Robust control design for zero-sum differential games problem based on off-policy reinforcement learning technique

This paper aims to figure out the robust zero-sum differential game problem using an off-policy reinforcement learning technique. The robust system model is first established based on the nominal one. The control strategy is proposed with the asymptotic stability and optimality being strictly proved. The off-policy reinforcement learning technique is built from the Bellman equation to generate the control policy. A potentially inaccurate system dynamic model’s influence is avoided because the outcome is attained from the system data set obtained. It is the first-time application of the off-policy RL algorithm on this robust two-player zero-sum differential game problem. Additionally, the final algorithm’s convergence is demonstrated, and a simulation example is run to confirm its efficacy.