Further, to improve the flight efficiency of the UAV swarm, a reinforcement learning model is provided to generate the optimal offset vector for each UAV according to its current subgoal. Moreover, to ...