Development of a parametric model and virtual machine to estimate task specific measurement uncertainty for a five-axis multi-sensor coordinate measuring machine

Any combination of contact and non contact sensors on a single coordinate measuring machine makes it a multi-sensor or hybrid coordinate measuring machine. The ability to change the probing device without re-establishing the part coordinate system is a significant advantage and consequently the reason for their wide acceptance and usage. In addition to the existing three linear stages in a conventional or hybrid coordinate measuring machine, the number of axes can be augmented with two rotary stages in order to make them more versatile. This not only complicates the metrology loop but also adds more parametric errors to existing list of known 21 errors for a three stage conventional measuring machine. Also different probing sensors in the system have different metrology loops. Literature survey reveals that there is no previous or current work on estimation of task specific uncertainty for machines of this class with a wide potential for measuring complex parts quickly and with great ease. The current work focuses on developing a virtual measuring machine of this class which incorporates a parametric model of a five-axis multi-sensor coordinate measuring machine, techniques to estimate the parametric errors in the model and error correction strategies for the same. The virtual machine is used to mimic real time measurements in order to come up with estimates for task specific measurement uncertainty.