Traversability analysis for autonomous driving in complex environment: A LiDAR‐based terrain modeling approach

Abstract

For autonomous driving, traversability analysis is one of the most basic and essential tasks. In this paper, we propose a novel LiDAR‐based terrain modeling approach, which could output stable, complete, and accurate terrain models and traversability analysis results. As terrain is an inherent property of the environment that does not change with different view angles, our approach adopts a multiframe information fusion strategy for terrain modeling. Specifically, a normal distributions transform mapping approach is adopted to accurately model the terrain by fusing information from consecutive LiDAR frames. Then the spatial‐temporal Bayesian generalized kernel inference and bilateral filtering are utilized to promote the stability and completeness of the results while simultaneously retaining the sharp terrain edges. Based on the terrain modeling results, the traversability of each region is obtained by performing geometric connectivity analysis between neighboring terrain regions. Experimental results show that the proposed method could run in real‐time and outperforms state‐of‐the‐art approaches.

A Review on Traversability Risk Assessments for Autonomous Ground Vehicles: Methods and Metrics

“…The DEM is the most widely used technique for environment modeling and traversability analysis. Therefore, several modifications have been proposed to improve its efficiency such as in [ 46 ], who considered the elevation to follow a Gaussian distribution, and the stores for each cell are not only the mean elevation, but also the variance. Although elevation maps offer compact representation, they may not adequately depict multi-level structures or even vertical structures (bridges, for instance).…”