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

TL;DR

This paper introduces an online, lidar-only, self-supervised traversability estimation method for autonomous robots that adapts in real-time to complex, densely vegetated terrains, enabling safe navigation with minimal data and limited computational resources.

Contribution

The paper presents a novel online adaptive traversability estimation approach using a probabilistic voxel model and sparse graph updates, improving real-time adaptability in unstructured environments.

Findings

Achieves MCC score of 0.63 with 8 minutes of data

Enables safe navigation in dense vegetation

Operates efficiently on limited hardware (25W GPU)

Abstract

Navigating densely vegetated environments poses significant challenges for autonomous ground vehicles. 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 paper presents a novel, lidar-only, online adaptive traversability estimation (TE) method that trains a model directly on the robot using self-supervised data collected through robot-environment interaction. The proposed approach utilises a probabilistic 3D voxel representation to integrate lidar measurements and robot experience, creating a salient environmental model. To ensure computational efficiency, a sparse graph-based representation is employed to update temporarily evolving voxel distributions. Extensive experiments with an unmanned ground vehicle in natural terrain demonstrate that the system adapts to complex environments with as little as 8 minutes 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 (25W GPU), achieving accuracy comparable to offline-trained models while maintaining reliable performance across varied environments.