Model free adaptive iterative learning fault-tolerant control for nonlinear multi-agent systems

This study develops an iterative learning control (ILC)-based trajectory tracking method for multi-agent systems, considering the impact of data packet loss and actuator faults. In this method, the system dynamics are unknown and nonlinear, and communication between agents is affected by packet loss during network transmission. To tackle this challenge, an ILC strategy is employed, combined with a fault compensation mechanism, enhancing the system’s robustness against actuator faults and data packet loss. By introducing feedback-based iterative updates, all agents are able to progressively approach the desired trajectory through multiple iterations. The study shows that the method can effectively reduce trajectory tracking errors and exhibits good robustness in the presence of actuator faults and data loss. Finally, simulation results are compared with traditional PID control methods, demonstrating the efficacy of the method in multi-agent trajectory tracking tasks.