Neuro/fuzzy behavior-based control of a mobile robot in unknown environments

Abstract

A new navigation method in an unknown environments based on human combination skills of elementary behaviors is discussed in our case. These elementary behaviors are achieved by means of fuzzy reasoning scheme. Neural networks are used to select different behaviors so that the motion speed and rotational velocity of the mobile robot are changed smoothly. The sharp shift of difference behaviors will exacerbate the absolute position errors. An explanation of the algorithm is presented in detail. Finally, the feasibility of the proposed design is verified by simulation experiments.