Fast 3D Perception for Collision Avoidance and SLAM in Domestic Environments

Autonomous service robots that assist in housekeeping, serve as butlers, guide visitors through exhibitions in museums and trade fairs, or provide care to elderly and disabled people could substantially ease everyday life for many people and present an enormous economic potential. However, these service robots face the challenging task of operating in real-world indoor and domestic environments, which tend to be cluttered, dynamic and populated by humans and domestic animals. In order to adequately react to sudden dynamic changes and avoid collisions, these robots need to be able to constantly acquire and process, in real-time, information about their environment.