Adaptive Output-Feedback Fault-Tolerant Control for Distributed Optimization of Nonlinear MASs Under Event-Triggered Communication
This article investigates the distributed adaptive optimization of multiagent systems with uncertain nonlinearities, utilizing a dynamic event-triggered mechanism. Distinct from existing algorithms that assume healthy operation, the proposed dynamic event-triggered strategy not only saves communication resources but also ensures convergence to a small neighborhood of the global minimizer even in the presence of actuator faults, thereby effectively avoiding the loss of optimality or optimization failure caused by malfunctions. Furthermore, considering that practical systems often possess only measurable outputs and exhibit complex nonlinear characteristics, a state observer is developed to estimate the unmeasured state variables. On this basis, by incorporating the backstepping design technique from the nonlinear fault-tolerant control theory, an adaptive event-triggered output feedback fault-tolerant control strategy is designed to compensate for actuator faults. Meanwhile, a high-order filter is introduced to resolve the incompatibility between the signal derivative discontinuity induced by event-triggering and the differentiability requirements of backstepping design. Thereby, the controller design and stability analysis can proceed within a continuous-time framework. Theoretical analysis demonstrates that the proposed method guarantees that all signals in the resulting closed-loop system are bounded. The validity of the proposed control strategy is further verified through a simulation.