The potential field method for unmanned vehicle path planning based on swarm intelligence optimization

Although the traditional artificial potential field method is efficient and fast, it still suffers from issues such as unreachable goals, path oscillation, and suboptimal paths. To address these limitations, this paper introduces the concept of swarm intelligence optimization and proposes an improved scheme based on the Hippopotamus Optimization Algorithm (HOA). The approach dynamically generates different repulsion factors according to the size of obstacles. By simulating the exploratory, defensive, and evasive behaviors of hippos, the algorithm adaptively adjusts the resolution and search speed of the search space to optimize the configuration of repulsion factors. Experimental results demonstrate that the proposed improved artificial potential field method significantly outperforms other algorithms in terms of path smoothness and safety, both in simulation tests using Python and in real-world tests conducted on the ROS platform.