ROS2 nodes vary depending on the use cases, data size, configurations, and DDS vendors. We conduct proof of concept for DDS approach to ROS and arrange DDS characteristic and guidelines from various evaluations. By highlighting the DDS capabilities, we explore and evaluate the potential and constraints of DDS and ROS2.
  ASSOC COMPUTING MACHINERY, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1145/2968478.2968502
  DOI ID：10.1145/2968478.2968502, Web of Science ID：WOS:000414220100005
• Componentizing an Operating System Feature Using a TECS Plugin　　　　　　　　　　　　　　　
  Tomoaki Kawada; Takuya Azumi; Hiroshi Oyama; Hiroaki Takada
  PROCEEDINGS OF 2016 IEEE 4TH INTERNATIONAL CONFERENCE ON CYBER-PHYSICAL SYSTEMS, NETWORKS, AND APPLICATIONS (CPSNA), 開始ページ：95, 終了ページ：99, 2016年, ［査読有り］
  This paper presents a componentized TOPPERS/ASP3 (a real-time operating system) time event notification using a plugin feature for the TOPPERS Embedded Component System (TECS). With emerging trends in cyber-physical systems, productivity is becoming a problem, owing to the complexity of such systems. To solve this problem, TECS has been proposed previously. Whereas TECS supports the generation of static API statements, however, it is limited insofar as it cannot generate certain complex statements, including those of the TOPPERS/ASP3 time event notification. Therefore, this paper proposes a TECS generator plugin that generates complex static API statements. The experimental results show that the runtime overhead is considerably low and that the proposed plugin is useful for componentizing the TOPPERS/ASP3 time event notification.
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/CPSNA.2016.27
  DOI ID：10.1109/CPSNA.2016.27, Web of Science ID：WOS:000391549000019
• HLBS: Heterogeneous Laxity-Based Scheduling Algorithm for DAG-based Real-Time Computing　　　　　　　　　　　　　　　
  Yuhei Suzuki; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato
  PROCEEDINGS OF 2016 IEEE 4TH INTERNATIONAL CONFERENCE ON CYBER-PHYSICAL SYSTEMS, NETWORKS, AND APPLICATIONS (CPSNA), 開始ページ：83, 終了ページ：88, 2016年, ［査読有り］
  Parallel scheduling of task graphs has been an important research topic in the field of parallel processing, and it is well known to be a challenging NP-complete problem. Scheduling in data flow systems can be represented by a directed acyclic graph (DAG). In parallel task scheduling using DAG, managing the dependent tasks reduce overall execution time. However, the algorithm cannot be applied to real-time systems with a mix of different deadlines and multi outputs, such as an autonomous driving system, because real-time systems can be represented by a DAG are mainly limited threads known as a fork-join structure. Real-time scheduling algorithm that does not depend on the structure of DAG can be widely used.
  This paper presents the Heterogeneous Laxity-Based Scheduling (HLBS) algorithm, which uses list scheduling heuristics to solve this problem. HLBS computes the rest time until deadlines, known as laxity, and preferentially assigns a task with shorter laxity to the processor. This enables scheduling of multiple deadlines to reduce deadline miss rate. The evaluation results demonstrate that HLBS reduced the deadline miss rate by an average of 45.6 % while achieving a performance level comparable to that of algorithms presented in previous studies.
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/CPSNA.2016.25
  DOI ID：10.1109/CPSNA.2016.25, Web of Science ID：WOS:000391549000017
• Lightweight Ruby Framework for Improving Embedded Software Efficiency　　　　　　　　　　　　　　　
  Takuro Yamamoto; Hiroshi Oyama; Takuya Azumi
  PROCEEDINGS OF 2016 IEEE 4TH INTERNATIONAL CONFERENCE ON CYBER-PHYSICAL SYSTEMS, NETWORKS, AND APPLICATIONS (CPSNA), 開始ページ：71, 終了ページ：76, 2016年, ［査読有り］
  Recently, the productivity of embedded systems applied to Cyber physical systems has become problematic due to their increasing complexity and scale. To improve the productivity, the mruby on TOPPERS embedded component system (TECS) framework, which employs scripting language (i.e., a lightweight Ruby) and supports component-based development, has been proposed. This paper proposes an extended framework of mruby on TECS that improves development efficiency more than the current version. In the proposed framework, a Bluetooth loader that receives an mruby bytecode, and a RiteVM scheduler that simplifies multitasking are implemented.
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/CPSNA.2016.23
  DOI ID：10.1109/CPSNA.2016.23, Web of Science ID：WOS:000391549000015
• Priority and Synchronization Support for ROS　　　　　　　　　　　　　　　
  Yukihiro Saito; Takuya Azumi; Shinpei Kato; Nobuhiko Nishio
  PROCEEDINGS OF 2016 IEEE 4TH INTERNATIONAL CONFERENCE ON CYBER-PHYSICAL SYSTEMS, NETWORKS, AND APPLICATIONS (CPSNA), 開始ページ：77, 終了ページ：82, 2016年, ［査読有り］
  Robot Operating System (ROS) is a software component framework developed for robot applications. It provides a rich set of software libraries and tools to construct robot components, including implementation of many popular perception, planning, and control algorithms. The state of the art in ROS, however, does not fundamentally support priority and synchronization among tasks, so-called nodes in ROS. Therefore, ROS may not be appropriate as a platform of real-time multi-tasking environments, despite the fact that most robot applications are multi-tasking and running under real-time constraints. In this paper, we explore real-time performance of ROS. We present a priority-based message transmission mechanism to reduce the execution time and time variance of high-priority ROS nodes. We also present a synchronization mechanism to harmonize multiple ROS nodes running at different frequencies. The presented mechanisms can be both used with legacy ROS applications without modification. Experiments using an autonomous driving system show a 62.3% reduction in a worst-case execution time and indicate that time differences between nodes are guaranteed below a constant time. In addition, the total CPU utilization is saved by up to about 10%.
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/CPSNA.2016.24
  DOI ID：10.1109/CPSNA.2016.24, Web of Science ID：WOS:000391549000016
• RTM-TECS: Collaboration Framework for Robot Technology Middleware and Embedded Component System　　　　　　　　　　　　　　　
  Ryo Hasegawa; Naofumi Yawata; Noriaki Ando; Nobuhiko Nishio; Takuya Azumi
  2016 IEEE 19TH INTERNATIONAL SYMPOSIUM ON REAL-TIME DISTRIBUTED COMPUTING (ISORC 2016), 開始ページ：212, 終了ページ：220, 2016年, ［査読有り］
  Robot technologies, such as robot technology middleware (RTM) that is a component-oriented platform, are popular. However, RTM does not ensure stable real-time processing in common object request broker architecture. In this paper, a collaboration framework of RTM and TOPPERS embedded component system (TECS) is proposed to address this problem. TECS, a system that satisfies real-time processing requirements, is employed to enhance real-time processing in the proposed framework. To implement the collaboration of RTM and TECS, we have adopted remote procedure call and one-way communication. In addition, extending a generator enables the generation of robot technology components from TECS components. We have evaluated the processor cycle counts of the proposed framework in comparison with those of a conventional method. In addition, we evaluated the execution time of serial communication and a motor application using the proposed framework. The evaluation results show that the proposed framework is functionally employed in a hard real-time system. Furthermore, we evaluated the amount of code generated by the proposed framework. The evaluation results reveal that the code generated by the proposed framework is reusable and can enhance productivity.
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ISORC.2016.37
  DOI ID：10.1109/ISORC.2016.37, Web of Science ID：WOS:000390839100026
• WiP Abstract: Preliminary Evaluation of ROS2　　　　　　　　　　　　　　　
  Yuya Maruyama; Shinpei Kato; Takuya Azumi
  2016 ACM/IEEE 7TH INTERNATIONAL CONFERENCE ON CYBER-PHYSICAL SYSTEMS (ICCPS), 2016年, ［査読有り］
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  ISSN：2375-8317, Web of Science ID：WOS:000386348200042
• Distributed processing for automotive data stream management system on mixed single- and multi-core processors.
  Jaeyong Rho; Takuya Azumi; Hiroshi Oyama; Kenya Sato; Nobuhiko Nishio
  SIGBED Review, 巻：13, 号：3, 開始ページ：15, 終了ページ：22, 2016年, ［査読有り］
  DBLP ID：journals/sigbed/RhoAOSN16
• Reservation-based scheduling for automotive DSMS under high overload condition　　　　　　　　　　　　　　　
  Jaeyong Rho; Takuya Azumi; Akihiro Yamaguchi; Kenya Sato; Nobuhiko Nishio
  Journal of Information Processing, 巻：24, 号：5, 開始ページ：751, 終了ページ：761, 2016年, ［査読有り］
  Recent automotive systems require various data, including data from on-board sensors and external sources to recognize environmental conditions. As the amount of sensor data used in automotive systems increases, processes that use such data become increasingly complicated. In addition, similar data processing can be duplicated over multiple applications. To address these issues, a data stream management system (DSMS) for automotive systems based on a data integration architecture has been developed. However, hard real-time deadlines cannot be guaranteed due to unpredictable load changes caused by data streams. For example, the arrival time and CPU utilization requested by data streams from vehicle-to-vehicle communications change rapidly depending on environmental conditions. We propose the reservation-based operator path earliest deadline first (ROP-EDF) scheduling algorithm for an automotive DSMS under overload conditions. The proposed algorithm reserves processor time preferentially for hard real-time tasks so that tasks can meet deadlines under overload conditions. ROP-EDF can be used for load testing on a single processor system. Experimental results show the effectiveness of the proposed algorithm compared with existing algorithms relative to the deadline miss ratio under overload conditions.
  Information Processing Society of Japan, 英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.2197/ipsjjip.24.751
  DOI ID：10.2197/ipsjjip.24.751, ISSN：1882-6652, DBLP ID：journals/jip/RhoAYSN16, SCOPUS ID：84987934456
• mruby on TECS: Component-based Framework for Running Script Program　　　　　　　　　　　　　　　
  Takuya Azumi; Yuki Nagahara; Hiroshi Oyama; Nobuhiko Nishio
  2015 IEEE 18th International Symposium on Real-Time Distributed Computing (ISORC), 開始ページ：252, 終了ページ：259, 2015年, ［査読有り］
  Scripting languages are attractive for embedded system due to their high productivity. However, it is difficult to use scripting languages in a practical application because their performance and libraries for managing embedded devices are immature compared to the C programming language. This paper proposes a framework for effectively running an mruby script program on embedded systems based on the TOPPERS Embedded Component System (TECS). TECS generates glue code for invocation from mruby programs to legacy code in C language. It also supports configuration of mruby. Experimental results demonstrate the effectiveness of the proposed framework.
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ISORC.2015.42
  DOI ID：10.1109/ISORC.2015.42, Web of Science ID：WOS:000380606500031
• Cloud-assisted sensing and supervision of multiple unmanned aerial vehicles by a single operator.　　　　　　　　　　　　　　　
  Takashi Shigekuni; Toshimitsu Ushio; Takuya Azumi
  Proceedings of the ACM/IEEE Sixth International Conference on Cyber-Physical Systems, ICCPS 2015, Seattle, WA, USA, April 14-16, 2015, 開始ページ：254, 2015年, ［査読有り］
  ACM
  DOI：https://doi.org/10.1145/2735960.2735987
  DOI ID：10.1145/2735960.2735987, DBLP ID：conf/iccps/ShigekuniUA15
• Adaptive Assignment of Deadline and Clock Frequency in Real-Time Embedded Control Systems　　　　　　　　　　　　　　　
  Tatsuya Yoshimoto; Toshimitsu Ushio; Takuya Azumi
  IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES, 巻：E98A, 号：1, 開始ページ：323, 終了ページ：330, 2015年01月, ［査読有り］
  Computing and power resources are often limited in real-time embedded control systems. In this paper, we resolve the trade-off problem between control performance and power consumption in a real-time embedded control system with a dynamic voltage and frequency scaling (DVFS) uniprocessor implementing multiple control tasks. We formulate an optimization problem whose cost function depends on both the control performance and the power consumption. We introduce an adapter into the real-time embedded control system that adaptively assigns deadlines of jobs and clock frequencies according to the plant's stability and schedulability by solving the optimization problem. In numerical simulations, we show that the proposed adapter can reduce the power consumption while maintaining the control performance.
  IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG, 英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.1587/transfun.E98.A.323
  DOI ID：10.1587/transfun.E98.A.323, ISSN：1745-1337, DBLP ID：journals/ieicet/YoshimotoUA15, Web of Science ID：WOS:000359464100034
• Connected smartphones and high-performance servers for remote object detection　　　　　　　　　　　　　　　
  Yuki Iida; Manato Hirabayashi; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato
  Proceedings - 2nd IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, CPSNA 2014, 開始ページ：71, 終了ページ：76, 2014年11月, ［査読有り］
  Object detection is one of the main challenges of cyber-physical systems for mobile applications. Mobile devices as embedded systems might not provide sufficient performance to meet the computational requirements for object detection, whereas it is not reasonable to enhance such devices with powerful processors. In this paper, we present a prototype system for mobile object detection based on smartphones and high-performance servers supplemented with graphics processing units (GPUs). The basic idea in the design of this system is to offload expensive image processing on a remote server over a wireless network, which allows mobile devices to support high-precision object detection. We describe an experimental evaluation using Android smartphones and high-end workstation with NVIDIA GPUs. Our conclusion is that mobile object detection is achievable at the expense of millisecond latency due to the communication between the mobile device and the remote server.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/CPSNA.2014.24
  DOI ID：10.1109/CPSNA.2014.24, SCOPUS ID：84916613212
• Reducing data copies between GPUs and NICs　　　　　　　　　　　　　　　
  Anh Nguyen; Yusuke Fujii; Yuki Iida; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato
  Proceedings - 2nd IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, CPSNA 2014, 開始ページ：37, 終了ページ：42, 2014年11月, ［査読有り］
  Cyber-physical systems (CPS) must perform complex algorithms at very high speed to monitor and control complex real-world phenomena. GPU, with a large number of cores and extremely high parallel processing, promises better computation if the data parallelism often found in real-world scenarios of CPS could be exploited. Nevertheless, its performance is limited by the latency incurred when data are transferred between GPU memory and I/O devices. This paper describes a method, based on zero-copy processing, for data transmission between GPUs and NICs. The arrangement enables NICs to directly transfer data to and from GPU. Experimental results show effective data throughput without packet loss.
  Institute of Electrical and Electronics Engineers Inc., 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/CPSNA.2014.15
  DOI ID：10.1109/CPSNA.2014.15, SCOPUS ID：84916602854
• TECS components providing functionalities of OSEK specifications for ITRON OS　　　　　　　　　　　　　　　
  Atsushi Ohno; Takuya Azumi; Nobuhiko Nishio
  Journal of Information Processing, 巻：22, 号：4, 開始ページ：584, 終了ページ：594, 2014年10月, ［査読有り］
  The number of electronic control units (ECUs) has increased to manage complicated vehicle systems. Many kinds of operating systems that run on ECUs exist: ITRON OS, OSEK OS, and so forth. Currently, developers implement the system control software according to the ITRON and OSEK specifications independently. For example, even though OSes provide similar functionalities, OSEK specifications have several differences from ITRON specifications such as scheduling policies (Non-preemptive scheduling), alarms, hook routines, and several system calls. Thus, when using legacy software following OSEK specifications on the ITRON OS, developers have to port the software to ITRON OS. This paper presents a component-based framework to fill the gap between OSEK and ITRON specifications by using TECS (TOPPERS Embedded Component System). The work required to port legacy OSEK applications built with TECS components to ITRON applications built for TECS is reduced by using our method. TECS is a high-level abstraction component system for enhancing the reusability of software. Examples for the characteristics of the framework are: (1) Non-preemptive scheduling tasks are implemented by changing the priority of the task to the highest priority
  (2) The system works as the OSEK alarm based on a counter value, which is incremented at an arbitrary time interval
  (3) OSEK hook routines are also available with a particular timing. Experimental results demonstrate that the overhead of the corresponding system calls compared to the original OSEK system calls is reduced to within 13.58 μsec.
  Information Processing Society of Japan, 英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.2197/ipsjjip.22.584
  DOI ID：10.2197/ipsjjip.22.584, ISSN：1882-6652, DBLP ID：journals/jip/OhnoAN14, SCOPUS ID：84908031104
• ROP-EDF: Reservation-Based OP-EDF Scheduling for Automotive Data Stream Management System　　　　　　　　　　　　　　　
  Jaeyong Rho; Akihiro Yamaguchi; Kenya Sato; Takuya Azumi; Nobuhiko Nishio
  WiP session of the IEEE 11th International Conference on Embedded Software and Systems (ICESS2014), 2014年08月, ［査読有り］
• VISA synthesis: Variation-aware instruction set architecture synthesis　　　　　　　　　　　　　　　
  Yuko Hara-Azumi; Takuya Azumi; Nikil D. Dutt
  Proceedings of the Asia and South Pacific Design Automation Conference, ASP-DAC, 開始ページ：243, 終了ページ：248, 2013年
  We present VISA: a novel Variation-aware Instruction Set Architecture synthesis approach that makes effective use of process variation from both software and hardware points of view. To achieve an efficient speedup, VISA selects custom instructions based on statistical static timing analysis (SSTA) for aggressive clocking. Furthermore, with minimum performance overhead, VISA dynamically detects and corrects timing faults resulting from aggressive clocking of the underlying processor. This hybrid software/hardware approach generates significant speedup without degrading the yield. Our experimental results on commonly used ISA synthesis benchmarks demonstrate that VISA achieves significant performance improvement compared with a traditional deterministic worst case-based approach (up to 78.0%) and an existing SSTA-based approach (up to 49.4%). © 2013 IEEE.
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ASPDAC.2013.6509603
  DOI ID：10.1109/ASPDAC.2013.6509603, SCOPUS ID：84877766061
• TECSCE: HW/SW Codesign Framework for Data Parallelism Based on Software Component.　　　　　　　　　　　　　　　
  Takuya Azumi; Yasaman Samei Syahkal; Yuko Hara-Azumi; Hiroshi Oyama 0002; Rainer Dömer
  Embedded Systems: Design, Analysis and Verification - 4th IFIP TC 10 International Embedded Systems Symposium(IESS), 開始ページ：1, 終了ページ：13, 2013年
  Springer, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1007/978-3-642-38853-8_1
  DOI ID：10.1007/978-3-642-38853-8_1, DBLP ID：conf/iess/AzumiSHOD13
• Detecting Wi-Fi base station behavior inappropriate for positioning method in participatory sensing logs　　　　　　　　　　　　　　　
  Nobuhiko Nishio; Yuuki Fukuzaki; Takuya Azumi
  UbiComp 2013 Adjunct - Adjunct Publication of the 2013 ACM Conference on Ubiquitous Computing, 開始ページ：665, 終了ページ：672, 2013年, ［査読有り］
  Recently mobile base stations are getting increased, which is considered harmful for the Wi-Fi positioning methods. In this paper, three approaches for detecting Wi-Fi base station behaviors inappropriate for Wi-Fi signature sampling are introduced and their performance evaluations are presented. First approach is for outdoor environment using GPS or Wi-Fi, second for indoor environment using Wi-Fi and accelerometers and last for the first contact stations using the Bayesian estimation method. Bayesian estimation is fine for stationary stations but much severe for mobile stations. Copyright © 2013 ACM.
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1145/2494091.2495987
  DOI ID：10.1145/2494091.2495987, SCOPUS ID：84885230009
• Pdr-based adaptation for user-progress in interactive navigation system　　　　　　　　　　　　　　　
  Shun Yoshimi; Takuya Azumi; Nobuhiko Nishio
  UbiComp 2013 Adjunct - Adjunct Publication of the 2013 ACM Conference on Ubiquitous Computing, 開始ページ：263, 終了ページ：266, 2013年, ［査読有り］
  Recently, indoor pedestrian navigation systems have attracted attentions of people. We have been developing an interactive pedestrian navigation system and found that the navigation system could not provide enough sense of reassurance for a user according to our preliminary experiment. In this study, we propose a method to improve the user's sense of reassurance by notifying user's level of achievement sensed by their smart phone sensors and conduct the experiment in the Osaka underground city. According to the questionnaire survey that we conducted after the experiment, people who chose affirmative opinions on a sense of reassurance increased by 45% due to the proposed method.
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1145/2494091.2494163
  DOI ID：10.1145/2494091.2494163, SCOPUS ID：84885215782
• Exploring microcontrollers in GPUs　　　　　　　　　　　　　　　
  Yusuke Fujii; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato
  Proceedings of the 4th Asia-Pacific Workshop on Systems, APSys 2013, 2013年, ［査読有り］
  Recent graphics processing units (GPUs) integrate wimpy microcontrollers on a chip. They are often used to execute firmware code configuring the functional units of GPUs. This paper opens up the programming of these microcontrollers and explores how to utilize them for GPU resource management. Our prototype system provides a compiler suite for NVIDIA's GPU microcontrollers with its basis on the Low Level Virtual Machine (LLVM) infrastructure. As a proof of concept, we develop fully-functional firmware using our compiler and provide a basic performance evaluation. The experimental results demonstrate that the overhead of introducing our firmware is suppressed to within 2.3%, as compared to the native proprietary firmware, while the impact of overhead is no greater than 0.01% of the total execution time according to microbenchmarks. We also show that a complementary use of microcontrollers can reduce the latency of data transfers with concurrent multiple data streams. Copyright 2013 ACM.
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1145/2500727.2500742
  DOI ID：10.1145/2500727.2500742, SCOPUS ID：84883060075
• HR-TECS: Component Technology for Embedded Systems with Memory Protection　　　　　　　　　　　　　　　
  Takuya Ishikawa; Takuya Azumi; Hiroshi Oyama; Hiroaki Takada
  2013 IEEE 16TH INTERNATIONAL SYMPOSIUM ON OBJECT/COMPONENT/SERVICE-ORIENTED REAL-TIME DISTRIBUTED COMPUTING (ISORC), 2013年, ［査読有り］
  A software partitioning has been used to develop safety-critical systems in recent years. In addition, software component technologies supporting a software partitioning have been developed. This paper describes the new component technology for embedded software that requires memory protection, which is one of the important features for the partitioning. HR-TECS is a new component technology based on the real-time operating system supporting the static memory layout. Developers can easily allocate components to partitions in order to protect memory areas. In addition, HR-TECS supports inter-partition communications so that developers can implement components without consideration for inter-partition communications. The results of evaluation demonstrate the effectiveness of HR-TECS.
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  ISSN：1555-0885, Web of Science ID：WOS:000365605900012
• Distributed Intent: Android Framework for Networked Devices Operation　　　　　　　　　　　　　　　
  Yuki Nagahara; Hiroshi Oyama; Takuya Azumi; Nobuhiko Nishio
  2013 IEEE 16TH INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND ENGINEERING (CSE 2013), 開始ページ：651, 終了ページ：658, 2013年, ［査読有り］
  We propose a distributed framework extending Intent in Android systems to improve collaboration with embedded devices. In the proposed framework, Android applications collaborate with embedded devices by sending serialized Intent messages to networked embedded devices. The applications can collaborate with embedded devices without proxy hardware. We indicate the framework's effectiveness by evaluating its execution times and memory usage.
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/CSE.2013.101
  DOI ID：10.1109/CSE.2013.101, ISSN：1949-0828, Web of Science ID：WOS:000351950300093
• Data Transfer Matters for GPU Computing　　　　　　　　　　　　　　　
  Yusuke Fujii; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato; Masato Edahiro
  2013 19TH IEEE INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED SYSTEMS (ICPADS 2013), 開始ページ：275, 終了ページ：282, 2013年, ［査読有り］
  Graphics processing units (GPUs) embrace many-core compute devices where massively parallel compute threads are offloaded from CPUs. This heterogeneous nature of GPU computing raises non-trivial data transfer problems especially against latency-critical real-time systems. However even the basic characteristics of data transfers associated with GPU computing are not well studied in the literature. In this paper, we investigate and characterize currently-achievable data transfer methods of cutting-edge GPU technology. We implement these methods using open-source software to compare their performance and latency for real-world systems. Our experimental results show that the hardware-assisted direct memory access (DMA) and the I/O read-and-write access methods are usually the most effective, while on-chip microcontrollers inside the GPU are useful in terms of reducing the data transfer latency for concurrent multiple data streams. We also disclose that CPU priorities can protect the performance of GPU data transfers.
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ICPADS.2013.47
  DOI ID：10.1109/ICPADS.2013.47, ISSN：1521-9097, Web of Science ID：WOS:000355315600031
• Lifelog-Based Active Movement Assistant System　　　　　　　　　　　　　　　
  Yusuke Satonaka; Takumi Kitazawa; Kazuki Suzuki; Yuki Fukuzaki; Takuya Azumi; Nobuhiko Nishio
  2013 IEEE 1ST INTERNATIONAL CONFERENCE ON CYBER-PHYSICAL SYSTEMS, NETWORKS, AND APPLICATIONS (CPSNA), 開始ページ：7, 終了ページ：12, 2013年, ［査読有り］
  In our daily life, we commute between home and office. Even if we leave at around the same time and on the the same route, the arrival time might not be the same due to the difference in walking speed, traffic conditions and trains to ride. Therefore, it is useful to develop a movement assisting system which can give the user timely pace-making advice. In order to give such advice automatically, the system has to know an accurate progress of the movement, the method of transportation and its mobility characteristics; walking speed, routes and traffic congestion characteristics. Hence, we propose a movement assisting system that proactively checks user's movement progress and gives timely advice upon automatically created personalized plan. Utilizing user's daily life log data of movements, it extracts key features as the places the often visited at and methods of transportation between them, and estimates arrival time to the destination according to their movement characteristics. In the case of movement by bus, our system could reduce 25% of arrival time estimation error compared with the conventional methods.
