Autonomous robots must be able to learn and maintain models of their environments. Research on mobile robot navigation has produced two major paradigms for mapping indoor environments: grid-based and topological. While grid-based methods produce accurate metric maps, their complexity often prohibits efficient planning and problem solving in large-scale indoor environments. Topological maps, on the other hand, can be used much more efficiently and facilitate interfacing with symbolic problem solvers such as GOLOG, yet accurate and consistent topological maps are considerably difficult to learn and maintain in large-scale environments.
In this talk I will present an approach that integrates both paradigms: grid-based and topological. Grid-based maps are learned using artificial neural networks and Bayesian integration. Topological maps are generated on top of the grid-based maps, by partitioning the latter into coherent regions. By combining both paradigms, the new approach gains the best of both worlds: accuracy/consistency and efficiency. I will discuss empirical results for autonomously operating a mobile robot equipped with sonar sensors in populated multi-room environments, and outline our current research on an autonomous robotic tour-guide.