Event-Triggered Fault-Tolerant Tracking Control for Autonomous Ground Vehicle via Descriptor Reduced-Order Observer

Autonomous ground vehicles (AGVs) play an important role in transportation, environmental exploration, and so on. However, they may suffer from inevitable sensor and actuator faults which degenerate the setpoint tracking performance. Within this context, this paper focuses on the issues of deploying event-triggered fault-tolerant controller based on reduced-order descriptor observer. First, allowing for the AGV kinematics and the setpoint tracking objective, we design a reduced-order descriptor sliding mode observer to synchronously reconstruct the tracking error states as well as physical faults. Then, we develop an event-triggered communication scheme in the control center side, leading to the tolerant controller by virtue of reconstructed tracking error dynamics with an improved transmission efficiency. In addition, we perform the strict theoretical guarantee of both estimation stability and tracking stability, based on which, the estimator and controller gains in conjunction with event trigger parameters are codesigned. Finally, the correctness and effectiveness of theoretical derivation are examined by the experimental study of setpoint tracking of AGV.