Abstract: In order to improve the accuracy of large-scale CMMs and correct spatial geometric errors, the laser tracing multi-station measurement method was used, instead of physical benchmark, to solve CMM geometric errors based on the elastic net algorithm. Based on laser tracing multi-station measurement, combined with L-M algorithm, the high-precision measurement of the volume errors of CMM space planning points was realized, and the measuring efficiency was improved. The elastic net algorithm was used to solve CMM quasi-rigid body model, and the problem of multicollinearity in the model solutions, and to realize the solution of geometric errors of the CMMs. The relationship model between volume errors and uniaxial geometric errors was combined with the part of coefficients of 0 in equations to solve geometric errors. In the experiments, a laser tracing multi-station measurement system was built, and the volume errors of the CMMs spatial planning points were measured. The results show that the method proposed herein may effectively solve the geometric errors of large-scale CMMs.

Key words: coordinate measuring machine(CMM), geometric error, elastic net, laser tracing multi-station measurement

摘要： 为提高大型三坐标测量机(CMM)的精度，修正空间几何误差，研究利用激光追踪仪多站位测量技术取代实物基准，提出了基于弹性网络算法求解CMM几何误差的方法。基于激光追踪仪多站位测量技术，结合L-M算法，实现CMM空间规划点体积误差的高精度测量，有效提高测量效率。利用弹性网络算法解算CMM准刚体模型，解决模型求解存在多重共线的难题，实现CMM几何误差的求解；将方程组中部分系数为0的项结合体积误差与单轴几何误差的关系模型来求解几何误差。实验搭建了激光追踪多站位测量系统，测量了CMM的空间待测点体积误差。实验结果表明，提出的方法可以有效求解大型CMM的几何误差。

关键词: 坐标测量机, 几何误差, 弹性网络, 激光追踪多站位测量