Automated, CAD-Guided Area Sensor Planning Process For Coordinate Measurement Machines (CMMs)
Publication Date: 2005-Sep-26
The IP.com Prior Art Database
New optical area sensor technologies have made it possible to acquire millions of points simultaneously on complex free-form surfaces in the time it would previously have taken to measure a single point [5,6]. This opens up the possibility of rapid, on-line measuring the entire surface of cars, components, stamping tools and dies. Coordinate measurement machines (CMMs) are typically equipped with touch probes for dimensional inspection, but can potentially be retrofitted with area sensors to drastically increase their data acquisition throughput, thereby reducing inspection time. Additionally, area sensing would allow CMMs to expend beyond their traditional role of dimension inspection into a much boarder realm of quality control involving, for instance, sheet metal surface quality and highlights. This would lead to reduced quality inspection cost and timing, and ultimately improved customer satisfaction.
Sensor planning is a critical step in operating a CMM to acquire intended data. With touch probes, the number of points to locate on a part is limited and planning is done by hand in what is known as a "Click-and-Check" fashion. This is currently the only method available in commercial software for sensor planning. The CAD model is visualized in a graphics system. The user interface allows the user to "click" and insert locations on the part to be inspected. Animation can be performed such that collisions or in feasible kinematics conditions are revealed and corrected by the operator. This process is repeated for all points of interest.
Path planning for robotically manipulated area sensors presents unique challenges not found in typical path planning operations, including: glare, visibility, focus, stand-off/field-of-view, resolution, and margin of area overlap. Because there are a large number of factors to be considered simultaneously, it is very difficult for a human to do this path planning manually. In addition, the number of viewpoints to plan for the sensor might be very large due to size of the part surface and the required data resolution.
Hence it was the first objective of this work to automate the sensor path planning task for a CMM equipped with an optical area sensor. It was the second objective of this work to develop a seamless dataflow that starts from the CAD model of a part to be inspected and ends with a color map showing the deviation of the part from the CAD model. It was the third objective of this work to develop a complete task flow in achieving the CMM-based optical area sensing. It was the ftheth objective of this work to apply this technology to a range of applications, including, but not limited to, metrology, spray painting, and chopper gun trajectory planning.
The overall concept of the work is illustrated in Figure 1. The internal performance requirement for such a system is shown in Figure 2. Th...