Research on Intelligent Control System of UAV Path Planning and Autonomous Control Based on Deep Reinforcement Learning

This paper studies the autonomy and intelligence of UAV. The system can effectively control the lifting of the UAV through the computer, and carry out linear motion between multiple operation points, so that its position deviation in the horizontal position is less than 0.14 meters. In this paper, a deep reinforcement learning PID method based on the COACH method of correction and correction is proposed, which provides the optimal path for UAV landing on the mobile platform. First, the reinforcement learning model is trained using the correction and correction framework in the simulation environment, and then the trained model is used to output the control parameters in the simulation environment and the real environment. Experiments show that the proposed method can effectively improve the time and smoothness of UAV trajectory planning, and improve the real-time performance of the algorithm, so as to meet the needs of obstacle avoidance and planning in complex scenes. The results show that this new type of aircraft can land smoothly on a maneuvering platform.