  IEEE, 英語, 研究論文（国際会議プロシーディングス）
  Web of Science ID：WOS:000339121000003
• TECS Components Providing Functionalities of OSEK Specification for ITRON OS　　　　　　　　　　　　　　　
  Atsushi Ohno; Takuya Azumi; Nobuhiko Nishio
  2012 IEEE 14TH INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING AND COMMUNICATIONS & 2012 IEEE 9TH INTERNATIONAL CONFERENCE ON EMBEDDED SOFTWARE AND SYSTEMS (HPCC-ICESS), 開始ページ：1434, 終了ページ：1441, 2012年, ［査読有り］
  Recently, vehicle systems have become more complicated and larger in scale. This makes two specifications, ITRON and OSEK, vital in the automotive embedded system. Therefore, developers have to apply the two specifications while implementing the system control software, though they provide the same features. It is difficult to provide the same functionalities, as software based on ITRON and OSEK must work independently of each other.
  This paper introduces a component-based approach to use the functionalities of the OSEK specification by using a TECS component system based on the ITRON specification. TECS is a high level abstraction component system. The characteristics of the prototype are listed below. Non-preemptive scheduling tasks are implemented by transiting the system state to the dispatching disable state. The system works as the OSEK alarm based on a counter value which is incremented at an arbitrary time interval. The OSEK resource will change a task's original priority. A set of tasks and alarms which are automatically started after the OS is started can be changed. In addition, the OSEK hook routines are also available with particular timing. The functionalities of the OSEK specification are available on an OS based the ITRON specification. In the end of this paper, the overhead of correspondence call system calls and the user restrictions are mentioned.
  IEEE COMPUTER SOC, 英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/HPCC.2012.210
  DOI ID：10.1109/HPCC.2012.210, Web of Science ID：WOS:000310377500202
• 学生および若手技術者による組込みシステム技術に関するサマースクールの実践　　　　　　　　　　　　　　　
  谷口 一徹; 安藤 友樹; 高瀬 英希; 安積 卓也; 松原 豊; 細合 晋太郎; 村上 靖明; 菅谷 みどり
  情報処理学会論文誌, 巻：52, 号：12, 開始ページ：3221-3237, 2011年12月
• TECS仕様に基づいたNXT用ソフトウェアプラットフォームの開発　　　　　　　　　　　　　　　
  石川拓也; 安積卓也; 一場利幸; 柴田誠也; 本田晋也; 高田広章
  コンピュータソフトウェア, 巻：28, 号：4, 開始ページ：158-174, 2011年11月
  DOI：https://doi.org/10.11309/jssst.28.4_158
  DOI ID：10.11309/jssst.28.4_158
• TECS-based software platform on mindstorms NXT　　　　　　　　　　　　　　　
  Takuya Ishikawa; Toshiyuki Ichiba; Seiya Shibata; Shinya Honda; Hiroaki Takada; Takuya Azumi
  Computer Software, 巻：28, 号：4, 開始ページ：158, 終了ページ：174, 2011年
  This paper presents a software platform for LEGO Mindstorms NXT, named ATON (ASP+TECS On NXT), which is based on a component system for embedded systems, TECS. ATON uses TECS so that memory usage can be optimized by detaching unused device drivers, and also application developers can write their programs smoothly from software models. In order to achieve such features, ATON includes TECS components such as tasks and semaphores which is provided by a Real-Time OS, and device drivers for devices of NXT. A case study on line trace robot development demonstrates effectiveness of ATON.
  日本語, 研究論文（学術雑誌）
  ISSN：0289-6540, SCOPUS ID：82755186867
• オープンソース組込みシステム向けシミュレータのマルチプロセッサ拡張　　　　　　　　　　　　　　　
  安積卓也; 古川貴士; 相庭裕史; 柴田誠也; 本田晋也; 冨山宏之; 高田広章
  コンピュータソフトウェア, 巻：27, 号：4, 開始ページ：24-42, 2010年11月
  DOI：https://doi.org/10.11309/jssst.27.4_24
  DOI ID：10.11309/jssst.27.4_24
• コンポーネントシステムを用いた組込みシステム向けアクセス制御機構
  安積卓也; 山田晋平; 大山博司; 中本幸一; 高田広章
  電子情報通信学会論文誌D, 巻：J93-D, 号：10, 開始ページ：2021-2031, 終了ページ：2031, 2010年10月
  社団法人電子情報通信学会, 日本語
  ISSN：1880-4535, CiNii Articles ID：120003147835
• Wheeled inverted pendulum with embedded component system: A case study　　　　　　　　　　　　　　　
  Takuya Azumi; Hiroaki Takada; Takayuki Ukai; Hiroshi Oyama
  ISORC 2010 - 2010 13th IEEE International Symposium on Object/Component/Service-Oriented Real-Time Distributed Computing, 巻：1, 開始ページ：151, 終了ページ：155, 2010年, ［査読有り］
  Software component techniques have been widely used for enhancement and the cost reduction of software development. We herein introduce a component system with a real-time operating system (RTOS). A case study of a two-wheeled inverted pendulum balancing robot with the component system is presented. The component system can deal with RTOS resources, such as tasks and semaphores, as components. Moreover, a trace functionality which is a new functionality to confirm the state of components or calling components without modification of C source code is introduced. © 2010 IEEE.
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/ISORC.2010.21
  DOI ID：10.1109/ISORC.2010.21, SCOPUS ID：77954773883
• メモリ共有を考慮したRPCシステム　　　　　　　　　　　　　　　
  安積卓也; 大山博司; 高田広章
  情報処理学会論文誌：プログラミング, 巻：2, 号：2, 開始ページ：37-53, 2009年03月
• 組込みシステムに適したコンポーネントシステムの実現と評価　　　　　　　　　　　　　　　
  安積卓也; 山本将也; 小南靖雄; 高木信尚; 鵜飼敬幸; 大山博司; 高田広章
  コンピュータ ソフトウェア, 巻：26, 号：4, 開始ページ：39, 終了ページ：55, 2009年
  Component system suitable for embedded systems and interface generator are presented. This component system adopts a static model which statically instantiates and connects components. The attributes of the components and interface codes for connecting components are statically generated and optimized by the generator. Hence, no instantiation overhead is introduced at runtime and runtime overhead of the interface code is minimized. A serial interface driver is demonstrated as a case study to evaluate the overhead and the software code size. The case study shows the effectiveness of our generator. © 2009, Japan Society for Software Science and Technology. All rights reserved.
  英語, 研究論文（学術雑誌）
  DOI：https://doi.org/10.11309/jssst.26.4_39
  DOI ID：10.11309/jssst.26.4_39, ISSN：0289-6540, SCOPUS ID：85024726401
• Optimization of component connections for an embedded component system　　　　　　　　　　　　　　　
  Takuya Azumi; Hiroaki Takada; Hiroshi Oyama
  Proceedings - 12th IEEE International Conference on Computational Science and Engineering, CSE 2009, 巻：2, 開始ページ：182, 終了ページ：188, 2009年, ［査読有り］
  Software component techniques are widely used to enhance productivity and reduce the cost of software systems development. This paper proposes optimization of component connections for a component system that is suitable for embedded systems. This component system adopts a static model that statically instantiates and connects components. The attributes of the components and the interface code for connecting the components are statically generated by the generator. No instantiation overhead is introduced at runtime, and the runtime overhead of the interface code is minimized. A case study using a serial interface driver is presented to evaluate the execution time overhead, the software code size, and the executable file size. The case study shows the effectiveness of optimization. © 2009 IEEE.
  英語, 研究論文（国際会議プロシーディングス）
  DOI：https://doi.org/10.1109/CSE.2009.97
  DOI ID：10.1109/CSE.2009.97, SCOPUS ID：70749158771
• A Generic Memory Protection Mechanism for An Embedded System and Its Application to Embedded Component Systems　　　　　　　　　　　　　　　
  Shimpei Yamada; Yukikazu Nakamoto; Takuya Azumi; Hiroshi Oyama; Hiroaki Takada
  IEEE 7th International Conference on Computer and Information Technology Workshops, 開始ページ：557, 終了ページ：562, 2008年07月, ［査読有り］
  英語, 研究論文（国際会議プロシーディングス）
  CiNii Articles ID：10026676867
• A memory allocator for efficient task communications by using RPC channels in an embedded component system　　　　　　　　　　　　　　　
  Takuya Azumi; Hiroshi Oyama; Hiroaki Takada
  Proceedings of the 9th IASTED International Conference on Software Engineering and Applications, SEA 2008, 開始ページ：204, 終了ページ：209, 2008年
  The present paper introduces a memory allocator component, two new keywords of component description, and a Remote Procedure Call (RPC) channel in an embedded component system. The memory allocator component is useful for embedded systems because it is possible to select a suitable memory allocator in a variety of memory allocators. The keywords are provided in order to efficiently use memory allocator component. The RPC channel is one of the components used in our component framework to facilitate the transparency of connected components in embedded systems. By supporting the RPC channel, it is possible to reuse the same component on different communication mechanisms without modification, including multiple task communications on the same or different processors and on networks. The effectiveness of the proposed system is demonstrated.
  英語, 研究論文（国際会議プロシーディングス）
  SCOPUS ID：70749123246
• A New Security Framework for Embedded Component Systems　　　　　　　　　　　　　　　
  Takuya Azumi; Shimpei Yamada; Hiroshi Oyama; Yukikazu Nakamoto; Hiroaki Takada
  Proceeding of IASTED Software Engineering and Applications 2007, 開始ページ：584, 終了ページ：589, 2007年11月, ［査読有り］
  英語, 研究論文（国際会議プロシーディングス）
• Visual Modeling Environment for Embedded Component Systems　　　　　　　　　　　　　　　
  Takuya Azumi; Shimpei Yamada; Hiroshi Oyama; Yukikazu Nakamoto; Hiroaki Takada
  Proceeding of IEEE 7th International Conference on Computer and Information Technology, 開始ページ：805, 終了ページ：810, 2007年10月, ［査読有り］
  英語
  CiNii Articles ID：10021844858
• New Environment for Event-Driven Programming　　　　　　　　　　　　　　　
  Takuya Azumi; Sho Narita; Atushi Hasegawa; Nikolay N. Mirenkov
  The Journal of Three Dimensional Images, 巻：21, 号：2, 開始ページ：99-104, 2007年06月
• A new specification of software components for embedded systems　　　　　　　　　　　　　　　
  Takuya Azumi; Masanari Yamamoto; Yasuo Kominami; Nobuhisa Takagi; Hiroshi Oyama; Hiroaki Takada
  10TH IEEE INTERNATIONAL SYMPOSIUM ON OBJECT AND COMPONENT-ORIENTED REAL-TIME DISTRIBUTED COMPUTING, PROCEEDINGS, 開始ページ：46, 終了ページ：+, 2007年, ［査読有り］
  In the last decade, the size and complexity of the software in embedded systems have increased. The present study attempts to decrease the complexity and difficulty of software development in embedded systems. We herein introduce a new component system that is suitable for embedded systems. It is possible to estimate the memory consumption of an entire application since the proposed system adopts a static configuration. In addition, this system takes into account to be used in several domains of embedded systems because several particle sizes of component are supported. Moreover the concept of the component for a distributed application is presented.
