I. Introduction
As a powerful robust control method for attenuating the disturbance effect [1], [2], control has received increasing attention in the past years [3]–[5]. From the perspective of minimax optimization problem, the control problem can be formulated as a two-player zero-sum (ZS) game [6], [7]. In order to obtain a controller that minimizes the performance index in the worst-case disturbance, ones need to find the Nash equilibrium solution by solving the Hamilton–Jacobi–Isaacs (HJI) equation. Nevertheless, it is intractable to give an analytic solution to the HJI equation in case of nonlinear systems.