Off-Road Path-Planning Based on Large Scale OpenStreetMap and DSM Data

In order to plan collision-free and shortest routes for vehicles in large-scale off-road scenarios, it is necessary to have a sufficient understanding and utilization of scene information. This paper proposes an off-road path planning method based on large scale OpenStreetMap (OSM) and Digital Surface Model (DSM). Our method comprehensively utilizes the terrain information from OpenStreetMap, slope information from DSM to analysis the traversability of the scene. We conduct experiments with a 2.5 km * 3.5 km area surrounding the Federal Institute of Technology in Lausanne, Switzerland. In order to more intuitively display the results of path planning in the terrain, we conducted a three-dimensional visualization of the results. The experimental results verify the effectiveness of the method and the collision-free and the shortest routes can be planned in the large scale off-road scenarios.