Abstract: To address issues such as high node count, non-smooth paths, and wide exploration range encountered by traditional algorithms in automated guided vehicles (AGV) path planning, an improved A^* algorithm is proposed. Firstly, a grid-based method is employed in the environmental modeling. Secondly, the heuristic function is enhanced by incorporating a weighted obstacle ratio. Additionally, collision avoidance functions are introduced to mitigate collision risks. Finally, corner optimization is applied to reduce the number of turns during AGV operations. Experimental results demonstrate significant improvements. In simple environments, the improved A^* algorithm reduces node traversal by 85.3% and 55.9% compared to two traditional algorithms. In complex environment, the number of traversal nodes of the improved A^* algorithm is reduced by 94.5% and 70.3%. Moreover, the optimized algorithm decreases the number of corners, ensures smoother paths, shortens planning times, enhances operational efficiency, and effectively lowers collision probabilities.