Knowledge Based Error Recovery in Industrial Robots

This paper describes the main aims of a new research project concerned with the implementation of automatic error recovery f ac i l i t i e s in industr ial robotics. An approach is discussed in which an existing manufacturing work ce l l is to be enhanced by the addition of a task event model contained in an error recovery knowledge base. This paper outlines the main design issues involved in this work. Reasons why conventional fault tolerance techniques are inadequate are given and the industr ial application is explained. A part icular approach to sensory monitoring and error diagnosis is described. The proposed system has more s imi lar i t ies with sensory driven expert systems than with body modelling contingency planners.