Automatic Navigation Map Generation for Mobile Robots in Urban Environments

TL;DR

An algorithm to automatically generate navigation maps using a typical navigation-oriented sensor setup: a single top-mounted 3D LiDAR sensor is proposed and shown to be able to produce accurate maps featuring different terrain types, positive obstacles of different heights as well as negative obstacles.

Abstract

A fundamental prerequisite for safe and efficient navigation of mobile robots is the availability of reliable navigation maps upon which trajectories can be planned. With the increasing industrial interest in mobile robotics, especially in urban environments, the process of generating navigation maps has become of particular interest, being a labor intensive step of the deployment process. Automating this step is challenging and becomes even more arduous when the perception capabilities are limited by cost considerations. This paper proposes an algorithm to automatically generate navigation maps using a typical navigation-oriented sensor setup: a single top-mounted 3D LiDAR sensor. The proposed method is designed and validated with the urban environment as the main use case: it is shown to be able to produce accurate maps featuring different terrain types, positive obstacles of different heights as well as negative obstacles. The algorithm is applied to data collected in a typical urban environment with a wheeled inverted pendulum robot, showing its robustness against localization, perception and dynamic uncertainties. The generated map is validated against a human-made map.

Figures (17)

• Figure 1: Yape, the robot used to test the proposed algorithm, featuring the two independent wheels and the top mounted LiDAR.
• Figure 2: Block scheme of the full mapping algorithm, including the preprocessing steps.
• Figure 3: Schematic examples of explored area extraction. Ground regions where no measurement falls are classified as unexplored (to the right of the obstacle in \ref{['fig:drawing_explored_area_shade']}). Ground regions which are not connected to the regions traversed by the robot are classified as unexplored (far right in \ref{['fig:drawing_explored_area_shade']} and after the step in \ref{['fig:drawing_explored_area_step']})
• Figure 4: Schematic examples of positive obstacles extraction. Areas with vertical or too steep slopes are classified as obstacles (red underline).
• Figure 5: Simulated LiDAR "ring" on a planar surface illustrating the effect of surface roughness \ref{['fig:lidar_roughness']} and a point cloud color coded with the traversability index value \ref{['fig:piazza_leonardo_traversability_index_single_point_cloud']}. Red annotated lines show the boundary between gravel and grass, pink line between gravel and asphalt and green between sidewalk and cobblestone road.