A uniform allowance matching method for point cloud based on the edge extraction under de-shaping center

Abstract

For automated grinding, such as robotic grinding, high stability of the manufacturing system is required during the machining process. Little attention has been paid to the effect of the uniformity of the matching margin on the machining stability as a result of visual matching. In fact, it has an important relationship with maintaining the stability of robotic grinding and keeping the uniformity of grinding depth. In this paper, important process features of the blade part are extracted to improve the matching efficiency with a point cloud, while the normal direction of the contour after finishing grinding is used as the normal direction for registration. The objective function of minimization of the allowance variance is constructed to improve the uniformity of the grinding depth. The overall uniformity of the calculation results is also considered, and the effect of no tangential distance tampering is achieved by constraining the shape centers of the two point cloud models. The method in this paper is compared with the classical matching method to verify the feasibility of the margin variance minimization algorithm based on the decentered point cloud edges in terms of algorithm speed, convergence, accuracy, spatial complexity, and the uniformity of the obtained results.