Anomaly Detection for Networked Motion Control Systems Using Finite-Time EID Approach

This work deals with the anomaly detection problem of networked motion control systems (NMCSs), and a finite-time equivalent-input-disturbance (FTEID) approach is proposed. First, both functional safety threats and cyber security threats are considered of the NMCS, and their overall effects are modeled to be a combined unknown signal. Then, an EID model is established of the system to implement the anomaly detection. A terminal-sliding-mode observer is introduced to construct the FTEID, thereby achieving a better anomaly-detection performance. In addition, the analysis of the system is straightforward. Finally, simulation experiment is carried out to verify the effectiveness of this study.