Abstract

In this paper we describe Magellan, an integrated architecture for mobile robotics. The system represents its spatial knowledge in terms of a topological network that connects a set of distinct places, each represented by evidence grids that contain probabilistic descriptions of occupancy. Magellan includes a module for place recognition that determines its initial location and when it has reached a goal, a module for continuous localization that maintains accurate estimates of the robot's position, and a module for navigation that generates path plans and executes them using reactive behaviors. Experiments in two laboratories with different characteristics suggest that the system can operate robustly across a range of environments, including ones that involve dynamic changes. This work was funded by the Office of Naval Research under Grant No. N00014-94-1-0505. We thank Hans Moravec for sharing his software, and we thank Nils Nilsson and Barbara Hayes-Roth for providing access to thei...

Citations