NSGA-II based trajectory optimization on 3D point cloud for spray painting robots

Abstract

Spraying trajectory planning for complex surface has always been a challenging problem. On the basis of a spraying path generated with the 3D point cloud slice, this paper proposes a Non-dominated Sorting Genetic Algorithm (NSGA-II) based spraying trajectory optimization method for complex surfaces. Firstly, the point cloud slicing method is used to generate the initial spraying path of complex surfaces. In particular, aiming at the possible path self-intersection problem near the high curvature region on complex surfaces, a path smoothing method based on monotone chain is proposed to eliminate invalid path loops. Then, a multi-objective NSGA-II spray trajectory optimization method is proposed to optimize spray efficiency and spray quality by improving the spray speed distribution along the spray path. Finally, the proposed approach is verified on real automotive surfaces, the results show its robustness and efficiency in solving spray trajectory planning problems on complex surfaces.