J. Acosta, J.M. Molina, and A. Sanchis (Spain)
Rules Extraction, Robot Control, Braitenberg Model,Classifier Systems
In some cases when applied biological inspired learning systems, the learned concepts are unreadable for humans. We present in this paper an alternative to obtaining rule representation of a Braitenberg robot controller. In the first step we acquire a data set resulting from the execution of the controller in different environments. A Classifier System (CS) is used for automatically generating the symbolic rules. A CS combine the execution capabilities of symbolic systems and the learning capabilities of Genetic Algorithms. The results show that the obtained symbolic rules system is very accurate modeling the controller.
Important Links:
Go Back
