Dynamic event-triggering adaptive dynamic programming for robust stabilization of partially unknown nonlinear systems

Abstract

In this paper, a new dynamic event-triggering (DET) mechanism based on adaptive dynamic programming (ADP) is developed to deal with the robust control problem of partially unknown uncertain systems. First, this paper completes the transition from the robust control problem to the optimal control problem by designing a nominal system. Meanwhile, the use of integral reinforcement learning (IRL) eliminates the need for prior knowledge of drift dynamics. Then, to improve resource utilization, a static event-triggering (SET) scheme is designed. Subsequently, a DET scheme is developed on the basis of SET to further improve resource utilization. It is proven that the developed DET controller guarantees the robustness of the partially unknown uncertain system. The neural network (NN) weight estimation errors are uniformly ultimately bounded (UUB) while the Zeno behavior is successfully avoided. Finally, an experiment is provided to demonstrate that the proposed DET algorithm has the fewest triggering samples while guaranteeing robustness.