초록 ▼

A method and system are provided for automatically determining the position and attitude of a three-dimensional body at a work station by utilizing three cameras each of which generates a non-overlapping plane of image data, including a single target point of the body without the use of structured l

A method and system are provided for automatically determining the position and attitude of a three-dimensional body at a work station by utilizing three cameras each of which generates a non-overlapping plane of image data, including a single target point of the body without the use of structured light. The locations of the target points are determined and processed within a programmed computer together with calibration data relating to the expected position of the body in the work station to provide data relating to the position and attitude of the body with respect to the work station. The resultant data relating to the position and attitude of the body are subsequently transformed into a first set of offset data relating to the difference between the actual and expected positions of the body. The first set of offset data is then transformed into a second set of offset data in the coordinate system of a peripheral device such as a robot, programmable controller, numerically controlled machine, etc. Finally, the programmed computer transfers the second set of offset data to the peripheral device which utilizes the data to modify its pre-programmed path.

대표청구항 ▼

Method for automatically determining the position and attitude of a three-dimensional body in space having at least three spaced apart target points at a vision station having a fixed coordinate system, said method comprising the steps of: generating calibration data defining a nominal position of t

Method for automatically determining the position and attitude of a three-dimensional body in space having at least three spaced apart target points at a vision station having a fixed coordinate system, said method comprising the steps of: generating calibration data defining a nominal position of the body at the vision station; generating substantially simultaneously a single plane of image data for each of said target points, each plane of image data containing its respective target point and each plane of image data being non-overlapping with respect to each of the other planes of image data; determining the locations of the target points within their respective image planes to obtain positional data; and processing the calibration data and the positional data together to provide data defining the position and attitude of the body with respect to the fixed coordinate system.