Fuzzy visual navigation using behavior primitives for small humanoid robot

Abstract

In this paper, we present a fuzzy visual navigation method that uses behavior primitives for humanoid robots. We define behavior primitives that consist of locomotion and motion primitives. The fuzzy navigation system consists of four control algorithms: autonomous walking, target tracking, obstacle avoidance, and behavior control based on marker recognition. We verify the proposed method through navigation experiments by using a developed small humanoid robot. The experimental results demonstrate that the humanoid robot can navigate efficiently and stably to the target and show improved performance of behavior planning for humanoid navigation. © 2013 Springer-Verlag.