A three-dimensional triangular vision-based contouring error detection system and method for machine tools

Abstract

Contouring error detection for machine tools can be used to effectively evaluate their dynamic performances. A triangular vision-based contouring error detection system and method is proposed in this paper, realizing the three-dimensional error measurement of an arbitrary trajectory in conditions of a high feed rate and wide motion range. First, a high-precision measurement fixture, which consists of high-precision circular coded markers and a highly uniform light source, is designed to accurately characterize the motion trajectory of a machine tool and realize the high-quality collection of an image sequence. Then, to improve the contouring error detection accuracy, a coded marker decoding and center location method for the automatic recognition and high-precision center positioning of the circular coded markers are applied. Using image preprocessing and matching, the markers’ three-dimensional coordinates in the camera coordinate system can be constructed. Moreover a data transformation method induced by the orthogonal motion of machine tools is proposed to obtain the three-dimensional trajectory in the machine tool coordinate frame and the contouring error can be calculated. Finally, a three-dimensional contouring error detection study of an equiangular spiral interpolation at different feed rates is performed in the laboratory. It is shown that the average contouring error for a feed rate of 1000mm/min is about 3µm, which verifies the vision measurement accuracy and feasibility.