B-TMS: Bayesian Traversable Terrain Modeling and Segmentation Across 3D LiDAR Scans and Maps for Enhanced Off-Road Navigation

TL;DR

The paper addresses robust traversable-terrain segmentation for off-road autonomous navigation by introducing B-TMS, a map-wide framework that couples Bayesian generalized kernel (BGK) based terrain completion with a tri-grid field (TGF). It combines an initial traversable search on a global TGF, BGK-based inference to predict unobserved terrain, and traversability-aware global model fitting to refine the terrain representation. Across SemanticKITTI, RELLIS-3D, and a newly collected extremely bumpy terrain dataset, B-TMS shows improved F1 and accuracy, reduced sensitivity to parameter changes, and effective handling of data distribution shifts and sunken/unobservable regions. This enables more reliable off-road navigation by predicting terrain in challenging areas and across map scales, with planned extensions to incorporate semantic cues and loop-closure for pose refinement.

Abstract

Recognizing traversable terrain from 3D point cloud data is critical, as it directly impacts the performance of autonomous navigation in off-road environments. However, existing segmentation algorithms often struggle with challenges related to changes in data distribution, environmental specificity, and sensor variations. Moreover, when encountering sunken areas, their performance is frequently compromised, and they may even fail to recognize them. To address these challenges, we introduce B-TMS, a novel approach that performs map-wise terrain modeling and segmentation by utilizing Bayesian generalized kernel (BGK) within the graph structure known as the tri-grid field (TGF). Our experiments encompass various data distributions, ranging from single scans to partial maps, utilizing both public datasets representing urban scenes and off-road environments, and our own dataset acquired from extremely bumpy terrains. Our results demonstrate notable contributions, particularly in terms of robustness to data distribution variations, adaptability to diverse environmental conditions, and resilience against the challenges associated with parameter changes.

Figures (4)

• Figure 1: Overview of map-wise Bayesian-based traversable terrain modeling and segmentation (B-TMS). B-TMS models and segments traversable terrain data in the given 3D point cloud map at once.
• Figure 2: Example scenes of our bumpy terrain dataset, which was acquired by traversing on the curved terrains of various heights and slopes.
• Figure 3: (L-R) The effect of the TGF resolution ($r^{\mathcal{T}}$), the inclination threshold ($\theta^{\mathcal{T}}$), the point-to-plane distance threshold ($\epsilon_{3}$) on ground segmentation for the 3D LiDAR scans and the partial maps from the SemanticKITTI dataset behley2019semantickitti, and RELLIS-3D dataset jiang2021rellis3d.
• Figure 4: Qualitative terrain segmentation results from a sequence of single scans of the RELLIS-3D dataset jiang2021rellis3d, comparing our previous work, TRAVEL oh22travel, with the proposed method, B-TMS. Green, red, blue, and black points represent true positives, false positives, false negatives, and true negatives, respectively. B-TMS, employing BGK-based terrain model completion, demonstrates robustness in narrow areas or rough off-road scenes where TRAVEL encounters difficulties, as highlighted by orange boxes. Although single scan data vary in distribution depending on the measured distance, a factor that can limit terrain modeling as highlighted by cyan boxes, B-TMS consistently shows robust results despite these challenges.