Route learning in mobile robots through self-organisation

Abstract
This paper describes route-learning experiments with an autonomous mobile robot in which mapbuilding is achieved through a process of unsupervised clustering of sensory data. The resulting topological mapping of the robot's per- ceptual space is used for subsequent navigation tasks such as route following. After the autonomous mapbuilding process is completed, the acquired generalised perceptions are associated with motor actions, enabling the robot to follow routes au- tonomously. The navigation system has been tested exten- sively on a Nomad 200 mobile robot, it is reliable and copes with noise and variation inherent in the environment. One important aspect of the mapbuilding and route fol- lowing system described here is that relevance or irrele- vance of perceptual features is determined autonomously by the robot, not through predefinition by the designer. Sec- ondly, further to previous work ((6)) the presented route learning system enables the robot to use the map for as- sociation of perception with action, rather than localisation alone.

This publication has 4 references indexed in Scilit: