Knowledge representation system for robot-based automated assembly

The complexity of the man-machine interface and the issues of controlling a robot and ensuring that the robot does indeed complete the task are significant obstacles to the use of robots in the assembly environment. We have developed a knowledge representation model for robots in the assembly environment. The proposed model provides a means of using apriori knowledge and reasoning capabilities for specifying, controlling, and monitoring the tasks performed by robots. This knowledge representation model facilitates robot programming and real time error prevention. We describe the object models for the proposed knowledge presentation system and present the constraint enforcing environment for the modeling of assemblies. This includes the representation of constraints and the rules for simplifying constraints and motion strategies for primitives and assemblies.