Adaptive state observer event-triggered consensus control for multi-agent systems with actuator failures

An adaptive neural network event-triggered consensus control method incorporating a state observer was proposed for a class of uncertain nonlinear multi-agent systems (MASs) with actuator failures. To begin, a state observer was constructed in an adaptive backstepping framework to estimate the MASs' unmeasurable states, and a radial basis function neural network (RBFNN) was employed to approximate the unknown nonlinear function of MASs. Meanwhile, to reduce the impact of actuator failure on the performance of MASs, the adaptive event-triggered mechanism (ETM) was designed to dynamically compensate for actuator failures, which alleviated the communication burden among individual agents by decreasing the update frequency of the control signals. Furthermore, all followers can track the leader's output signal with the synchronization errors converging to zero. Finally, simulation examples were used to verify the effectiveness of the proposed control strategy.