Reactive navigation of mobile robot by temporal radial basis function approach

Abstract

In this article we present, a new approach based on temporal radial basis functions "TRBF", in aim to developing a reactive navigation module based on neural networks. We applied this module to a structured type like interior of building; in such case the robot must assure the navigation in middle while avoiding obstacles without wandering, with the possibility to take in account its decisions in his past lasting trajectory. Our TRBF is based on orthogonal least square algorithm "ols".