Robust Cooperative Control for Nonlinear Multi-agent Systems with Input-Disturbances via Adaptive Dynamic Programming
Qiuxia Qu, Juan Wang, Qiong Xia, Liangliang Sun, Yang Cui · 2021
Considering the leader-following consensus problem for the nonlinear multi-agent systems with bounded input-disturbances under fixed topology, a novel distributed robust protocol is designed to guarantee all followers synchronize to the leader by investigating the gain of the Nash Equilibrium. The robustness restrictions are given through Lyapunov theory. To get the Nash solution, critic neural networks are trained based on adaptive dynamic programming algorithm in an online and forward-in-time manner to solve the coupled Hamilton-Jacobi equations. An additional term is added to the neural network weight tuning law to avoid the requirement for the initial admissible control law.