| 摘要 |
PURPOSE:To enhance percentage of automatization of work by storing features of a part to be worked as its shade and shadow and recognizing the part to be worked on a basis of this stored shade and shadow. CONSTITUTION:In a robot used for assembling process, positional deviation between a working part and the part to be worked is corrected by the picture signal from an image pickup means such as a camera or the like. In this case, depths of valley parts of the shade and shadow of the part to be worked are detected in the first detecting means B in accordance with said picture signal, and widths of valley parts are detected in the second detecting means C. Both detection signals are inputted to an operating means D, and depths are divided by windths of valley parts to attain sharpness of valley parts. Pertinent data corresponding to plural working parts are selected by a selecting means E in accordance with these sharpness and depths and are stored in a storage means F in the teaching mode. When the play back mode is set by a mode setting means A, the part to be worked is recognized in a means G by said stored depth and sharpness data of the part to be worked, the result in the means B of a working start target point, and the result in the means D, and the position is corrected on a basis of recognition result. |
