Event-triggered Global Adaptive Dynamic Programming for Multi-agent Consistency

Abstract

An algorithm based on event-triggered global adaptive dynamic programming is proposed for optimal control of multi-agent system consistency. The algorithm converts the multi-agent problem of consistency control to solving the Hamilton-Jacobi-Bellman equation of the optimal solution, and a method of the sum of squares iteration is used to calculate the optimal control strategy. The process of approximating the optimal control strategy and cost function by neural network training through a large number of basic functions is eliminated, to reduce the computation complexity of the system. By introducing event trigger conditions, the update times of the controller and actuator in the multi-agent system are reduced, and the frequency of information transmission between adjacent agents is also reduced. Using the optimal control theory and Lyapunov stability theory, the convergence of the system in a period of time after the event is triggered is analyzed. Finally, the effectiveness of the theoretical results is verified by MATLAB simulation.