  IEEE COMPUTER SOC, 英語, 研究論文（国際会議プロシーディングス）
  Web of Science ID：WOS:000246826300006

• Embedded Component-based Framework for Robot Technology Middleware　　　　　　　　　　　　　　　
  Ryo Hasegawa; Naofumi Yawata; Noriaki Ando; Nobuhiko Nishio; Takuya Azumi
  情報処理学会論文誌, 巻：58, 号：8, 2017年08月15日
  This paper presents a component-based framework for robot technology middleware (RTM) to address real-time issues with RTM. To handle real-time applications, the proposed framework achieves collaboration between RTM and the TOPPERS embedded component system (TECS). TECS is employed to enhance real-time processing in the proposed framework. To implement the collaboration of RTM and TECS, we have adopted remote procedure call. In addition, extending a generator enables the generation of robot technology components from TECS components with source code generated by a model-based development tool such as MATLAB/Simulink. We have evaluated the processor cycle counts of the proposed framework in comparison with those of a conventional method. Moreover, we evaluated the execution time of serial communication and a motor application using the proposed framework. The evaluation results show that the proposed framework is functionally employed in a hard real-time system. Furthermore, we evaluated the amount of code generated by the proposed framework. The evaluation results reveal that the code generated by the proposed framework is reusable and can enhance productivity.------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.25(2017) (online)DOI http://dx.doi.org/10.2197/ipsjjip.25.811------------------------------This paper presents a component-based framework for robot technology middleware (RTM) to address real-time issues with RTM. To handle real-time applications, the proposed framework achieves collaboration between RTM and the TOPPERS embedded component system (TECS). TECS is employed to enhance real-time processing in the proposed framework. To implement the collaboration of RTM and TECS, we have adopted remote procedure call. In addition, extending a generator enables the generation of robot technology components from TECS components with source code generated by a model-based development tool such as MATLAB/Simulink. We have evaluated the processor cycle counts of the proposed framework in comparison with those of a conventional method. Moreover, we evaluated the execution time of serial communication and a motor application using the proposed framework. The evaluation results show that the proposed framework is functionally employed in a hard real-time system. Furthermore, we evaluated the amount of code generated by the proposed framework. The evaluation results reveal that the code generated by the proposed framework is reusable and can enhance productivity.------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.25(2017) (online)DOI http://dx.doi.org/10.2197/ipsjjip.25.811------------------------------
  英語
  ISSN：1882-7764, CiNii Articles ID：170000148865, CiNii Books ID：AN00116647
• 組込みシステムに適した動的メモリアロケータのコンポーネント化　　　　　　　　　　　　　　　
  山本 拓朗; 大山 博司; 安積 卓也
  情報処理学会関西支部支部大会講演論文集, 開始ページ：6p, 2017年
  [情報処理学会関西支部], 日本語
  ISSN：1884-197X, CiNii Articles ID：40021349049, CiNii Books ID：AA12496678
• TINET plus TECS: Component-Based TCP/IP Protocol Stack for Embedded Systems　　　　　　　　　　　　　　　
  Yamamoto Takuro; Hara Takuma; Ishikawa Takuya; Oyama Hiroshi; Takada Hiroaki; Azumi Takuya
  2017 16TH IEEE INTERNATIONAL CONFERENCE ON TRUST, SECURITY AND PRIVACY IN COMPUTING AND COMMUNICATIONS / 11TH IEEE INTERNATIONAL CONFERENCE ON BIG DATA SCIENCE AND ENGINEERING / 14TH IEEE INTERNATIONAL CONFERENCE ON EMBEDDED SOFTWARE AND SYSTEMS, 開始ページ：784-791, 2017年
  英語, 速報，短報，研究ノート等（学術雑誌）
  DOI：https://doi.org/10.1109/Trustcom/BigDataSE/ICESS.2017.313
  DOI ID：10.1109/Trustcom/BigDataSE/ICESS.2017.313, ISSN：2324-9013
• Reservation-Based Scheduling for Automotive DSMS under High Overload Condition　　　　　　　　　　　　　　　
  Jaeyong Rho; Takuya Azumi; Akihiro Yamaguchi; Kenya Sato; Nobuhiko Nishio
  情報処理学会論文誌, 巻：57, 号：8, 2016年08月15日
  Recent automotive systems require various data, including data from on-board sensors and external sources to recognize environmental conditions. As the amount of sensor data used in automotive systems increases, processes that use such data become increasingly complicated. In addition, similar data processing can be duplicated over multiple applications. To address these issues, a data stream management system (DSMS) for automotive systems based on a data integration architecture has been developed. However, hard real-time deadlines cannot be guaranteed due to unpredictable load changes caused by data streams. For example, the arrival time and CPU utilization requested by data streams from vehicle-to-vehicle communications change rapidly depending on environmental conditions. We propose the reservation-based operator path earliest deadline first (ROP-EDF) scheduling algorithm for an automotive DSMS under overload conditions. The proposed algorithm reserves processor time preferentially for hard real-time tasks so that tasks can meet deadlines under overload conditions. ROP-EDF can be used for load testing on a single processor system. Experimental results show the effectiveness of the proposed algorithm compared with existing algorithms relative to the deadline miss ratio under overload conditions.------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.24(2016) No.5 (online)------------------------------Recent automotive systems require various data, including data from on-board sensors and external sources to recognize environmental conditions. As the amount of sensor data used in automotive systems increases, processes that use such data become increasingly complicated. In addition, similar data processing can be duplicated over multiple applications. To address these issues, a data stream management system (DSMS) for automotive systems based on a data integration architecture has been developed. However, hard real-time deadlines cannot be guaranteed due to unpredictable load changes caused by data streams. For example, the arrival time and CPU utilization requested by data streams from vehicle-to-vehicle communications change rapidly depending on environmental conditions. We propose the reservation-based operator path earliest deadline first (ROP-EDF) scheduling algorithm for an automotive DSMS under overload conditions. The proposed algorithm reserves processor time preferentially for hard real-time tasks so that tasks can meet deadlines under overload conditions. ROP-EDF can be used for load testing on a single processor system. Experimental results show the effectiveness of the proposed algorithm compared with existing algorithms relative to the deadline miss ratio under overload conditions.------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.24(2016) No.5 (online)------------------------------
  英語
  ISSN：1882-7764, CiNii Articles ID：170000131016, CiNii Books ID：AN00116647
• OSの機能のTECSプラグインを用いたコンポーネント化 (ディペンダブルコンピューティング) -- (組込み技術とネットワークに関するワークショップETNET2016)
  河田 智明; 安積 卓也; 大山 博司; 高田 広章
  電子情報通信学会技術研究報告 = IEICE technical report : 信学技報, 巻：115, 号：519, 開始ページ：31, 終了ページ：36, 2016年03月24日
  電子情報通信学会, 日本語
  ISSN：0913-5685, CiNii Articles ID：40020791136
• Bluetoothを用いたマルチVM対応mrubyバイトコードローダ (コンピュータシステム) -- (組込み技術とネットワークに関するワークショップETNET2016)
  山本 拓朗; 大山 博司; 安積 卓也
  電子情報通信学会技術研究報告 = IEICE technical report : 信学技報, 巻：115, 号：518, 開始ページ：43, 終了ページ：48, 2016年03月24日
  電子情報通信学会, 日本語
  ISSN：0913-5685, CiNii Articles ID：40020790007
• プロセッサ故障にともなう過負荷状態の回避を考慮した最適出力フィードバック制御器の設計　　　　　　　　　　　　　　　
  吉本 達也; 潮 俊光; 安積 卓也
  情報処理学会論文誌, 巻：56, 号：8, 開始ページ：1560, 終了ページ：1567, 2015年08月15日
  組込みシステムにおいて，プロセッサの故障にともなう過負荷状態を回避し，制御性能の劣化を最小化することはシステムの信頼性，安全性を保つうえで重要である．プロセッサが過負荷状態に陥ると，デッドラインを満たせないジョブが発生し，それらのジョブの実行に時間を要したために他のジョブのデッドラインミスにも影響する．そこで，過負荷状態において受理するジョブと破棄するジョブを選択し，受理されたジョブのデッドラインが必ず満たされる方法としてジョブスキッピングが考える．しかし，従来研究ではジョブスキッピングパターンを決定するものの，制御性能の劣化を抑える制御器の設計に関してはあまり検討されていない．本論文では．ジョブスキッピングに基づくスケジューリングを用いたリアルタイム制御システム上の最適出力フィードバック制御器の設計を行う．シミュレーションでは，ジョブスキッピングが発生したときに従来の最適出力フィードバック制御器に比べて提案する制御器によって制御性能の劣化が抑えられることを示す．In embedded control systems, it is important to avoid an overload situation and minimize a degradation of control performance caused by a failure of a processor for keeping reliability and safety. If the processor is under the overload situation, several jobs cannot meet their deadlines and their fruitless execution compress the other jobs' deadlines. Then, we consider a job skipping approach which is a policy to classify released jobs into accepted and skipped jobs to meet deadlines of all accepted jobs. However, conventional approaches had not addressed a design that suppresses an effect of job skipping. In this paper, we design an optimal output-feedback controller on a real-time system with a scheduler using the job skipping policy. By a numerical simulation, we show that the proposed optimal controller suppresses the degradation of control performance when jobs are skipped.
  日本語
  ISSN：1882-7764, CiNii Articles ID：170000130683, CiNii Books ID：AN00116647
• 組込みコンポーネントシステムの呼び出しフロー解析ツールの開発 (コンピュータシステム)　　　　　　　　　　　　　　　
  成瀬 有美; 石川 拓也; 安積 卓也; 大山 博司; 高田 広章
  電子情報通信学会技術研究報告 = IEICE technical report : 信学技報, 巻：114, 号：506, 開始ページ：237, 終了ページ：242, 2015年03月06日
  組込みソフトウェア向けのコンポーネントベース開発を支援するコンポーネントシステムにTECSがある. TECSは,コンポーネント記述に独自のコンポーネント記述言語(TECS CDL),実装言語にC言語を採用している.一般的にソフトウェアは,フロー解析を用いて検証を行うことで,信頼性の向上が期待できる. TECSを利用して開発したシステム全体のフロー解析を行うためには,C言語のフロー解析に加え, CDLの解析を行ってC言語の解析結果と統合を行う必要がある.そのため,TECS向けのフロー解析を行う開発支援ツールTECS-flowを開発した.本ツールによって,TECSを使用して開発したシステム全体のフローを可視化し,コンポーネント間の関数呼び出し順序を知ることができる.加えて,排他制御が必要な箇所の検出やデッドコード検出も可能である.なお,本研究の一部はenPiTプロジェクトの一環として実施した,
  一般社団法人電子情報通信学会, 日本語
  ISSN：0913-5685, CiNii Articles ID：110010021774, CiNii Books ID：AN10013141
• 組込みコンポーネントシステムの呼び出しフロー解析ツールの開発 (ディペンダブルコンピューティング)　　　　　　　　　　　　　　　
  成瀬 有美; 石川 拓也; 安積 卓也; 大山 博司; 高田 広章
  電子情報通信学会技術研究報告 = IEICE technical report : 信学技報, 巻：114, 号：507, 開始ページ：237, 終了ページ：242, 2015年03月06日
  組込みソフトウェア向けのコンポーネントベース開発を支援するコンポーネントシステムにTECSがある. TECSは,コンポーネント記述に独自のコンポーネント記述言語(TECS CDL),実装言語にC言語を採用している.一般的にソフトウェアは,フロー解析を用いて検証を行うことで,信頼性の向上が期待できる. TECSを利用して開発したシステム全体のフロー解析を行うためには, C言語のフロー解析に加え,CDLの解析を行ってC言語の解析結果と統合を行う必要がある.そのため,TECS向けのフロー解析を行う開発支援ツールTECS-flowを開発した.本ツールによって,TECSを使用して開発したシステム全体のフローを可視化し,コンポーネント間の関数呼び出し順序を知ることができる.加えて,排他制御が必要な箇所の検出やデッドコード検出も可能である.なお,本研究の一部はenPiTプロジェクトの一環として実施した,
  一般社団法人電子情報通信学会, 日本語
  ISSN：0913-5685, CiNii Articles ID：110010021464, CiNii Books ID：AA11645397
• 組込みコンポーネントシステムの呼び出しフロー解析ツールの開発　　　　　　　　　　　　　　　
  成瀬 有美; 石川 拓也; 安積 卓也; 大山 博司; 高田 広章
  研究報告組込みシステム（EMB）, 巻：2015, 号：27, 開始ページ：1, 終了ページ：6, 2015年02月27日
  組込みソフトウェア向けのコンポーネントベース開発を支援するコンポーネントシステムに TECS がある．TECS は，コンポーネント記述に独自のコンポーネント記述言語 （TECS CDL），実装言語に C 言語を採用している．一般的にソフトウェアは，フロー解析を用いて検証を行うことで，信頼性の向上が期待できる．TECS を利用して開発したシステム全体のフロー解析を行うためには，C 言語のフロー解析に加え，CDL の解析を行って C 言語の解析結果と統合を行う必要がある．そのため，TECS 向けのフロー解析を行う開発支援ツール TECS-flow を開発した．本ツールによって，TECS を使用して開発したシステム全体のフローを可視化し，コンポーネント間の関数呼び出し順序を知ることができる．加えて，排他制御が必要な箇所の検出やデッドコード検出も可能である．なお，本研究の一部は enPiT プロジェクトの一環として実施した．
  一般社団法人情報処理学会, 日本語
  ISSN：0919-6072, CiNii Articles ID：110009881315, CiNii Books ID：AA12149313
• メモリ保護を考慮した組込みシステム向けソフトウェアコンポーネント技術　　　　　　　　　　　　　　　
  石川拓也; 安積卓也; 大山博司; 本田晋也; 高田広章
  コンピュータソフトウェア (日本ソフトウェア科学会論文誌), 巻：31, 号：4, 開始ページ：105-130, 終了ページ：4_130, 2014年11月, ［査読有り］
  本論文では，メモリ保護が必要な組込みソフトウェアの開発を支援するコンポーネント技術HR-TECSについて述べる． HR-TECSでは，組込みソフトウェア向けのコンポーネント技術であるTECSを拡張することで，メモリ保護を考慮した組込みソフトウェアのコンポーネントベース開発を可能とし，メモリ保護機能を持ったリアルタイムOSの機能を利用して，メモリ保護機能を実現する．また，コンポーネント間の通信について，個々のコンポーネントにおける通信APIの記述は，それらのコンポーネントが配置される保護領域に依存せず，共通のAPIで開発できる．さらに，HR-TECSは，非特権モードで動作するアプリケーションだけでなく，特権モードで動作するデバイスドライバなどの開発にも利用できる．本論文では，メモリ保護機能を持ったリアルタイムOSであるTOPPERS/HRP2カーネルを利用した場合のHR-TECSの実現方法を示し，評価実験により，HR-TECSを用いることによって生じる実行時間のオーバヘッドが小さく，かつ，予測が可能であること，および，メモリ使用量のオーバヘッドが小さいことを示す．
  Japan Society for Software Science and Technology, 日本語, 速報，短報，研究ノート等（学術雑誌）
  DOI：https://doi.org/10.11309/jssst.31.4_105
  DOI ID：10.11309/jssst.31.4_105, ISSN：0289-6540, CiNii Articles ID：130004706727
• 分散インテントプラグイン:組込み機器向け分散インテントの自動生成機構
  長原 裕希; 大山 博司; 安積 卓也; 西尾 信彦
  組込みシステムシンポジウム2013論文集, 巻：2013, 開始ページ：114, 終了ページ：122, 2013年10月09日
  日本語
  CiNii Articles ID：170000078521
• 広場通過を考慮した屋内ナビゲーションの手法検討　　　　　　　　　　　　　　　
  PhanDuyPhuoc; 宮島 智大; 枇榔 晃裕; 安積 卓也; 西尾 信彦
  研究報告ユビキタスコンピューティングシステム（UBI）, 巻：2013, 号：56, 開始ページ：1, 終了ページ：6, 2013年05月09日
  近年，ナビゲーションのような多くの移動支援サービスが提供されており，屋外だけでなく，測位インフラの整っていない地下街など屋内にもサービスの範囲が広がってきている．地下街など屋内のナビゲーションでは通路よりも広場の通過が問題となる．広場は出入口が多い上に，目印となるランドマークが少なく，ナビゲーションに必要な道路ネットワークは自動生成すると複雑なものとなる傾向がある．本稿では，広場内を最小限の曲がり角で表現する道路ネットワークの自動生成と，天井吊り案内標示を利用した広場通過のナビゲーションの必要条件を検討する．Recently, many navigation services have been provided. These services have been used widespreadly not only outdoor, but also indoor where users can use no positioning technology. As the indoor environment such as a underground center, non-corridor area caused of problems rather than passages. Indoor non-corridor area have many gateways, nevertheless, there are few landmarks. In addition, it is difficult to make road network for a navigation. This paper proposes the mean to generate a road network which has the minimum junction for navigation automatically, and to navigate in an open space with ceiling guidance labels.
