Research on AGV Path Planning Algorithm Integrating Adaptive A* and Improved APF Algorithm

Aiming at the issues of low search efficiency, insufficient kinematic adaptability, and weak obstacle avoidance robustness in automated guided vehicle (AGV) global path planning under static environments, this paper proposes a fusion algorithm that integrates an adaptive A* algorithm with an improved artificial potential field (APF) algorithm. The algorithm dynamically adjusts the heuristic weight of the A* algorithm based on local obstacle complexity and incorporates an Ackermann kinematic constraint-based steering traj ectory generation mechanism for node exploration. This mechanism employs a hierarchical collision detection strategy to ensure steering traj ectories remain within feasible domain boundaries. Furthermore, the repulsion function of the APF algorithm was optimized during the fusion process. By introducing vehicle geometric feature parameters and a dynamic safety threshold compensation factor, an adaptive repulsive gradient field was constructed to enhance obstacle avoidance range optimization in complex obstacle clusters. Simulation experiments demonstrate that the proposed algorithm improves both obstacle avoidance performance and path cost under low- and high-obstacle-density environments, validating its advantages in motion continuity and environmental adaptability.