Learning autonomous navigation abilities using radial basis functions networks

摘要

A system that learns how to react to visual inputs in order to accomplish simple autonomous navigation tasks is presented. The technique of radial basis functions networks along with their applications in problems of learning from examples is first outlined, and the various stages of the training process are then described in detail. Experiments are reported which show how, in driving a robot along a corridor, the system is able to attain a level of performances which is very close-at least as far as simulations are concerned-to the one displayed by its human trainers.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>