  一般社団法人情報処理学会, 日本語
  ISSN：0913-5685, CiNii Articles ID：110009579772, CiNii Books ID：AA11838947
• 屋内細粒度行動認識のためのWi-Fi標本の自動管理機構　　　　　　　　　　　　　　　
  藤井 陽光; 坂口 拓嗣; 安積 卓也; 西尾 信彦
  研究報告ユビキタスコンピューティングシステム（UBI）, 巻：2013, 号：57, 開始ページ：1, 終了ページ：7, 2013年05月09日
  近年屋内では利用が困難な GPS を代替するため，Wi-Fi の信号強度を利用した測位手法が注目されている．しかし，この手法では Wi-Fi 基地局の緯度経度情報などの管理作業に，多大なコストが必要である．この問題点を解決するため，我々は先行研究として現在地を識別するのみにとどまるが，Wi-Fi 基地局の統計的特徴量から生成される標本を利用することでコストの大きな管理作業を必要としない測位手法を提案してきた．しかし，長期の観測では同一室内の多数の地点が標本化されており，標本化された地点間で互いの識別を困難にする問題や，断続的にしか観測されない Wi-Fi 基地局のためユーザ測位の際にノイズとなる問題がある．本研究では，近在する標本をひとくくりにすることで標本を適切に生成，アップデートする手法を提案し，評価をおこなった。In recent years, Studies on localization methods using Wi-Fi signal strength have been conducted to substitute for GPS which is not proper to indoor use.Unfortunately, this method costs a lot of money and takes a lot of works to collect information such as latitude and longitude of Wi-Fi base stations. To solve those problems, We proposed Indoor localization method which identify the current position but does not require lots of works. There were many points to be sampled when we measured in the same room by long term. but there were some problems of recognition the location of point with some standardized points and occuring noises by the intermittent Wi-Fi transmissions. in this paper, we propose a method to integrate the nearby samples, and evaluate it.
  一般社団法人情報処理学会, 日本語
  ISSN：0913-5685, CiNii Articles ID：110009579773, CiNii Books ID：AA11838947
• 屋内細粒度行動認識のためのWi-Fi標本の自動管理機構(位置情報・行動認識,フレッシュマンセッション及び一般)　　　　　　　　　　　　　　　
  藤井 陽光; 坂口 拓嗣; 安積 卓也; 西尾 信彦
  電子情報通信学会技術研究報告. ASN, 知的環境とセンサネットワーク, 巻：113, 号：38, 開始ページ：319, 終了ページ：325, 2013年05月09日
  近年屋内では利用が困難なGpsを代替するため,Wi-Fiの信号強度を利用した測位手法が注目されている.しかし,この手法ではWi-Fi基地局の緯度経度情報などの管理作業に,多大なコストが必要である.この問題点を解決するため,我々は先行研究として現在地を識別するのみにとどまるが,Wi-Fi基地局の統計的特徴量から生成される標本を利用することでコストの大きな管理作業を必要としない測位手法を提案してきた.しかし,長期の観測では同一室内の多数の地点が標本化されており,標本化された地点間で互いの識別を困難にする問題や,断続的にしか観測されないWi-Fi基地局のためユーザ測位の際にノイズとなる問題がある.本研究では,近在する標本をひとくくりにすることで標本を適切に生成,アップデートする手法を提案し,評価をおこなった。
  一般社団法人電子情報通信学会, 日本語
  CiNii Articles ID：110009771496, CiNii Books ID：AA12618979
• 広場通過を考慮した屋内ナビゲーションの手法検討(位置情報・行動認識,フレッシュマンセッション及び一般)　　　　　　　　　　　　　　　
  PHUOC PHAN DUY; 宮島 智大; 枇榔 晃裕; 安積 卓也; 西尾 信彦
  電子情報通信学会技術研究報告. ASN, 知的環境とセンサネットワーク, 巻：113, 号：38, 開始ページ：313, 終了ページ：318, 2013年05月09日
  近年,ナビゲーションのような多くの移動支援サービスが提供されており,屋外だけでなく,測位インフラの整っていない地下街など屋内にもサービスの範囲が広がってきている.地下街など屋内のナビゲーションでは通路よりも広場の通過が問題となる.広場は出入口が多い上に,目印となるランドマークが少なく,ナビゲーションに必要な道路ネットワークは自動生成すると複雑なものとなる傾向がある.本稿では,広場内を最小限の曲がり角で表現する道路ネットワークの自動生成と,天井吊り案内標示を利用した広場通過のナビゲーションの必要条件を検討する.
  一般社団法人電子情報通信学会, 日本語
  CiNii Articles ID：110009771495, CiNii Books ID：AA12618979
• GPU制御用ファームウェア開発環境
  藤居 祐輔; 安積 卓也; 西尾 信彦; 加藤 真平
  組込みシステムシンポジウム2012論文集, 巻：2012, 開始ページ：136, 終了ページ：143, 2012年10月10日
  日本語
  CiNii Articles ID：170000072425
• OpaqueRPCを用いたRTMとTECS間通信の実現
  八幡 尚文; 安積 卓也; 西尾 信彦
  組込みシステムシンポジウム2012論文集, 巻：2012, 開始ページ：170, 終了ページ：177, 2012年10月10日
  日本語
  CiNii Articles ID：170000072429
• 複数センサーを用いた屋内細粒度行動認識の自動化　　　　　　　　　　　　　　　
  西尾 信彦; 藤井 陽光; 安積 卓也
  電子情報通信学会技術研究報告. USN, ユビキタス・センサネットワーク : IEICE technical report, 巻：112, 号：31, 開始ページ：79, 終了ページ：84, 2012年05月10日
  ユーザの細粒度での屋内行動を認識するために,複数のセンサーを用い普段滞在する居場所を自動的に学習し標本化する機構と,それを用い現在の状況を標本から抽出する機構を実装し評価した.その結果,新規の標本を生成すべきか,既存の標本をアップデートすべきかの閾値を獲得する必要があることがわかり,そのための検証実験を実施し,比較的短期間のセンシングでも自動的に標本をアップデートして,ほぼ0.1%未満の誤認識率に押えられる見込みが得られた.
  一般社団法人電子情報通信学会, 日本語
  ISSN：0913-5685, CiNii Articles ID：110009569473, CiNii Books ID：AA1221489X
• カリフォルニアの研究生活(<シリーズ特集>世界のAI,日本のAI〔第21回〕)　　　　　　　　　　　　　　　
  安積 卓也; Takuya Azumi; College of Information Science and Engineering Ritsumeikan University
  人工知能学会誌 = Journal of Japanese Society for Artificial Intelligence, 巻：27, 号：2, 開始ページ：207, 終了ページ：210, 2012年03月01日
  人工知能学会, 日本語
  ISSN：0912-8085, CiNii Articles ID：110009419563, CiNii Books ID：AN10067140
• HW/SW Cosimulation Framework Based on Software Component System (組込みシステム(EMB) Vol.2011-EMB-20)
  AZUMI TAKUYA; HARA-AZUMI YUKO; HONDA SHINYA
  情報処理学会研究報告, 巻：2010, 号：6, 開始ページ：1, 終了ページ：6, 2011年04月
  情報処理学会, 英語
  ISSN：2186-2583, CiNii Articles ID：40019913882
• HW/SW cosimulation framework based on software component system (コンピュータシステム)
  Azumi Takuya; Hara-Azumi Yuko; Honda Shinya
  電子情報通信学会技術研究報告, 巻：110, 号：473, 開始ページ：261, 終了ページ：266, 2011年03月18日
  電子情報通信学会, 英語
  ISSN：0913-5685, CiNii Articles ID：40018763372
• Soft error-aware scheduling in high-level synthesis (コンピュータシステム)
  Hara-Azumi Yuko; Tomiyama Hiroyuki; Azumi Takuya
  電子情報通信学会技術研究報告, 巻：110, 号：473, 開始ページ：111, 終了ページ：116, 2011年03月18日
  電子情報通信学会, 英語
  ISSN：0913-5685, CiNii Articles ID：40018763347
• Soft Error-Aware Scheduling in High-Level Synthesis (組込みシステム(EMB) Vol.2011-EMB-20)　　　　　　　　　　　　　　　
  YukoHara-Azumi; Hiroyuki Tomiyama; Takuya Azumi; Shigeru Yamashita; NikilD.Dutt; Hiroaki Takada
  研究報告システムLSI設計技術（SLDM）, 巻：2011, 号：19, 開始ページ：1, 終了ページ：6, 2011年03月11日
  Due to the continuous reduction in chip feature size and supply voltage, soft errors are becoming a serious problem in the today's LSI design. This paper proposes a soft error-aware scheduling method in high-level synthesis. The reliability of the datapath circuit is determined not only by those of its computations, which depend on their assigned hardware resources, but also by those of its values, which are affected by their lifetime length. By considering both influences, our proposed method schedules operations for maximizing the reliability of the datapath circuit.Due to the continuous reduction in chip feature size and supply voltage, soft errors are becoming a serious problem in the today's LSI design. This paper proposes a soft error-aware scheduling method in high-level synthesis. The reliability of the datapath circuit is determined not only by those of its computations, which depend on their assigned hardware resources, but also by those of its values, which are affected by their lifetime length. By considering both influences, our proposed method schedules operations for maximizing the reliability of the datapath circuit.
