Three dimensional object location in the eye-on-hand robot system

The algorithm of three dimensional object location for the eye-on-hand robot system is given. Based on the modified DLT algorithm the camera and robot system calibration algorithms are introduced. The parameters of the system calibration are obtained using the pseudo-inverse method. Two images are taken in different positions for the same scene. Based on the calibration parameters two sets of sight line equations are derived for various feature points in the two images. The corresponding points are solved using the least squares method according to the geometry constraints. Error matching can be avoided using the third image taken for the same scene. Finally, the three dimensional locations of object points are obtained accurately