SCIENCE@home

Online Adaptive Traversability Estimation Through Interaction for Unstructured, Densely Vegetated Environments

Fabio Ruetz,Nicholas Lawrance,E. Hern'andez,Paulo V. K. Borges,Thierry Peynot

Published 2025 in IEEE Transactions on Field Robotics

ABSTRACT

Navigating densely vegetated environments poses significant challenges for autonomous ground vehicles (AGVs). Learning-based systems typically use prior and in situ data to predict terrain traversability, but often degrade in performance when encountering out-of-distribution elements caused by rapid environmental changes or novel conditions. This article presents a novel, lidar-only, online adaptive traversability estimation (TE) method that trains a model directly on the robot using self-labeled data collected through robot–environment interaction. The proposed approach utilizes a probabilistic 3-D voxel representation to integrate lidar measurements and robot experience over time, creating a salient environmental model. To ensure computational efficiency, a sparse graph-based representation is employed to update temporally evolving voxel distributions. Extensive experiments with an autonomous ground vehicle in natural terrain demonstrate that the system adapts to complex environments with as little as 8 min of operational data, achieving a Matthews correlation coefficient (MCC) score of 0.63 and enabling safe navigation in densely vegetated environments. This work examines different training strategies for voxel-based TE methods and offers recommendations for training strategies to improve adaptability. The proposed method is validated on a robotic platform with limited computational resources (25-W GPU), achieving an accuracy comparable with offline-trained models while maintaining reliable performance across varied environments.

PUBLICATION RECORD

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-27 of 27 references · Page 1 of 1

CITED BY

  • No citing papers are available for this paper.

Showing 0-0 of 0 citing papers · Page 1 of 1