Dynamic Event-Triggered Optimal Consensus Fault-Tolerant Control for Unknown Stochastic Nonlinear Multi-Agent Systems With Time-Delays

The optimal consensus fault-tolerant control (FTC) problem for unknown time-delays nonlinear stochastic multi-agent systems (MASs) with hybrid actuator faults is investigated via adaptive dynamic programming (ADP) under the dynamic event-triggered control (ETC) mechanism in this paper. In control design, FLSs with delayed states are utilized to design an adaptive state identifier for estimating the unknown delayed dynamics in stochastic MASs. The fault estimator is constructed to detect actuator bias fault in real-time. By utilizing the ADP algorithm with identifier-actor-critic structure, a simplified fuzzy adaptive optimal consensus FTC strategy is developed, which does not require persistent excitation (PE) assumption. Furthermore, a dynamic ETC strategy is designed to further optimize the data transmission efficiency, effectively enhancing the system flexibility and resource utilization. Theoretical analysis indicates that the proposed control scheme can ensure that all signals of the nonlinear stochastic MASs are semi-globally uniformly ultimately bounded (SGUUB) in mean square, while achieving consensus between the follower’s states and the state of the leader. Finally, a numerical simulation validates the efficacy and feasibility of the proposed control strategy. Note to Practitioners—With the intensification of the global energy crisis, energy consumption optimization has become a crucial performance indicator of modern control design. In this paper, the optimal consensus FTC strategy is developed for unknown nonlinear stochastic MASs with time delays and actuator failures, which can achieve control requirements while minimizing control costs. The proposed control strategy can be applied to practical scenarios such as aerospace, formation control, and power systems. Compared with the existing control algorithms, the proposed control strategy avoids complex dynamic modeling and strong assumptions of nonlinear functions, and does not require PE assumption. In addition, a novel dynamic ETC strategy has been developed, which effectively improves resource utilization and data transmission efficiency compared with traditional continuous-time control and general ETC strategies. Consequently, the control strategy designed in this paper exhibits broader applicability and stronger practical value.