Observer-Based Adaptive Dynamic Programming Control For Stackelberg Differential Games Of Input Constrained Two-Player Nonlinear Systems*

This article proposes a method of control for the Stackelberg differential game problem in nonlinear systems with unknown dynamic input constraints, leveraging adaptive dynamic programming (ADP). Initially, a neural network (NN) observer is created to ascertain the system dynamics, easing the requirements on the system dynamics using input and output data. Subsequently, the neural network is employed to approximate the value function for obtaining the equilibrium solution of the Stackelberg differential game. The stability of the closed-loop system is then proven through Lyapunov stability theory. Finally, the effectiveness of the ADP method is confirmed through simulation studies.