UAV Path Planning Based on DDQN for Mountain Rescue

The high calculation cost and low success rate of strategy generation caused by unknown environment and emergencies, highly limited the usage of traditional UAV path planning methods. A path planning method based on DDQN algorithm is proposed to solve the mountain rescue problem in a complex environment. First, the state space and action space of the UAV are determined, then the reward function is designed based on the positional relationship between target, obstacle and UAV, and finally the DDQN reinforcement learning model based on the greedy strategy is established. Simulation experiments in the two scenarios confirm that the DDQN algorithm can not only achieve path planning for rescuing static goals, but also realize fast-tracking of moving targets through quadratic programming At the same time, compared with the results of the DQN algorithm, it shows that the method has faster network convergence and better path planning ability.