AGV Path Planning Based on Improved A-star Algorithm
Dengxing Zhang, Chen Chen, Guanyu Zhang
Abstract
Traditional A-star algorithms encounter challenges such as an excess of turning points, significant turning angles, and proximity of turning points to obstacles. These issues can lead to increased energy consumption, hinder Automated Guided Vehicles(AGV) motion control, and elevate the risk of collisions, particularly when turning points cluster near obstacles. In response, we present an improved A-star based path planning algorithm. Our approach addresses these challenges by incorporating innovative modifications to the cost function, dynamically adjusting weights based on diverse turning angles to curtail unnecessary turns. Additionally, we integrate the artificial potential field method, introducing a unique heuristic function. This function includes a penalty term accounting for obstacle information at turning points, ensuring that these points are strategically positioned away from obstacles. Simulation results demonstrate the effectiveness of our algorithm in comparison to traditional A-star algorithms. It successfully reduces the number of turning points in the path, maintains a safe distance from obstacles, resulting in smoother paths, and proves advantageous for precise path tracking and control.