Voxel Map to Occupancy Map Conversion Using Free Space Projection for Efficient Map Representation for Aerial and Ground Robots

TL;DR

This paper presents a novel method to convert 3D voxel maps into 2D occupancy maps for both UAVs and UGVs by projecting free space and enriching the 2D representation with height and slope information. The approach yields two dedicated 2D occupancy maps, one tailored for aerial navigation and another for ground robots, plus a height map and a slope map derived via least-squares, enabling reliable 2D path planning with 3D path lifting. The method is implemented in ROS using UFOMap, validated in cave and outdoor environments, and shown to dramatically reduce map size while supporting real-time updates during exploration. Such a framework facilitates cross-domain planning, faster inter-agent map sharing, and the reuse of 2D-planner techniques in 3D environments, with practical implications for bandwidth-limited multi-robot systems.

Abstract

This article introduces a novel method for converting 3D voxel maps, commonly utilized by robots for localization and navigation, into 2D occupancy maps for both unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs). The generated 2D maps can be used for more efficient global navigation for both UAVs and UGVs, in enabling algorithms developed for 2D maps to be useful in 3D applications, and allowing for faster transfer of maps between multiple agents in bandwidth-limited scenarios. The proposed method uses the free space representation in the UFOMap mapping solution to generate 2D occupancy maps. During the 3D to 2D map conversion, the method conducts safety checks and eliminates free spaces in the map with dimensions (in the height axis) lower than the robot's safety margins. This ensures that an aerial or ground robot can navigate safely, relying primarily on the 2D map generated by the method. Additionally, the method extracts the height of navigable free space and a local estimate of the slope of the floor from the 3D voxel map. The height data is utilized in converting paths generated using the 2D map into paths in 3D space for both UAVs and UGVs. The slope data identifies areas too steep for a ground robot to traverse, marking them as occupied, thus enabling a more accurate representation of the terrain for ground robots. The effectiveness of the proposed method in enabling computationally efficient navigation for both aerial and ground robots is validated in two different environments, over both static maps and in online implementation in an exploration mission. The methods proposed within this article have been implemented in the popular robotics framework ROS and are open-sourced. The code is available at: https://github.com/LTU-RAI/Map-Conversion-3D-Voxel-Map-to-2D-Occupancy-Map.

Figures (5)

• Figure 1: An overview of the proposed methodology for 3D Voxel map to 2D map conversion, along with path planning for the UAVs and UGVs, and 2D to 3D path conversion.
• Figure 2: The computation time to update the 2D map generated from the voxel map under two exploration missions. The scenarios marked with '*' are the time required by the method, excluding the time needed to read the UFOmap.
• Figure 3: The computation time to update the 2D map generated from the voxel map over the total mission time, under two exploration missions. The scenarios marked with '*' are the time required by the method, excluding the time needed to read the UFOmap.
• Figure 4: 2D maps and paths generated by the proposed methods for UAV and UGV in the cave environment. The results from the 2D path to 3D path conversion are shown in subfigures e) and f).
• Figure 5: 2D maps and paths generated by the proposed methods for UAV and UGV in the cave environment. The results from the 2D path to 3D path conversion are shown in subfigures e) and f).