Research on Obstacle Avoidance Algorithm of Multi-UAV Consistent Formation Based on Improved Dynamic Window Approach

Abstract

The artificial potential field method is widely used in UAV formation and obstacle avoidance due to its concise algorithm and easy implementation, but it is easy to fall into a local optimal solution in a large-scale environment, and the algorithm does not consider the kinematics of the controlled object; this paper proposes An improved dynamic window approach, considering the UAV kinematics model, realizes the effective obstacle avoidance of UAV formations under the consistency algorithm. Firstly, the target distance correction azimuth angle evaluation function is introduced, and the A star algorithm is integrated to replace the fixed weight of the azimuth angle evaluation function, which improves the search ability of the UAV to navigate to the target point in the unknown environment. Secondly, a new rotation cost is added to the evaluation function, and a penalty is imposed on the large rotation angle to ensure the smoothness of the trajectory. Then, the improved obstacle avoidance strategy is applied to the leading-following consistent formation algorithm, which can effectively achieve formation keeping, obstacle avoidance and collision avoidance between aircraft in unknown environments. Finally, the simulation verification based on Matlab shows that the proposed improved dynamic window strategy can significantly improve the UAV path planning and obstacle avoidance ability in the location environment, and it can be applied to the consistent formation algorithm while maintaining the formation.