The UAV formation control has such problems as insufficient intelligence level and lack of self-learning ability.Aiming at the problems,a UAV formation controller based on DDQN algorithm in deep reinforcement learning is designed.The controller can control the speed and heading channels simultaneously, so that the followers can track the leader and maintain the formation through self-learning,and the intelligence level of the UAV is improved.To verify the effectiveness of the designed controller,the traditional PID controller is used for comparison.The simulation results show that the DDQN-based controller can effectively create UAV formation and meet the requirements of UAV formation.The study is an effective exploration for intelligent control of UAV formation.