This paper describes a novel technique for position error compensations of robots based on a fuzzy error interpolation method. A traditional robot calibration implements either model or modelless methods. The compensation of position error in a model-less method is to move the robot's end-effector to a target position in the robot workspace, and to find the target position error online based on the measured neighboring four-point errors around the target position. For this purpose, a stereo camera or other measurement device can be used to measure offline the position errors of the robot's end-effector at predefined grid points. By using the proposed fuzzy error interpolation technique, the accuracy of the position error compensation can be greatly improved, which is confirmed by the simulation results given in this paper. A comparison study among various interpolation methods, such as bilinear, cubic spline, and the fuzzy error interpolation technique is also made via simulation. The simulation results show that more accurate compensation results can be achieved using the fuzzy error interpolation technique compared with its bilinear and cubic spline counterparts.
