Minimum viewpoint planning for dimensional inspection of sheet metal parts

In automotive manufacturing, many sheet metal body parts need to be dimensionally inspected for quality control purpose. To efficiently inspect these parts, it is highly desirable to obtain the minimum number of camera viewpoints with each satisfying all the given task constraints. However, due to the NP-hardness of the general minimum-viewpoint problem, new methods should be found to handle it efficiently. In this paper, a novel approach is developed to solve this problem into its sub-optimality. Based on a decomposition approach, this approach first generates a pool of candidate viewpoints. Then an integer optimization problem-set-partition problem is formulated to find the minimum viewpoints. Experimental results on realworld parts demonstrate the effectiveness of the new approach.