SPATIAL-LEARNING FOR NAVIGATION IN DYNAMIC ENVIRONMENTS

This article describes techniques that have been developed for spatial learning in dynamic environments and a mobile robot system, ELDEN, th at integrates these techniques for exploration and navigation. In this research, we introduce the concept of adaptive place networks, increm entally constructed spatial representations that incorporate variable- confidence links to model uncertainty about topological adjacency. The se networks guide the robot's navigation while constantly adapting to any topological changes that are encountered. ELDEN integrates these n etworks with a reactive controller that is robust to transient changes in the environment and a relocalization system that uses evidence gri ds to recalibrate dead reckoning.