Four-rotor unmanned aerial vehicle route following control method based on deep reinforcement learning

Abstract

The invention provides a four-rotor unmanned aerial vehicle route following control method based on deep reinforcement learning. The method comprises the following steps of firstly, establishing a Markov model of a four-rotor unmanned aerial vehicle route following deep reinforcement learning algorithm, and then performing deep reinforcement learning by adopting a deep deterministic policy gradient (DDPG) algorithm. The problems of relatively low control precision, continuous control incapability, unstable learning process and the like in a conventional method based on reinforcement learning are solved, and high-precision four-rotor unmanned aerial vehicle route following control is achieved. According to the method, the reinforcement learning is combined with a deep neural network, so that the learning ability and the generalization ability of the model are improved, and the complexity and carelessness of manually operating flight of an unmanned aerial vehicle in the uncertainty environment are avoided, so that completion of a route following task by the unmanned aerial vehicle is safer and more efficient; and meanwhile, the method has good application prospects in scenes of unmanned aerial vehicle target tracking, autonomous obstacle avoidance and the like.

Description

technical field [0001] The invention belongs to the field of intelligent control, and in particular relates to a method for controlling the flight path of an unmanned aerial vehicle. Background technique [0002] In recent years, as quadrotor UAVs shine in many fields such as industrial inspection, emergency rescue and disaster relief, and life assistance, it has gradually become a new frontier and hot spot in military aviation academic research. For UAVs to complete high-altitude route following, target tracking and other mission scenarios where humans cannot reach the on-site operation, ensuring the autonomy and controllability of UAV flight is the most basic and important functional requirement. premise of the task. Due to many reasons, autonomous decision-making and control of UAVs still face enormous challenges in the field of intelligent control. First, the UAV flight control has a large amount of input and output, and its kinematics and dynamics models are complex, ...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the deficiencies in the prior art, the present invention proposes a four-rotor UAV route following control method based on deep reinforcement learning. The method first establishes the Markov model of the quadrotor UAV route following deep reinforcement learning algorithm , and then use the Deep Deterministic Policy Gradient (DDPG) algorithm for deep reinforcement learning to overcome the problems of low control accuracy, inability to achieve continuous control and unstable learning process in previous methods based on reinforcement learning, and realize high-precision quadrotor drones course following control

Images