Adaptive compensation control for nonlinear stochastic multi‐agent systems: An event‐triggered mechanism

This paper proposes an adaptive compensation control algorithm for solving the actuator failures problem of nonlinear stochastic multi‐agent systems (MASs) under the directed communication topology. It should be emphasized that the coexistence of unknown nonlinearities, stochastic perturbations and actuator failures makes the implementation of control protocol very difficult and extremely challenging. To achieve the control objective, fuzzy logic systems (FLSs) are first employed to deal with the unknown nonlinearities of each agent. Then, the threshold‐based event‐triggered mechanism is further considered to reduce the communication burden of the system in the case of limited communication resources. Moreover, the issue of “explosion of complexity” is solved by using dynamic surface control (DSC) technique in the process of backstepping design. With these efforts, the actuator failures are circumvented and the outputs of the followers converge to the convex hull spanned by the multiple leaders' outputs. Finally, the simulation results of multiple single‐link robots show the validity of the proposed design scheme. First, the studied stochastic nonlinear MASs consisting of N leaders is considered, in which the dynamics of the leaders can be completely unknowable. Therefore, the leader models can be more general compared to the previous control results [42–46]. Moreover, we cleverly apply a novel coordinate transformation to deal with the consensus problem of multi‐leader. Second, in this paper, the adaptive compensation controller is constructed to effectively compensate the uncertainties of the actuator failure models. In addition, event‐triggered control mechanism for the input of the actuator failure model is used to address the problem of resources transmission limitation. Third, the proposed communication protocol can ensure that the outputs of the followers converge to the convex hull spanned by the outputs of the multiple leaders. Furthermore, the problem of “explosion of complexity” in the derivation process is solved by fusing the backstepping and DSC technology.