Based on Deep Reinforcement Learning(DRL),the path planning of UAV formation is studied.Aiming at the shortcomings of slow convergence speed and sparse rewards of reinforcement learning algorithm models in the formation control problem,artificial potential field method is introduced into the deep reinforcement learning,and the UAV formation path planning network training framework is established.Meanwhile,according to the formation control goal,the formation switching reward function is designed for training.Based on AirSim and UE4 simulator,a UAV reinforcement learning formation path planning simulation training environment is built to realize the UAV formation path planning control in the threatened environment.Through comparative experiments,it is verified that the proposed algorithm has superior performance and faster convergence speed in terms of formation stability and collision rate compared with the baseline algorithm.