Learning force-based assembly skills from human demonstration for execution in unstructured environments

M. Skubic, R.A. Volz
Proceedings. 1998 IEEE International Conference on Robotics and Automation (Cat. No.98CH36146)  
Robots have been used successfully in structured settings, where the environment is controlled; this research is inspired by the vision of robots moving beyond the structured, controlled settings. The work focuses on the problem of learning low-level force-based assembly skills from human demonstration. To avoid position dependencies, force-based discrete states are used to describe qualitatively how contact is being made with the environment. Sensorimotor skills are modeled using a hybrid
more » ... ol model, which provides a mechanism for combining continuous low level force control with higher level discrete event control. A change in qualitative, discrete state constitutes an event and triggers a new control command to the robot. In this way, the skill execution is not dependent on absolute position but rather responds to changes in the forcebased qualitative state. Experimental results are presented which validate the approach and show how skill acquisition can be accomplished even with an imperfect demonstration. 1 Introduction Robots have been used successfully in manufacturing settings, where the environment is very structured and the tasks performed are repetitive and relatively simple. As long as the environment is controlled and the workpieces are con ned to precise positions and orientations, the robot can continue to execute its task. This research is inspired by the vision of robots moving beyond the structured, controlled settings, and still performing successful assembly operations. Although many challenging problems still exist, it is realistic to think of robots being used as intelligent tools and assistants which make the job easier for the human worker, extend his capability, or allow
doi:10.1109/robot.1998.677279 dblp:conf/icra/SkubicV98 fatcat:ik3m6retc5d7fe2dgqyl3ecfhy