Actor-Critic Learning Hierarchical Sliding Mode Control for a Class of Underactuated Systems

Aiming at the stability control problem of a class of second-order uncertain under-actuated systems, an adaptive hierarchical sliding mode control method based on Actor-Critic learning is proposed. First, a new type of PID sliding surface is designed to construct a sliding mode controller. Then, the Actor-Critic learning method is used to optimize the control parameters to enhance the adaptability of the controller. Among them, the actor’s strategy function and Critic’s value function are approximated by the radial basis function network. Finally, a simulation experiment is performed on the automobile inverted pendulum system. Comparative experimental simulation results show that the control method can suppress interference more effectively and has stronger robustness.