Multi-UUV Path Planning Study with Improved Ant Colony Algorithm and DDQN Algorithm

Abstract

For the Unmanned Undersea Vehicle (UUV) path planning problem, an improved ant colony algorithm is proposed to solve the global static path planning and an improved DDQN algorithm is proposed to solve the local dynamic collision avoidance problem. The traditional ant colony algorithm is optimized by adding the constraints of collision avoidance among UUVs to improve the accuracy of the algorithm. The Q algorithm idea is introduced to improve the DDQN algorithm to realize the UUV’s anticipation of future environmental state changes. This algorithm can not only solve the local collision avoidance problem, but also realize the accurate prediction of future environment. Finally, the algorithm is validated with a real UUV. The experimental results show that the proposed method solves the problems of collision avoidance and path finding for multiple UUVs in complex environments, and can effectively avoid collisions for obstacles.