In this paper, we present a system level approach tosmartwheelchairsystem(SWS)navigationinurbanenvironments. The proposed SWS ecosystem has two primary components: a mapping service which generates large-scale landmark maps, and the SWS vehicle itself, which is a clientofthemappingservice.TheSWSprototypeintegrates 3D LIDAR/imaging systems which provide robust perception in unstructured, outdoor environments. It also leverages these same sensors for map-based localization. In demonstrating the efficacy of the approach, the SWS navigated autonomously over a distance of more than 12 km in a representative urban environment without once losing localization, and without the use of GPS.

EB00000003557K

Available

• A smart wheelchair ecosystem for autonomous navigation in urban environments