Mapping of probe pretravel in dimensional measurements using neural networks computational technique

Abstract Pretravel is a major error source in dimensional measurements on coordinate measuring machine (CMM) and machine tools using trigger touch probes in modern manufacturing systems. The direction-dependent characteristic of pretravel needs effective and efficient computational solutions. In this paper, a back-propagation neural network is designed and implemented to establish a pretravel map for touch trigger probes. Experimental pretravel information from a CMM is used to train the network, and the network is used to predict pretravel in various probe approach directions. Performance of the trained neural network has been evaluated and effective pretravel predication is demonstrated. Thanks to the parallel-computing nature of neural network technique, pretravel mapping can be implemented in real time once the off-line-trained network is obtained. The proposed approach can be incorporated into the machine systems to provide real-time pretravel correction in dimensional measurement processes. We also reported the advantages of neural network technology over the traditional method used in this application.