  情報処理学会, 英語
  ISSN：2186-2583, CiNii Articles ID：110008583300, CiNii Books ID：AA11451459
• HW/SW Cosimulation Framework Based on Software Component System (システムLSI設計技術(SLDM) Vol.2011-SLDM-149)　　　　　　　　　　　　　　　
  Takuya Azumi; YukoHara-Azumi; Shinya Honda; Hiroaki Takada
  研究報告システムLSI設計技術（SLDM）, 巻：2011, 号：44, 開始ページ：1, 終了ページ：6, 2011年03月11日
  As the size and complexity of software applications increase, software component systems are getting used due to its reusability, reduced time-to-market, and hence, reduced software production cost. This paper proposes SysTECS, a highly automated hardware/software (HW/SW) cosimulation framework which integrates a system-level design tool, SystemBuilder, and a software component system for embedded systems, TECS. In existing HW/SW codesign technologies, designers need to manually add or modify HW/SW communication sources (e.g., their size, direction, and allocator) in input behavioral descriptions, which is complex to specify. On the contrary, in SysTECS, the designers can design the overall system at higher abstraction level and have no need to specify the HW/SW communication in the input descriptions because TECS specifically defines the interface between components and the communication sources are automatically generated. This paper demonstrates the effectiveness of Sys- TECS through a case study with an SHA application.As the size and complexity of software applications increase, software component systems are getting used due to its reusability, reduced time-to-market, and hence, reduced software production cost. This paper proposes SysTECS, a highly automated hardware/software (HW/SW) cosimulation framework which integrates a system-level design tool, SystemBuilder, and a software component system for embedded systems, TECS. In existing HW/SW codesign technologies, designers need to manually add or modify HW/SW communication sources (e.g., their size, direction, and allocator) in input behavioral descriptions, which is complex to specify. On the contrary, in SysTECS, the designers can design the overall system at higher abstraction level and have no need to specify the HW/SW communication in the input descriptions because TECS specifically defines the interface between components and the communication sources are automatically generated. This paper demonstrates the effectiveness of Sys- TECS through a case study with an SHA application.
  情報処理学会, 英語
  ISSN：2186-2583, CiNii Articles ID：110008583324, CiNii Books ID：AA11451459
• UMLモデルのC言語実装におけるTECSの適用事例　　　　　　　　　　　　　　　
  石川 拓也; 安積 卓也; 一場 利幸; 柴田 誠也; 高田 広章
  情報処理学会論文誌プログラミング（PRO）, 巻：3, 号：4, 開始ページ：59, 終了ページ：59, 2010年09月22日
  本発表では，UMLによって設計された組み込みソフトウェアをC言語で実装する過程の事例として，組み込みシステム向けコンポーネントシステムを用いた倒立二輪ライントレースロボットの制御ソフトウェアの開発について述べる．組み込みソフトウェアは，大規模化・複雑化が進んでおり，生産性の向上が求められている．C++やJavaなどのオブジェクト指向言語を用いたソフトウェア開発では，生産性を向上させるために，UMLを用いた設計が行われている．UMLはオブジェクト指向を意識したモデル言語であるため，クラス図などのUMLモデルからソースコードの自動生成が容易で，モデルとソースコードとの一貫性を保ちやすい．一方，組み込みシステムのソフトウェア開発では，実装言語として手続き型言語であるC言語が広く用いられている．C言語はオブジェクト指向言語ではないため，UMLモデルからC言語ソースコードを自動生成するツールを利用したとしても，モデルとソースコードの間に隔たりがある．この問題を解決するために，組み込みシステムに適したコンポーネントシステムであるTECSを利用した．TECSは独自のコンポーネント図を入力とし，コンポーネントの結合部分および，コンポーネントの提供する機能のテンプレートのC言語ソースコードを自動生成する．自動生成により，コンポーネント図とソースコードの一貫性を保つことができる．本事例により，UMLモデルのクラス図からTECSのコンポーネント図への変換方法および，コンポーネント図とソースコードの一貫性を保持できることを示す．In this presentation, we will describe the development of control software for a two-wheeled self-balancing and line-tracing robot using a component system for embedded systems, as the case study of an embedded software design in UML and an implementation in C. Embedded systems have become large-scale and complex. Therefore, improving software productivity has been demanded. In software development using an object-oriented language, UML is often used in software design to improve software productivity. Since UML is based on object-oriented concepts, it is easy to automatically generate source codes from a UML model. As a result, it is easy to keep a consistency between a UML model and source codes. Meanwhile, in embedded software development where C language is generally used, UML has not been an effective solution for improving productivity because C language is not the object-oriented language. There can be gaps between a UML model and C source codes, even if automatic generation tools are used. To fill the gaps, we have taken advantage of TECS, which is a component system suitable for embedded systems. TECS takes an original component diagram as an input, and generates C source codes, that describe connections between components and templates of functions that components provide. Therefore, it is possible to keep a consistency between a TECS component diagram and source codes. This case study shows a way to transform a UML class diagram into a TECS component diagram, and that TECS enables to keep the consistency.
  日本語
  ISSN：1882-7802, CiNii Articles ID：110007970958, CiNii Books ID：AA11464814
• History of summer school on embedded system technologies organized by students and young engineers
  Hideki Takase; Takuya Azumi; Ittetsu Taniguchi; Yutaka Matsubara; Hayato Kanai; Shintaro Hosoai; Midori Sugaya
  Proceedings - 2009 Workshop on Embedded Systems Education, WESE 2009, 開始ページ：19, 終了ページ：26, 2009年12月01日
  This paper reports an embedded educational activity named Summer School on Embedded System Technologies (SSEST) in Japan. We, the committee members made up of students and young engineers gathering from all over the country, set down beginners as main targets since they have few chance to experience a series of processes of an embedded system development. This activity aims to provide an opportunity to learn elementary knowledge and skills of embedded system technologies through a practice of whole development process. Communication skills of participants can be enhanced by an interchange among them of different universities and companies through a group work. In this paper, we introduce a learning curriculum and contents of SSEST. Our curriculum consists of prior individual training and boot camp. Participants develop a line trace car through the embedded system development process. Then, we discuss the educational effects on this activity and achievements of the purpose. According to answers of the questionnaire, the 94% of participants satisfied or almost satisfied for SSEST. © ACM 2009.
  英語
  DOI：https://doi.org/10.1145/1719010.1719015
  Scopus：https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=77950923697&origin=inward
  Scopus Citedby：https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=77950923697&origin=inward
  DOI ID：10.1145/1719010.1719015, SCOPUS ID：77950923697
• A realization of RPC in embedded component systems (システムLSI設計技術・組込みシステム・組込技術とネットワークに関するワークショップETNET2008)　　　　　　　　　　　　　　　
  Takuya Azumi; Hiroshi Oyama; Hiroaki Takada
  情報処理学会研究報告システムLSI設計技術（SLDM）, 巻：2008, 号：32, 開始ページ：97, 終了ページ：102, 2008年03月27日
  This paper presents a Remote Procedure Call (RPC) channel in an embedded component system. The RPC channel is one of the components used in our component framework to facilitate the transparency of connected components in embedded systems. By supporting the RPC channel it is possible to reuse the same component with out modification on different communication mechanisms including different task communications on the same or different processors and on networks. In addition a memory allocator component and new keywords of a component description are provided. The effectiveness of our work is demonstrated.This paper presents a Remote Procedure Call (RPC) channel in an embedded component system. The RPC channel is one of the components used in our component framework to facilitate the transparency of connected components in embedded systems. By supporting the RPC channel, it is possible to reuse the same component with out modification on different communication mechanisms, including different task communications on the same or different processors and on networks. In addition, a memory allocator component and new keywords of a component description are provided. The effectiveness of our work is demonstrated.
  一般社団法人情報処理学会, 英語
  ISSN：0919-6072, CiNii Articles ID：110006820438, CiNii Books ID：AA11451459
• 組込みコンポーネントシステム向け保護機構の開発　　　　　　　　　　　　　　　
  山田 晋平; 中本 幸一; 安積 卓也; 高田 広章; 大山 博司
  情報処理学会研究報告システムLSI設計技術（SLDM）, 巻：2008, 号：32, 開始ページ：91, 終了ページ：96, 2008年03月27日
  近年，ネットワークや汎用コンピュータシステムとの通信，統合が必要な組込みシステムが増加したため，セキュリティを強化する機構が必要とされている．また，大規模化・複雑化する組込みシステム開発に対して，コンポーネントの使用による開発効率の向上が期待されている．このような背景から，本稿では組込みコンポーネントシステム用にアクセス制御機構とメモリ保護機構を開発する．アクセス制御機構は主に，アクセス制御を行うセキュリティコンポーネントと，アクセスの可否を判断するリファレンスモニタから成る．メモリ保護機構は，複数領域や入れ子構造の保護や，動的なメモリ保護パターンの変更が可能など柔軟な設計となっている．Recently, embedded systems connected to network have increased; therefore it is needed to strengthen security. In addition, software component-based development is expected improvement of embedded software in productivity. We present design of security enhancement for embedded component system, access control and memory protection. Firstly, Access control controls what a subject access to an object by the access rule. We design and implement the access control mechanism for embedded component system. The second one is memory protection mechanisms. We also design the generic memory protection mechanism including the nested memory protection and the dynamic protection pattern change.
  一般社団法人情報処理学会, 日本語
  ISSN：0919-6072, CiNii Articles ID：110006820437, CiNii Books ID：AA11451459
• A realization of RPC in embedded component systems (ディペンダブルコンピューティング・組込技術とネットワークに関するワークショップETNET2008)　　　　　　　　　　　　　　　
  Azumi Takuya; Oyama Hiroshi; Takada Hiroaki
  電子情報通信学会技術研究報告. DC, ディペンダブルコンピューティング, 巻：107, 号：559, 開始ページ：97, 終了ページ：102, 2008年03月20日
  This paper presents a Remote Procedure Call (RFC) channel in an embedded component system. The RPC channel is one of the components used in our component framework to facilitate the transparency of connected components in embedded systems. By supporting the RPC channel, it is possible to reuse the same component without modification on different communication mechanisms, including different task communications on the same or different processors and on networks. In addition, a memory allocator component and new keywords of a component description are provided. The effectiveness of our work is demonstrated.
  一般社団法人電子情報通信学会, 英語
  ISSN：0913-5685, CiNii Articles ID：110006781737, CiNii Books ID：AA11645397
• 組込みコンポーネントシステム向け保護機構の開発　　　　　　　　　　　　　　　
  山田 晋平; 中本 幸一; 安積 卓也; 高田 広章; 大山 博司
  電子情報通信学会技術研究報告. DC, ディペンダブルコンピューティング, 巻：107, 号：559, 開始ページ：91, 終了ページ：96, 2008年03月20日
  近年,ネットワークや汎用コンピュータシステムとの通信,統合が必要な組込みシステムが増加したため,セキュリティを強化する機構が必要とされている.また,大規模化・複雑化する組込みシステム開発に対して,コンポーネントの使用による開発効率の向上が期待されている.このような背景から,本稿では組込みコンポーネントシステム用にアクセス制御機構とメモリ保護機構を開発する.アクセス制御機構は主に,アクセス制御を行うセキュリティコンポーネントと,アクセスの可否を判断するリファレンスモニタから成る.メモリ保護機構は,複数領域や入れ子構造の保護や,動的なメモリ保護パターンの変更が可能など柔軟な設計となっている.
