Data-Based Event-Triggered Control of Zero-Sum Games with Completely Unknown Dynamics

In our design, we develop a model-free optimal control method of zero-sum games (ZSG) with unknown system dynamics under the event-triggered mechanism. Firstly, based on the adaptive dynamic programming (ADP), the optimal policies are obtained by solving the Hamilton-Jacobi-Issacs (HJI) equation. Secondly, a data-based optimal control approach is designed via integral reinforcement learning (IRL) algorithm. Moreover, to reduce the communication burden, an event-triggered IRL-based control method is proposed for ZSG of completely unknown system. The stability analysis is given via Lyapunov principle. Finally, a simulation example is illustrated to show the effectiveness of the designed algorithm.