Reinforcement Learning based Anti-UAV Three-dimensional Pursuit-evasion Game for Substation Security

The illegal intrusion of unmanned aerial vehicle (UAV) into substations will have a serious impact on the safe operation of the power grid. This article will adopt the method of equipping UAV in the substation system to capture illegal UAV. Specifically, the pursuit-evasion game of UAV will be transformed into solving the Hamilton-Jacobi-Bellman equation using the principle of differential games, and then the integral reinforcement learning algorithm will be used to solve the pursuit-evasion game problem. Finally, the simulation results demonstrate the effectiveness of the proposed algorithm in the UAV pursuit-evasion game problem.