  一般社団法人電子情報通信学会, 日本語
  ISSN：0913-5685, CiNii Articles ID：110006781736, CiNii Books ID：AA11645397

• 自動運転　　　　　　　　　　　　　　　
  コロナ社, 2021年01月
  日本語, 総ページ数：xi, 273p
  CiNii Books：http://ci.nii.ac.jp/ncid/BC04848058
  ISBN：9784339027754, CiNii Books ID：BC04848058
• Interface 2011年7月号 コンポーネントで組み上げる MINDSTORMS NXT用プラットホームTOPPERS/ASP+TECSを使いこなす　　　　　　　　　　　　　　　
  石川拓也, 147-157
  CQ出版社, 2011年05月
• 組込みプレスvol.15:OSSの組込みコンポーネントシステム TECSを使ってみよう　　　　　　　　　　　　　　　
  大山博司, 137-145
  技術評論社, 2009年05月

• Mapping Method of MATLAB/Simulink Model for Embedded Many-Core Platform　　　　　　　　　　　　　　　
  Kentaro Honda; Sasuga Kojima; Hiroshi Fujimoto; Masato Edahiro; Takuya Azumi
  the 28th Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP 2020), 2020年03月, ［国際会議］
  英語, 口頭発表（一般）
• Accurate Contention Estimate Scheduling Method Using Multiple Clusters of Many-core Platform for Applications Requiring Parallel Computations　　　　　　　　　　　　　　　
  Shingo Igarashi; Yuto Kitagawa; Takuro Fukunaga; Takuya Azumi
  the 28th Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP 2020), 2020年03月, ［国際会議］
  英語, 口頭発表（一般）
• Autoware Toolbox: MATLAB/Simulink Benchmark Suite for ROS-based Self-driving Software Platform　　　　　　　　　　　　　　　
  Keita Miura; Shota Tokunaga; Noriyuki Ota; Yoshiharu Tange; Takuya Azumi
  the 30th International Workshop on Rapid System Prototyping (RSP), 2019年10月, ［国際会議］
  英語, 口頭発表（一般）
• Multi-rate DAG Scheduling Considering Communication Contention for NoC-based Embedded Many-core Processor　　　　　　　　　　　　　　　
  Shingo Igarashi; Yuto Kitagawa; Tasuku Ishigooka; Tatsuya Horiguchi; Takuya Azumi
  the 23rd International Symposium on Distributed Simulation and Real Time Applications (DS-RT2019), 2019年09月, ［国際会議］
  英語, 口頭発表（一般）
• erformance Estimation for Many-core Processor in Model-Based Development　　　　　　　　　　　　　　　
  Kentaro Honda; Takuya Azumi
  the 8th Mediterranean Conference on Embedded Computing (MECO2019), 2019年06月, ［国際会議］
  英語, 口頭発表（一般）
• System on a Chip and Software Platforms for Fully Autonomous Driving　　　　　　　　　　　　　　　
  Hiroyuki Chishiro; Kazutoshi Suito; Tsutomu Ito; Seiya Maeda; Takuya Azumi; Kenji Funaoka; Shinpei Kato
  he IEEE 15th International Conference on Embedded Software and Systems (ICESS2019), 2019年06月, ［招待有り］, ［国際会議］
  英語, 口頭発表（一般）
• Autoware Toolbox: MATLAB/Simlink Benchmark Suite for ROS-based Self-driving Software Platform　　　　　　　　　　　　　　　
  Shota Tokunaga; Keita Miura; Takuya Azumi
  the 21st IEEE International Symposium on Real-time Computing (ISORC 2018), 2019年05月, ［国際会議］
  英語, 口頭発表（一般）
• Demo Abstract: MATLAB/Simulink Benchmark Suite for ROS-based Self-driving System　　　　　　　　　　　　　　　
  Shota Tokunaga; Noriyuki Ota; Yoshiharu Tange; Keita Miura; Takuya Azumi
  Demo Session of the 10th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS2019), 2019年04月, ［国際会議］
  英語, ポスター発表
• Resource Manager for Scalable Performance in ROS Distributed Environments　　　　　　　　　　　　　　　
  Daisuke Fukutomi; Takuya Azumi; Shinpei Kato; Nobuhiko Nishi
  the Design, Automation and Test in Europe (DATE 2019), 2019年03月, ［国際会議］
  英語, 口頭発表（一般）
• IDF-Autoware: Integrated Development Framework for ROS-based Self-driving Systems Using MATLAB/Simulink　　　　　　　　　　　　　　　
  Workshop of Autonomous; Systems Design (A
  2019年03月, ［国際会議］
  英語
• Runtime Component Information on Embedded Component Systems　　　　　　　　　　　　　　　
  Seito Shirata; Hiroshi Oyama; Takuya Azumi
  In Proceedings of the 16th IEEE International Conference on Embedded and Ubiquitous Computing (EUC 2018), 2018年10月, ［国際会議］
  英語, 口頭発表（一般）
• DAG Scheduling Algorithm for a Cluster-Based Many-Core Architecture　　　　　　　　　　　　　　　
  Yuto Kitagawa; Tasuku Ishigooka; Takuya Azumi
  n Proceedings of the 16th IEEE International Conference on Embedded and Ubiquitous Computing (EUC 2018), 2018年10月, ［国際会議］
  英語, 口頭発表（一般）
• ROSCH: Real-Time Scheduling Framework for RO　　　　　　　　　　　　　　　
  Yukihiro Saito; Futoshi Sato; Takuya Azumi; Shinpei Kato; Nobuhiko Nishio
  In Proceedings of the 24th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA 2018), 2018年08月, ［国際会議］
  英語, 口頭発表（一般）
• DAG Scheduling Algorithm Considering Large-scalecalculation Tasks Using Many-core Architecture　　　　　　　　　　　　　　　
  Yuto Kitagawa; Takuya Azumi
  the 9th International Real-Time Scheduling Open Problems Seminar (RTSOPS 2018), 2018年07月, ［国際会議］
  英語, 口頭発表（一般）
• Real-Time ROS Extension on Transparent CPU/GPU Coordination Mechanism　　　　　　　　　　　　　　　
  Yuhei Suzuki; Takuya Azumi; Shinpei Kato; Nobuhiko Nishio
  In Proceedings of the 21st IEEE International Symposium on Real-time Computing (ISORC 2018), 2018年05月, ［国際会議］
  英語, 口頭発表（一般）
• Demo Abstract: MATLAB/Simulink Benchmark Suite for ROS-based Self-driving System　　　　　　　　　　　　　　　
  Shota Tokunaga; Noriyuki Ota; Yoshiharu Tange; Keita Miura; Takuya Azumi
  2018年04月, ［国際会議］
  英語, ポスター発表
• ROS-based Support System for Supervision of Multiple UAVs by a Single Operator　　　　　　　　　　　　　　　
  Hiroki Hayakawa; Takuya Azumi; Akinori Sakaguchi; Toshimitsu Ushio
  WiP session of the 9th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS2018), 2018年04月, ［国際会議］
  英語, 口頭発表（一般）
• Autoware on Board: Enabling Autonomous Vehicles with Embedded Systems　　　　　　　　　　　　　　　
  Shinpei Kato; Shota Tokunaga; Yuya Maruyama; Seiya Maeda; Manato Hirabayashi; Yuki Kitsukawa; Abraham Monrroy; Tomohito Ando; Yusuke Fujii; Takuya Azumi
  In Proceedings of the 9th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS2018), 2018年04月, ［国際会議］
  英語, 口頭発表（一般）
• Remapping Method to Minimize Makespan of Simulink Model for Embedded Multi-core Systems　　　　　　　　　　　　　　　
  Sasuga Kojima; Masato Edahiro; Takuya Azumi
  In Proceedings of the 33rd International Conference on Computers and Their Applications (CATA2018), 2018年03月, ［国際会議］
  英語, 口頭発表（一般）
• GPUhd: Augmenting YARN with GPU Resource Management　　　　　　　　　　　　　　　
  Daisuke Fukutomi; Yuki Iida; Takuya Azumi
  In Proceedings of the International Conference on High Performance Computing in Asia-Pacific Region (HPC Asia 2018), 2018年01月, ［国際会議］
  英語, 口頭発表（一般）
• HVMD_CC : Heterogeneous Value with Multiple Deadlines and Communication Contention　　　　　　　　　　　　　　　
  Mayo Nakagawa; Takuya Azumi; Yuto Kitagawa; Nobuhiko Nishio
  the 10th International Workshop on Compositional Theory and Technology for Real-Time Embedded Systems (CRTS 2017), 2017年12月, ［国際会議］
  英語, 口頭発表（一般）
• Anomaly Prediction Based on k-means Clustering for Memory-constrained Embedded Devices　　　　　　　　　　　　　　　
  Yuto Kitagawa; Tasuku Ishigooka; Takuya Azumi
  In Proceedings of 16th IEEE International Conference on Machine Learning and Applications (ICMLA 2017), 2017年12月, ［国際会議］
  英語, 口頭発表（一般）
• Exploring Scalable Data Allocation and Parallel Computing on NoC-Based Embedded Many Cores.　　　　　　　　　　　　　　　
  Yuya Maruyama; Shinpei Kato; Takuya Azumi
  2017 IEEE International Conference on Computer Design, ICCD 2017, Boston, MA, USA, November 5-8, 2017, 2017年11月
• TINET+TECS: Component-Based TCP/IP Protocol Stack for Embedded Systems.　　　　　　　　　　　　　　　
  Takuro Yamamoto; Takuma Hara; Takuya Ishikawa; Hiroshi Oyama; Hiroaki Takada; Takuya Azumi
  2017 IEEE Trustcom/BigDataSE/ICESS, 2017年08月
• Demo Abstract: Co-simulation Framework for Autonomous Driving Systems with MATLAB/Simulink.　　　　　　　　　　　　　　　
  Shota Tokunaga; Takuya Azumi
  2017 IEEE Real-Time and Embedded Technology and Applications Symposium, RTAS 2017, Pittsburg, PA, USA, April 18-21, 2017, 2017年04月
• WiP Abstract: Preliminary Evaluation of ROS2.　　　　　　　　　　　　　　　
  Yuya Maruyama; Shinpei Kato; Takuya Azumi
  7th ACM/IEEE International Conference on Cyber-Physical Systems, ICCPS 2016, Vienna, Austria, April 11-14, 2016, 2016年
• Extended mapping algorithm based on modularity from synchronous block diagrams to AUTOSAR runnables.　　　　　　　　　　　　　　　
  Shunsuke Hori; Takuya Azumi
  21st IEEE International Conference on Emerging Technologies and Factory Automation, ETFA 2016, Berlin, Germany, September 6-9, 2016, 2016年
• Exploring the performance of ROS2.　　　　　　　　　　　　　　　
  Yuya Maruyama; Shinpei Kato; Takuya Azumi
  2016 International Conference on Embedded Software, EMSOFT 2016, Pittsburgh, Pennsylvania, USA, October 1-7, 2016, 2016年
• Componentizing an Operating System Feature Using a TECS Plugin.　　　　　　　　　　　　　　　
  Tomoaki Kawada; Takuya Azumi; Hiroshi Oyama; Hiroaki Takada
  4th IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, CPSNA 2016, Nagoya, Japan, October 6-7, 2016, 2016年
• HLBS: Heterogeneous Laxity-Based Scheduling Algorithm for DAG-Based Real-Time Computing.　　　　　　　　　　　　　　　
  Yuhei Suzuki; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato
  4th IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, CPSNA 2016, Nagoya, Japan, October 6-7, 2016, 2016年
• Priority and Synchronization Support for ROS.　　　　　　　　　　　　　　　
  Yukihiro Saito; Takuya Azumi; Shinpei Kato; Nobuhiko Nishio
  4th IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, CPSNA 2016, Nagoya, Japan, October 6-7, 2016, 2016年
• Lightweight Ruby Framework for Improving Embedded Software Efficiency.　　　　　　　　　　　　　　　
  Takuro Yamamoto; Hiroshi Oyama; Takuya Azumi
  4th IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, CPSNA 2016, Nagoya, Japan, October 6-7, 2016, 2016年
• RTM-TECS: Collaboration Framework for Robot Technology Middleware and Embedded Component System.　　　　　　　　　　　　　　　
  Ryo Hasegawa; Naofumi Yawata; Noriaki Ando; Nobuhiko Nishio; Takuya Azumi
  19th IEEE International Symposium on Real-Time Distributed Computing, ISORC 2016, York, United Kingdom, May 17-20, 2016, 2016年
• mruby on TECS: Component-Based Framework for Running Script Program.　　　　　　　　　　　　　　　
  Takuya Azumi; Yuki Nagahara; Hiroshi Oyama; Nobuhiko Nishio
  IEEE 18th International Symposium on Real-Time Distributed Computing, ISORC 2015, Auckland, New Zealand, 13-17 April, 2015, 2015年
• Cloud-assisted sensing and supervision of multiple unmanned aerial vehicles by a single operator.　　　　　　　　　　　　　　　
  Takashi Shigekuni; Toshimitsu Ushio; Takuya Azumi
  Proceedings of the ACM/IEEE Sixth International Conference on Cyber-Physical Systems, ICCPS 2015, Seattle, WA, USA, April 14-16, 2015, 2015年
• Connected Smartphones and High-Performance Servers for Remote Object Detection.　　　　　　　　　　　　　　　
  Yuki Iida; Manato Hirabayashi; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato
  2014 IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, CPSNA 2014, Hong Kong, China, August 25-26, 2014, 2014年
• Reducing Data Copies between GPUs and NICs.　　　　　　　　　　　　　　　
