A 3-D flexible vision system for robotic assembly cell: an artificial intelligence based approach

Determining the identities, position and orientations of randomly placed objects in 3-D space is of fundamental importance in industrial robotics. To precisely accomplish this, a robot must be equipped with a 3-D vision sensor and must be able to interpret vision data to acquire information yielding intrinsic characteristics of objects. This paper presents a 3-D flexible vision system which has been developed using an artificial intelligence and knowledge based approach for a robotic assembly cell. The developed system can robustly recognise any number of objects of any shape, where the objects could be with any random orientation and position, and lying on any side or face. Depending on the shape and size of the objects introduced to the system, it automatically selects a minimum number of important recognition features by using its artificial intelligence program for robustly recognising all the objects, and it automatically modifies the recognition program based on the selected recognition features. After recognition of the object, in order to manipulate the objects by the robot, the system generates 3-D position and orientation data using the moment methods and the data in the knowledge based system