Optimal Fault-Tolerant Formation Control of Multi-UAVs Based on Event-Triggered Control
Tong Mei, Ruian Wang, Wenlai Ma, Wei Hao · IFAC-PapersOnLine · 2025
This paper investigates the problem of fault-tolerant formation control (FTFC) for multi-UAV systems subject to actuator faults and external disturbances. To address lumped system uncertainties, an adaptive neural network-based backstepping control strategy is proposed. A neural network observer is developed to estimate unmeasurable system states in real time, enabling timely compensation and enhancing the overall fault-tolerant capability. An event-triggered mechanism is incorporated to reduce unnecessary control inputs, thereby saving computational resources while maintaining formation consistency. Furthermore, a reinforcement learning (RL) framework with an actor-critic architecture is employed to enhance control performance by adaptively tuning the controller parameters without requiring the persistence of excitation condition. Lyapunov-based analysis guarantees the stability of the closed-loop system and the convergence of tracking errors. Finally, numerical simulation results verify the effectiveness of the proposed control approach.