SEARCH OF STEADY-STATE GENETIC ALGORITHMS FOR VISION-BASED MOBILE ROBOTS

摘要

This chapter discusses the searching capability of genetic algorithms for vision-based mobile robots from the viewpoints of ecological psychology. The perceptual system of a mobile robot is restricted by the action system in the spatio-temporal context of the facing environment. In this chapter, we propose a time-series of a searching method in the visual perception according to the current situation and action outputs. Furthermore, we show several experimental results of two different mobile robots.