Decentralized Fault Estimation and Distributed Fault-tolerant Tracking Control Co-design for Sensor Faulty Multi-agent Systems with Bidirectional Couplings

This study proposes a co-design framework of decentralized fault estimation and distributed fault-tolerant tracking control schemes of Lipschitz nonlinear multi-agent systems with external disturbances and unpredicted sensor faults. To begin with, the sensor fault is actively hidden in the extended state through augmented transformation, and the decentralized unknown input observer based on extended dynamics is applied in synchronously estimating system state and sensor fault. Then, the updated link-based fault-tolerant tracking control protocol is proposed by virtue of the estimated information from estimation dynamics and the relative output signal from neighboring agents in a distributed fashion. The proposed co-designed algorithm guarantees the state consensus tracking property and overcomes the bi-directional couplings between the estimation and tolerance systems. Simulation example of multi-machine power systems verifies the effectiveness of the proposed co-designed algorithm