Path planning is a critical component of autonomous vehicles, intelligent transportation, and robotic navigation. However, conventional algorithms such as A*, Dijkstra, bidirectional A*, and RRT often suffer from local optima, redundant turning points, excessive cumulative turning angles, and computational overhead in maintaining safe distances—factors that limit their real-time performance and stability in complex, high-resolution maps.
To address these issues, we propose a path planning optimization framework that integrates morphological preprocessing with polar-coordinate-based keypoint extraction. Morphological dilation is applied to map representations to adaptively enforce obstacle clearance during planning, improving robustness and efficiency. Subsequently, a polar-coordinate keypoint extraction strategy reduces redundant nodes, minimizes cumulative turning angles, and alleviates local optima.
The approach was implemented on the ROS–Gazebo simulation platform and systematically compared with multiple baseline algorithms across maps of varying sizes. Experimental results demonstrate that the proposed method reduces path nodes by over 50%, decreases cumulative turning angles by 85%, shortens path length by 23.8%, and improves planning response time by 69%. These results highlight the effectiveness of the method in enhancing path smoothness, computational efficiency, and global optimality, providing a reliable basis for safe autonomous navigation in complex environments.
If you have any questions about submitting your review, please email us at [email protected].