This paper proposes an optimized path planning method for robotic systems in automated fiber placement (AFP) on complex surfaces. The purpose of this paper is to enhance path accuracy and continuity by addressing cumulative error issues on nonuniform surfaces.
The method builds upon an improved fast marching method and incorporates an angle bisector method and back propagation method to refine path planning. A post-processing strategy is also introduced to mitigate computation dependencies, enabling smoother and more accurate path generation across intricate surface geometries.
Experimental results indicate that the proposed path planning method significantly improves both smoothness and precision compared to conventional approaches, showing consistent performance across diverse complex surface models.
This study provides a robust path planning framework for robotic AFP applications, offering a solution with enhanced accuracy and reliability. The angle bisector, back propagation and post-processing strategy introduced in this paper may be extended to other robotic systems requiring precise path planning on complex geometries.
