| 摘要 |
<P>PROBLEM TO BE SOLVED: To teach gripping action of a work whose a shape and a three-dimensional position are unknown to a robot by intuitive and simple input operation by an operator. <P>SOLUTION: (a) A photographed image of a working space is displayed on a display device 21, and (b) operation for second-dimensionally designating a recognition area including a gripping portion in the work 90 gripped by a hand 11 on an image of the work 90 displayed on the display device 21 is received. (c) Operation for designating the original shape model fitted to the gripping portion from the plurality of original shape models is received, and (d) the designated original shape model is fitted to three-dimensional position data of a space corresponding to the recognition area, thereby, a group of parameters for designating the shape, the position and the attitude of the original shape model is determined. (e) Database 15 in which a gripping pattern applicable with the hand 11 relative to the respective original shape models is described are retrieved, and the gripping pattern applicable for gripping of the work 90 is selected. <P>COPYRIGHT: (C)2009,JPO&INPIT |
