Online Event-Triggered Control of Partially Unknown Networked Systems via Adaptive Dynamic Programming

For a class of partially unknown multi-player nonzero-sum games in nonlinear continuous-time networked systems, this article proposes a control method which is adaptive and includes an event trigger mechanism. Because of the system dynamic is unknown, a specific identifier based on fuzzy logic model (FLM) is built. To acquire an approximate Nash equilibrium solution, it is necessary to use critic neural network (NN) to estimate the value function and control strategy. In this work, a new state-dependent event triggering mechanism is provided, which is different from the conventional ones. This new method can ensure the stability of the controlled systems and the calculation and communication burdensome can be reduced significantly. In addition, the system state and the estimation errors of the critic NN weight are proved to be uniform ultimate bounded (UUB). In the end, two numerical simulation examples are used to prove the correctness of the theoretical analysis.