Transactions of the Institute of Systems, Control and Information Engineers
Online ISSN : 2185-811X
Print ISSN : 1342-5668
ISSN-L : 1342-5668
Volume 37, Issue 3
Displaying 1-3 of 3 articles from this issue
Special Issue Paper
  • Syota Yoshida, Shin Wakitani
    Article type: Paper
    2024 Volume 37 Issue 3 Pages 65-72
    Published: March 15, 2024
    Released on J-STAGE: June 15, 2024
    JOURNAL FREE ACCESS

    A model error compensator (MEC) was proposed as a method to improve the control loop performance by suppressing modeling errors. MEC suppresses modeling errors by providing feedback on the error between control output and model output. In MEC, the performance of the modeling error suppression is strongly affected by the structure of the compensator and the tuning of control gains. Model-based design methods based on robust control and parameter tuning methods based on data-driven approaches have been proposed as design methods for MEC. However, an appropriate compensator structure and parameters are required to suppress modeling errors effectively. This study proposes a design method for structure-free MEC to directly determine compensation inputs using a database-driven approach. Thanks to the proposed method, the optimal output from the compensator can be obtained using only data. This paper verifies the effectiveness through numerical examples and implementation to the hydraulic motor control system.

    Download PDF (1792K)
  • Reo Okatani, Tomoyuki Iori, Yasumasa Fujisaki
    Article type: Paper
    2024 Volume 37 Issue 3 Pages 73-79
    Published: March 15, 2024
    Released on J-STAGE: June 15, 2024
    JOURNAL FREE ACCESS

    Contraction analysis is a method for stability analysis of nonlinear systems, which focuses on the distance between trajectories rather than that from an equilibrium or a target trajectory. In earlier studies, relationships among the three characteristics, i.e., contraction, incrementally exponential stability (IES), and exponential convergence, have been investigated. This paper shows the equivalency among semi-contraction, incremental stability, and convergence, all of which are weaker versions of contraction, IES, and exponential convergence, respectively, without exponential decay. Finally, through a numerical example, we show that the equivalency holds for a nonlinear system having a limit cycle, to which the properties with exponential decay cannot be applied.

    Download PDF (1219K)
Regular Issue Paper
  • —Action Recognition using LSTM
    Takumi Ozaki, Atsushi Harada
    Article type: Paper
    2024 Volume 37 Issue 3 Pages 80-90
    Published: March 15, 2024
    Released on J-STAGE: June 15, 2024
    JOURNAL FREE ACCESS

    The purpose of this study is to develop a system to watch over the increasing number of elderly people living alone. We aimed to construct a system with high privacy protection and a high recognition rate by using LSTM based on skeletal coordinates obtained from images. In this study, LSTM models were trained with various optimizers and learning rates, and the model with the best results was verified. Furthermore, we constructed a system that can recognize actions in real time using LSTM models. As a result, we recorded a maximum accuracy of 0.999886 when evaluating the LSTM model on the split training dataset, and a maximum average recall of 0.834148 on the validation dataset that included only the abnormal actions that we captured. In addition, a client-server type system was constructed, and only skeletal information is used for mutual communication to ensure a high degree of privacy protection.

    Download PDF (1626K)
feedback
Top