GPU accelerated real-time traversability mapping

Abstract

The navigation of autonomous mobile robots requires effective localization and mapping modules. Dense map representation of the robot surroundings, which contains detailed information of the drivable region can be easily used for motion planning. To build a dense map on mobile robots, the main challenge is that the system has to be efficient due to the limited computational resources. In this paper, we propose a novel approach to generate a dense map with drivable information. First, the dense map with elevation information is generated by the proprioceptive localization results acquired from kinematic and inertial measurement, as well as the accumulated raw data from the range sensor. Then, we calculate slope and roughness of each grid on the map to assess whether this area is accessible. Combining the data in these two steps, we can form the dense map with drivable information. The entire system accelerated by GPU performs well in handling dynamic obstacles. For implementations, we demonstrate the effectiveness of our approach with mobile robot in a complex outdoor environment and have a detailed comparison with other methods.