Learning model-free motor control

Conference Article


European Conference on Artificial Intelligence (ECAI)





Doc link


Download the digital copy of the doc pdf document


Some robotic tasks require an accurate control to follow the desired trajectory in the presence of unforeseen external disturbances and system parameters variations. In this case conventional control techniques such as PID must be constantly readjusted and a compromise solution must be adopted. This problem can be avoided using a learning process that automatically learns the appropriate control law and adapts to ongoing system variations. But a drawback of many learning systems is that they are not effective for non-toy problems. In this paper we present the results obtained with a categorization and learning algorithm able to perform efficient generalization of the observed situations, and learn accurate control policies in a short time without any previous knowledge of the plant.


control theory, generalisation (artificial intelligence), intelligent robots, learning (artificial intelligence).

Author keywords

reinforcement learning, categorization

Scientific reference

A. Agostini and E. Celaya. Learning model-free motor control, 16th European Conference on Artificial Intelligence, 2004, València, Espanya, in ECAI 2004: Proceedings of the 16th European Conference on Artificial Intelligence, Vol 110 of Frontiers in Artificial Intelligence and Applications, pp. 947-948, 2004, IOS Press, Amsterdam, Països Baixos.