Abstract
We present a method for autonomous learning of dextrous manipulation skills with multifingered robot hands. We use heuristics derived from observations made on human hands to reduce the degrees of freedom of the task and make learning tractable. Our approach consists of learning and storing a few basic manipulation primitives for a few prototypical objects and then using an associative memory to obtain the required parameters for new objects and/or manipulations. The parameter space of the robot is searched using a modified version of the evolution strategy, which is robust to the noise normally present in real-world complex robotic tasks. Given the difficulty of modeling and simulating accurately the interactions of multiple fingers and an object, and to ensure that the learned skills are applicable in the real world, our system does not rely on simulation; all the experimentation is performed by a physical robot, in this case the 16-degree-of-freedom Utah/MIT hand. E xperimental results show that accurate dextrous manipulation skills can be learned by the robot in a short period of time. We also show the application of the learned primitives to perform an assembly task and how the primitives generalize to objects that are different from those used during the learning phase.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Arbib, M., Iberall, T. & Lyons, D. (1983). Coordinated control programs for movements of the hand. Technical Report 83–25, Department of Computer and Information Science, University of Massachusetts at Amherst, Amherst, Massachusetts.
Arts, E.H.L. & Korst, J. (1989). Simulated Annealing and Bolzmann Machines. Wiley, Chichester.
Bellman, R.E. (1957). Dynamic Programming. Princeton University Press, Princeton, NJ.
Born, J. (1978). Evolutionsstrategien zur numerischen Lösung von Adaptationsaufgaben. PhD thesis, Humboldt Universität, Berlin, Germany.
Christiansen, A.D., Mason, M.T. & Mitchell, T.M. (1990). Learning Reliable Manipulation Strategies without Initial Physical Models. In Proceedings of the 1990 IEEE International Conference on Robotics and Automation, pages 1224–1230, Cincinnati, Ohio.
Dunn, G.B. & Segen, J. (1988). Automatic discovery of robotic grasping configurations. In Proceedings of the 1988 IEEE International Conference on Robotics and Automation, pages 396–401.
Eiben, A.E., Aarts, E.H.L. & Van Hee, K.M. (1991). Global convergence of genetic algorithms: an infinite Markov chain analysis. In Proceedings of the First International Conference on Parallel Problem Solving from Nature, pages 4–12, Berlin, Germany. Springer.
Fuentes, O. & Nelson, R.C. (1996a). Experiments on dextrous manipulation without prior object models. In Proceedings of the 1996 IEEE International Symposium on Intelligent Control, Dearborn, Michigan.
Fuentes, O. & Nelson, R.C. (1996b). The virtual tool approach to dextrous telemanipulation. In Proceedings of the 1996 IEEE International Conference on Robotics and Automation, pages 1700–1705, Minneapolis, Minnesota.
Grossman, T. & Davidor, Y. (1992). An investigation of a genetic algorithm in continuous parameter space. Technical Report CS92–20, The Weizmann Institute of Science, Department of Applied Mathematics and Computer Science, Revhovot, Israel.
Iberall, T. (1987). The nature of human prehension: Three dextrous hands in one. In Proceedings of the 1987 IEEE International Conference on Robotics and Automation, pages 396–401, Raleigh, North Carolina.
Jacobsen, S., Iversen, E., Knutti, D., Johnson, R. & Bigger, K. (1986). Design of the Utah/MIT Dextrous Hand. In Proceedings of the 1986 IEEE International Conference on Robotics and Automation, pages 96–102.
Jägersand, M., Fuentes, O. & Nelson, R.C. (1996). Acquiring visual-motor models for precision manipulation with robot hands. In Proceedings of the Fourth European Conference on Computer Vision, Cambridge, U. K.
Kamon, I., Flash, T. & Edelman, S. (1996). Learning to grasp using visual information. In Proceedings of the 1996 IEEE International Conference on Robotics and Automation, pages 2470–2476, Minneapolis, Minnesota.
Maes, P. & Brooks, R.A. (1990). Learning to coordinate behaviors. In Proceedings of AAAI-90, pages 796–802.
Matarić, M. (1994). Reward functions for accelerated learning. In Proceedings of the Eleventh International Conference on Machine Learning, pages 181–189.
Matarić, M. (1997). Learning social behavior. Robotics and Autonomous Systems, 20:191–204.
Michelman, P. & Allen, P. (1993). Compliant manipulation with a dexterous robot hand. In Proceedings of the 1993 IEEE International Conference on Robotics and Automation, pages 711–716, Atlanta, Georgia.
Narasimhan, S. (1988). Dexterous robot hands: Kinematics and control. Master's thesis, MIT Artificial Intelligence Laboratory, Cambridge, Massachusetts.
Nelson, R.C. Jägersand, M. & Fuentes, O. (1995). Virtual tools: A framework for simplifying sensory-motor control in complex robotic systems. In Proceedings of the 1995 Workshop on Vision for Robots, Pittsburgh, PA.
Quinlan, J.R. (1986). Induction of decision trees. Machine Learning, 1:81–106.
Rechenberg, I. (1973). Evolutionsstrategie: Optimierung technischer Systeme nach Prinzipien der biologischen Evolution. Frommann-Holzboog Verlag, Stuttgart.
Salganicoff, M., Kunin, L.G. & Ungar, L.H. (1994). Active exploration based ID-3 learning for robot grasping. In 1994 Workshop on Robot Learning, New Brunswick, NJ.
Schwefel, H.-P. (1981). Numerical Optimization of Computer Models. John Wiley & Sons, Ltd.
Speeter, T.H. (1991). Primitive Based Control of the Utah/MIT Dextrous Hand. In Proceedings of the IEEE International Conference on Robotics and Automation, pages 866–877, Sacramento, California.
Rights and permissions
About this article
Cite this article
Fuentes, O., Nelson, R.C. Learning Dextrous Manipulation Skills for Multifingered Robot Hands Using the Evolution Strategy. Machine Learning 31, 223–237 (1998). https://doi.org/10.1023/A:1007409228154
Issue Date:
DOI: https://doi.org/10.1023/A:1007409228154