A Unique Sensor Fusion System for Coordinate Measuring Machine Tasks

This paper describes a real-time hierarchical system that combines (fuses) data from vision and touch sensors to simplify and improve the operation of a coordinate measuring machine (CMM) used for dimensional inspection tasks. Our emphasis is on sensory processing techniques that can aid CMM applications rather than on the analysis of CMM performance measurements. Our system consists of sensory processing, world modeling, and task decomposition modules. It uses the strengths of each sensor-- the accuracy of the CMM scales and the analog touch probe and the global information provided by a low resolution camera--to simplify the inspection task while maintaining CMM accuracy. In the experiment described, the vision module performs all computations in image coordinate space. The system fuses data obtained from the vision system in image coordinates with the velocity and probe position provided by the CMM controller.The fused information provides feedback to the motion controller as it guides the probe during a raster scan. We also describe a method for combining information from the vision system and the probe in real- time to simplify the data acquisition process required for camera calibration tasks. We autonomously register 2-D and 3-D points as the probe moves along a pre-programmed path. These corresponding points are used as input to a calibration algorithm