  Anh Nguyen; Yusuke Fujii; Yuki Iida; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato
  2014 IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, CPSNA 2014, Hong Kong, China, August 25-26, 2014, 2014年
• HR-TECS: Component Technology for Embedded Systems with Memory Protection　　　　　　　　　　　　　　　
  The 16th IEEE International Symposium on Object/component/service-oriented Real-time distributed computing, 2013年06月
• TECSCE: HW/SW Codesign Framework for Data Parallelism Based on Software Component　　　　　　　　　　　　　　　
  International Embedded Systems Symposium 2013, 2013年06月
• VISA Synthesis: Variation-Aware Instruction Set Architecture Synthesis　　　　　　　　　　　　　　　
  Asia and South Pacific Design Automation Conference (ASP-DAC13), 2013年01月
• Data Transfer Matters for GPU Computing.　　　　　　　　　　　　　　　
  Yusuke Fujii; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato; Masato Edahiro
  19th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2013, Seoul, Korea, December 15-18, 2013, 2013年
• Detecting wi-fi base station behavior inappropriate for positioning method in participatory sensing logs.　　　　　　　　　　　　　　　
  Nobuhiko Nishio; Yuuki Fukuzaki; Takuya Azumi
  The 2013 ACM International Joint Conference on Pervasive and Ubiquitous Computing, UbiComp '13, Zurich, Switzerland, September 8-12, 2013 - Adjunct Publication, 2013年
• Exploring microcontrollers in GPUs.　　　　　　　　　　　　　　　
  Yusuke Fujii; Takuya Azumi; Nobuhiko Nishio; Shinpei Kato
  Asia-Pacific Workshop on Systems, APSys '13, Singapore, Singapore, July 29-30, 2013, 2013年
• Lifelog-based active movement assistant system.　　　　　　　　　　　　　　　
  Yusuke Satonaka; Takumi Kitazawa; Kazuki Suzuki; Yuki Fukuzaki; Takuya Azumi; Nobuhiko Nishio
  1st IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, CPSNA 2013, Taipei, Taiwan, August 19-20, 2013, 2013年
• PDR-based adaptation for user-progress in interactive navigation system.　　　　　　　　　　　　　　　
  Shun Yoshimi; Takuya Azumi; Nobuhiko Nishio
  The 2013 ACM International Joint Conference on Pervasive and Ubiquitous Computing, UbiComp '13, Zurich, Switzerland, September 8-12, 2013 - Adjunct Publication, 2013年
• Distributed Intent: Android Framework for Networked Devices Operation.　　　　　　　　　　　　　　　
  Yuki Nagahara; Hiroshi Oyama; Takuya Azumi; Nobuhiko Nishio
  16th IEEE International Conference on Computational Science and Engineering, CSE 2013, December 3-5, 2013, Sydney, Australia, 2013年
• Enhancement of Real-time Processing by Cooperation of RTM and TECS　　　　　　　　　　　　　　　
  Naofumi Yawata; Takuya Azumi; Nobuhiko Nishio
  Work-in-Progress Session of the 18th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, 2012年08月
• TECS Components Providing Functionalities of OSEK Specification for ITRON OS　　　　　　　　　　　　　　　
  The 9th IEEE International Conference on Embedded Software and Systems, 2012年06月
• Automatic Generation of Personal Indoor Activity Map　　　　　　　　　　　　　　　
  the 10th International Conference on Pervasive Computing, 2012年06月
• Virtual Platform Generation Using TECS Software Component and SCE　　　　　　　　　　　　　　　
  Design Automation & Test in Europe Workshop(Quo Vadis, Virtual Platforms? Challenges and Solutions for Today and Tomorrow), 2012年03月
• Extending Intent in Android for Distributed Collaboration Framework　　　　　　　　　　　　　　　
  The 17th Workshop on Synthesis And System Integration of Mixed Information Technologies, 2012年03月
• Software Component-Based HW/SW Cosimulation Framework: A Case Study　　　　　　　　　　　　　　　
  The 17th IEEE Real-Time and Embedded Technology and Applications Symposium Work-in-Progress Session of, 2011年04月
• Wheeled Inverted Pendulum with Embedded Component System: a Case Study　　　　　　　　　　　　　　　
  The 13th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing, 2010年05月
• History of Summer School on Embedded System Technologies Organized by Students and Young Engineers　　　　　　　　　　　　　　　
  Workshop on Embedded Systems Education, 2009年10月
• Optimization of Component Connections for an Embedded Component System　　　　　　　　　　　　　　　
  IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, 2009年08月
• A Memory Allocator for Efficient Task Communications by Using RPC Channels in an Embedded Component System　　　　　　　　　　　　　　　
  The 12th IASTED International Conference on Software Engineering and Applications, 2008年11月
• A Realization of RPC in Embedded Component Systems　　　　　　　　　　　　　　　
  IEEE International Conference on Embedded and Real-Time Computing Systems and Applications Work-in-Progress Session, 2008年08月
• Generic Memory Protection Mechanism for Embedded System and Its Application to Embedded Component System　　　　　　　　　　　　　　　
  IEEE 8th International Conference on Computer and Information Technology, 2008年07月
• A New Security Framework for Embedded Component Systems　　　　　　　　　　　　　　　
  The 11th IASTED International Conference on Software Engineering and Applications, 2007年11月
• A Visual Modeling Environment for Embedded Component Systems　　　　　　　　　　　　　　　
  IEEE 7th International Conference on Computer and Information Technology, 2007年10月
• A New Specification of Software Components for Embedded Systems　　　　　　　　　　　　　　　
  The 10th IEEE International Symposium on Object/component/service-oriented Real-time distributed Computing, 2007年05月
• New Environment for Event-Driven Programming　　　　　　　　　　　　　　　
  The 9th International Conference on Humans and Computers, 2006年09月
• Multimedia Manual for SORA Mail　　　　　　　　　　　　　　　
  The Ninth IASTED International Conference on Software Engineering and Applications, 2005年11月
• Networked Speaker Array Streaming Back to Client:theWorld’s Most Expensive Sound Spatializer?　　　　　　　　　　　　　　　
  The 12th International Conference on Artificial Reality and Telexistence, 2002年12月

• 情報システム学実験Ⅰ（ロボット実験）, 立命館大学
• 卒業研究２, 立命館大学
• オペレーティングシステム, 埼玉大学
• データ構造とアルゴリズム, 埼玉大学
• 情報倫理, 埼玉大学
• 情報セキュリティ, 埼玉大学
• 基礎ゼミ, 大阪大学
• 知能システム学実験C（制御）, 大阪大学
• 知能システム学実験B（ロボットアーム）, 大阪大学
• 情報システム学実験II（アセンブラ・パイプライン）, 立命館大学
• 組込みシステム, 立命館大学

• 電子情報通信学会
• 情報処理学会
• ソフトウェア科学会
• IEEE

• モビリティ革新に向けた大規模点群のリアルタイム推論のための分散機械学習理論と基盤　　　　　　　　　　　　　　　
  日本学術振興会, 科学研究費助成事業, 基盤研究(A), 2023年04月01日 - 2027年03月31日
  安積 卓也, 埼玉大学
  配分額（総額）：46280000, 配分額（直接経費）：35600000, 配分額（間接経費）：10680000
  課題番号：23H00464
  論文ID：49457152
• 低消費自律駆動型モビリティ向けソフトウェアプラットフォームの構築　　　　　　　　　　　　　　　
  科学技術振興機構, 戦略的な研究開発の推進 戦略的創造研究推進事業 さきがけ, 2021年 - 2024年
  安積 卓也, 埼玉大学, 研究代表者
  本研究では、組込みシステム向けの高性能・低消費のハードウェアを搭載したエッジデバイスを用いて高度化・複雑化に対応する自律駆動型モビリティ向けのソフトウェアプラットフォームを構築します。具体的には、バッテリ駆動する案内・搬送ロボットやパーソナルモビリティ等小型の自律駆動型モビリティに不可欠となる認知・判断・動作に係る処理を低消費電力かつリアルタイム処理で実現します。
• 大容量データをリアルタイム処理するメニーコア向けソフトウェアプラットフォームの構築　　　　　　　　　　　　　　　
  JST, さきがけ（新しい社会システムデザインに向けた情報基盤技術の創出）, 2017年10月 - 2021年03月
  安積 卓也, 研究代表者
  競争的資金
• マルチ・メニーコア向け高並列組み込み制御システム設計・実装手法の研究　　　　　　　　　　　　　　　
  日本学術振興会, 科学研究費助成事業, 基盤研究(B), 2016年04月01日 - 2019年03月31日
  枝廣 正人; 道木 慎二; 安積 卓也; 中本 幸一; 本田 晋也, 名古屋大学
  配分額（総額）：17160000, 配分額（直接経費）：13200000, 配分額（間接経費）：3960000
  一つの半導体チップに複数のプロセッサを搭載するマルチ・メニーコアプロセッサ（以下，マルチコア）向け高並列組込み制御システム設計・実装手法の確立を目的とし，以下の４項目を実現した．(ア)高並列化を実現する制御アルゴリズムを確立，(イ)マルチコア活用を目指す制御設計とソフトウェア実装を統合するクロスレイヤ設計手法を確立し，車載制御を意識した実際の環境において実証，(ウ)モデルベース並列化における最大実行時間最小化手法を提案，(エ)PCを用いた組込み制御高速シミュレーション手法を提案．
  課題番号：16H02800
• 組込みシステム開発基盤のための革新的動作検証プロトタイプシステムの研究　　　　　　　　　　　　　　　
  日本学術振興会, 科学研究費,：若手研究（A）, 2015年04月 - 2019年03月
  安積 卓也, 研究代表者
  競争的資金
• 安全性・生産コストを考慮したロボット向けソフトウェアプラットフォームの構築　　　　　　　　　　　　　　　
  日本学術振興会, 科学研究費：若手（B）, 2012年04月 - 2015年03月
  安積 卓也, 研究代表者
  競争的資金
• 組込みシステム向けコンポーネントシステムを用いたシステムレベル設計　　　　　　　　　　　　　　　
  日本学術振興会, 科学研究費,：研究活動スタート支援, 研究活動スタート支援, 2011年 - 2013年03月
  安積 卓也, 立命館大学, 研究代表者
  携帯電話、家電、自動車などの組込みシステムは、大規模化、複雑化しており、組込みソフトウェアと同様にハードウェア設計の生産性も重要な問題である。本研究課題では、ソフトウェアの生産性を向上する組込みシステム向けのコンポーネントシステム(TECS)を活用し、ソフトウェア資源をハードウェア設計に直接利用することでハードウェア設計を効率化するフレームワークの研究・開発を行った。
  競争的資金, 課題番号：22800071
• 組込みシステム向けコンポーネントシステムの車載向けリアルタイムOSへの適用　　　　　　　　　　　　　　　
  研究成果最適展開支援プログラム A-STEP：探索タイプ, 2011年 - 2011年
  競争的資金
• 組込みシステムに適したコンポーネント技術についての研究　　　　　　　　　　　　　　　
  日本学術振興会, 科学研究費助成事業, 特別研究員奨励費, 2008年 - 2009年
  安積 卓也, 名古屋大学
  配分額（総額）：1200000, 配分額（直接経費）：1200000
  組込みシステムの複雑化・大規模化に伴いソフトウェア生産性が問題となっている。この問題を解決するために、組込みシステムに適したコンポーネントシステムを提案してきた。本コンポーネントシステムは、コンポーネントを静的にインスタンス化・結合する静的なコンポーネントモデルを採用する。
  昨年度に引き続き、申請書に書いた通りアクセス制御機構、最適化機構、RPC機構の完成度を高め、その成果を論文にまとめた。
  今年度新たに、コンポーネントベース開発におけるソフトウェアの生産性の向上を目的とし、コンポーネントの結合部分に影響を与えずにコンポーネントを挿入する機構を提案した。コンポーネントの結合部分に新たなコンポーネントを挿入することで、さまざまな応用が可能である。たとえば、トレースログ機構、アクセス制御機構、RPCチャネル、性能評価、排他制御機構など広範囲に適用できる。提案機構はインタフェースジェネレータのプラグインとして実装した。そのプラグインを選択することで、挿入されるコンポーネントはコンポーネントの結合に適したコンポーネント記述、およびC言語のソースコードが自動生成され利用される。
  さらに、コンポーネントシステムの適用事例として「リアルタイムOSのカーネルオブジェクトのコンポーネント化」、「倒立振子」、「UMLで設計された組込みソフトウェアの開発事例」を行った。本年度の成果として論文誌1件、国際会議3件、国内学会で3件の発表を行った。
  課題番号：08J06327
• RTOS上でのセキュリティフレームワークの構築　　　　　　　　　　　　　　　
  2006年度下期未踏ソフトウェア創造事業, 2006年 - 2007年
  競争的資金