Intelligent Path Planning Algorithm for UAV Group Based on Machine Learning

Abstract

Aiming at the minimum and local minimum problems existing in the artificial potential field algorithm for obstacle avoidance of UAV, a new path planning method is proposed by using the improved artificial potential field algorithm. Different from the current artificial potential field method, this model starts from the interaction between two aircraft. On the basis of obstacle repulsion, the repulsion force between UAV is increased, and the pre-centroid of the cluster is defined as another source of gravity. The analysis of the algorithm shows that this method can effectively prevent the UAV from falling into the local minimum and enhance the control and obstacle avoidance ability of the UAV cluster. Based on the UAV control model, the path planning design is given and the simulation experiment is carried out. The experimental results show that the UAV cluster control based on this model has better obstacle avoidance performance and target tracking ability.