Coordinate measuring machines (CMMs) are already widely used as a measuring tool in the manufacturing industry. Fast probing is now the trend for next generation CMMs. However, increases in the measuring velocity of CMMs are limited by dynamic errors that occur in CMMs.
In this paper, theoretical analysis and experimental research is used to create a systematic approach for modeling the dynamic errors of a touch-trigger probe CMM. First, an overall analysis of the dynamic errors of CMMs is given, and methods to improve the stiffness of air bearings are presented. Weak elements of the CMM are identified with a laser interferometer. The probing process, as conducted with a touch-trigger probe, is analyzed and dynamic errors are measured. Based on these analyses, the dynamic errors in touch-trigger probing are modeled using neural networks. In turn, dynamic errors are predicted. An approach to achieving software error compensation is discussed. Finally, the method and results from this study illustrate that it is possible to compensate for dynamic errors of CMMs.