Dynamic Event-Triggered Robust Stabilization of Continuous-Time Nonaffine Nonlinear Systems Based on ADP

This paper investigates the robust optimal stabilization problem of continuous-time nonaffine nonlinear systems and designs a dynamic event-triggered adaptive robust controller using adaptive dynamic programming (ADP) and dynamic event-triggered mechanism (ETM). Initially, the original problem is transformed into the stabilization problem of affine systems with the introduction of an affine-form pre-compensator. Then, the unknown disturbances are estimated by designing a disturbance observer. And the estimated results are employed to develop a novel adaptive cost function to suppress the impacts of disturbances. Based on the established cost function, the event-triggered robust control policy is derived by training a single critic neural network (NN), where the initial stabilizing control policy is not required. In comparison with traditional time-triggered and static event-triggered ADP schemes, the involvement of dynamic ETM with the help of a dynamic variable reduces more transmission and computational loads. According to Lyapunov stability theory, the uniform ultimate boundedness (UUB) of the closed-loop system is demonstrated under the proposed control strategy. Simulation results further verify the effectiveness of the design for the stabilization problem of nonaffine nonlinear systems with unknown disturbances.