A UAV Autonomous Maneuver Decision-Making Algorithm for Route Guidance

In order to improve the autonomy of UAV for route guidance, an UAV autonomous maneuver decision- making algorithm is proposed. The optimal UAV maneuver policy is developed in an interactive environment. Firstly, the UAV maneuver decision-making model based on MDPs was established, where the flight state space, the flight action space and the flight assessment function are designed. Then, we proposed the algorithm based on Double Deep Q-Learning with Prioritized Experience Replay. Finally, according to the simulation results, the efficiency and autonomy of algorithm we proposed is proved.