Improved Artificial Potential Field Method for UAV Path Planning

Path planning for uncrewed aerial vehicles (UAVs) is a core technology for autonomous navigation. The artificial potential field (APF) method, known for its computational efficiency and strong real-time performance, has been widely applied. However, the traditional approach suffers from three inherent defects: target unreachability, local minima, and local path oscillation. To address these issues, this paper proposes an Improved Artificial Potential Field (IAPF) method, innovatively incorporating three types of optimization mechanisms: 1) A target point distance-weighted function to reconstruct the repulsive field, ensuring terminal convergence; 2) A local exploration factor to achieve escape from local minima; 3) A directional weighting factor and a forward-looking decision-making mechanism to suppress path oscillations. Moreover, experiments have demonstrated the ability of the proposed algorithm to address the issues in traditional APF, as well as its robustness and superiority in complex three-dimensional environments.