Learning Bayesian models of sensorimotor interaction: from random exploration toward the discovery of new behaviors

Eva Simonin (1) , Julien Diard (1) , Pierre Bessiere (1)

1 E-MOTION - Geometry and Probability for Motion and Action

We are interested in probabilistic models of space and navigation. We describe an experiment where a Koala robot uses experimental data, gathered by randomly exploring the sensorimotor space, so as to learn a model of its interaction with the environment. This model is then used to generate a variety of new behaviors, from obstacle avoidance to wall following to ball pushing, which were previously unknown by the robot. The learned model can be seen as a building block for a hierarchical control architecture based on the Bayesian Map formalism.

Domaines: Autre [cs.OH]

Eva Simonin, Julien Diard, Pierre Bessiere. Learning Bayesian models of sensorimotor interaction: from random exploration toward the discovery of new behaviors. Proc. of the IEEE-RSJ Int. Conf. on Intelligent Robots and Systems, 2005, Edmonton, Canada. pp.1226--1231. ⟨inria-00182040⟩