Mobile Robot Mapping and Localization in Non-Static Environments

Book One

Volume

Proceedings of the AAAI Conference on Artificial Intelligence, 20

Robotics

Whenever mobile robots act in the real world, they need to be able to deal with non-static objects. In the context of mapping, a common technique to deal with dynamic objects is to filter out the spurious measurements corresponding to such objects. In this paper, we present a novel approach to estimate typical configurations of dynamic areas in the environment of a mobile robot. Our approach clusters local grid maps to identify the possible configurations. We furthermore describe how these clusters can be utilized within a Rao-Blackwellized particle filter to localize a mobile robot in a non-static environment. In practical experiments carried out with a mobile robot in a typical office environment, we demonstrate the advantages of our approach compared to alternative techniques for mapping and localization in dynamic environments.