Abstract
In this paper we describe a hybrid approach to solve a real-world robotic task with uncertainty. The solution is based on the integration of unsupervised learning of task features and reinforcement learning of the correspondence between situations and actions. We seek for inspiration in the behavior of people performing manipulation tasks. The proposed approach clearly separates the programmed skills from the learned knowledge. A real-world example is presented which shows how the robot, starting from a pure random strategy, improves its performance and becomes more skillful with the task.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Kohonen, T., Self-Organizing Maps, Springer Series in Information Sciences, 30, Springer, 1995.
Sutton, R. S. (editor), Reinforcement Learning, Kluwer Academic Publishers, (reprinted from Machine Learning, 8, 1992).
Watkins, C. J. C. H. and Dayan, P., Q-Learning, Machine Learning, 8, 279–292, 1992.
Erdmann, M., Randomization in Robot Tasks, International Journal of Robotics Research, 11, 5, 399–436, 1992.
Hongler, M., Badano, F., Betemps, M., and Jutard, A., A Random Exploration Approach for Automatic Chamferless Insertion. International Journal of Robotics Research, 14, 161–173, 1995.
Cervera, E., del Pobil, A. P., Marta, E., and Serna, M. A., Perception-Based Learning for Motion in Contact in Task Planning. Journal of Intelligent and Robotic Systems, 17, 283–308, 1996.
Sun, R., and Bookman, L. A. (editors), Computational Architectures Integrating Neural and Symbolic Processes, Kluwer Academic Publishers, 1995.
Cervera, E., del Pobil, A. P., Marta, E., Serna, M. A., A Sensor-Based Approach for Motion in Contact in Task Planning, Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, IROS'95, vol. 2, pp. 468–473, Pittsburgh, USA, 1995.
Ultsch, A., Self-Organized feature maps for monitoring and knowledge acquisition of a chemical process, Proc. Int. Conf. Artif. Neural Nets. (ICANN93), pp. 864–867, Springer-Verlag, 1993.
Cervera, E. and del Pobil, A.P. (1997). Force-Based Robot Control for Dexterous Tasks. In: Proc. IFAC-IFIP-IMACS Conference on Control of Industrial Systems, Belfort, France. Edited by: L. Grujic, P. Borne and M. Fernet, Elsevier Science (in press).
Gullapalli, V., J. Franklin, and H. Benbrahim (1994). Acquiring Robot Skills via Reinforcement Learning. IEEE Control Systems, 1, 13–24.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Cervera, E., del Pobil, A.P. (1997). Integration of Self-Organizing Feature Maps and reinforcement learning in robotics. In: Mira, J., Moreno-Díaz, R., Cabestany, J. (eds) Biological and Artificial Computation: From Neuroscience to Technology. IWANN 1997. Lecture Notes in Computer Science, vol 1240. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0032595
Download citation
DOI: https://doi.org/10.1007/BFb0032595
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63047-0
Online ISBN: 978-3-540-69074-0
eBook Packages: Springer Book Archive