In event-triggered control, which effectively reduces communication frequency in networked control systems, detecting network disconnections is difficult. To solve this problem, a previous study proposed event-triggered control systems using an integrator with an auxiliary signal for detecting communication disconnections. However, this method has not yet been applied to servo systems, and it sometimes takes a long time to detect anomalies. Therefore, we extend the previously proposed method to servo systems in this study. Additionally, we apply an anomaly detection approach based on Generative Adversarial Networks (GANs), which have recently demonstrated notable effectiveness in anomaly detection tasks. Numerical experiments confirm that the proposed GAN-based anomaly detection method is effective for servo systems, achieving faster detection of anomalies than the previous approach.
Yamamoto et al. (Sat,) studied this question.