ROBOT CONTROL BASED ON IMAGERY INFORMATION WITH AMBIGUITY.

Human-like control method of an industrial arm-robot with five joints is presented. The main idea is based on the concept of probabilistic sets in extended fuzzy expression. Ambiguous instructions in terms of membership and vagueness are generated by the robot itself using imagery data from ITV cameras. These instructions are divided into several classes based on the imagery information about the object such as the size and the shape, and the positional relationships between the hand of the robot and the object which consist of two factors (direction and distance). The robot automatically moves his hand toward the object according to the instructions and finally catches the object.