Neuro-fuzzy based vector field model: an unified representation for mobile robot guiding styles

Abstract

The development of autonomous robot navigation systems has been a challenge for robot specialists for decades. Different guiding or navigation styles are required for autonomous robots to operate in the intelligent space efficiently and safely under different situations. In this paper an extension of the classical potential based guiding model is presented. The vector field based guiding model is proposed as a unified representation for guiding style model to be used in intelligent space. Simulations have been conducted to demonstrate the effectiveness of the proposed guiding style model and its neuro-fuzzy realization.