Observer-based event triggered ADP approach for input-constrained nonlinear systems with disturbances

This paper proposes an ADP-based event-triggered control (ETC) approach for input-constrained nonlinear systems with external disturbances to effectively save computational cost. An observer is established to estimate unknown disturbances, and a novel performance index function is introduced to suppress the negative effect of disturbances and tackle the control constraints. Under the event-triggered mechanism, the corresponding nonquadratic Hamilton-Jacobi-Bellman equation is further formulated. With the aid of the modified performance index, an event-triggered condition is designed to reduce the controller updating times with guaranteed system stability. Furthermore, a single critic neural-network (NN) is constructed to approximate the optimal control policy by using the ADP technique. In order to relax the requirement of initial admissible control, a modification term is added to the critic weight updating law. According to the Lyapunov stability theorem, the closed-loop nonlinear system and the weight estimation error of the critic NN are both guaranteed to be uniformly ultimately bounded. Finally, simulation results and comparison studies justify the effectiveness of the proposed ADP-based ETC method.