PERCEPTION-BASED LEARNING FOR MOTION IN CONTACT IN TASK PLANNING

This paper presents a new approach to error detection during motion in contact under uncertainty for robotic manufacturing tasks. In this ap proach, artificial neural networks are used for perception-based learn ing. The six force-and-torque signals from the wrist sensor of a robot arm are fed into the network. A self-organizing map is what learns th e different contact states in an unsupervised way. The method is inten ded to work properly in complex real-world manufacturing environments, for which existent approaches based on geometric analytical models ma y not be feasible, or may be too difficult. It is used for different t asks involving motion in contact, particularly the peg-in-hole inserti on task, and complex insertion or extraction operations in a flexible manufacturing system. Several real examples for these cases are presen ted.