Distilling Global Traversability Priors for Image-based Affordance Prediction in Off-road Environments

2026-07-20

AI Summary

This work addresses the challenge of short-sighted navigation in unstructured off-road environments, where limited perception hinders autonomous systems. To overcome this, the authors propose a traversability prediction method that integrates prior knowledge from satellite imagery. By leveraging satellite-derived global feasible paths as supervisory signals, the approach distills long-range traversability priors into a monocular vision model for the first time, enabling end-to-end prediction of distant traversable regions. This strategy substantially reduces reliance on human demonstration data, achieving performance gains exceeding 10% on multiple offline benchmarks. Furthermore, real-world off-road experiments demonstrate a significant reduction in human interventions, thereby enhancing the system’s capability for sustained autonomous navigation over extended missions.

Abstract

Standard methods for autonomous navigation in unstructured terrain are prone to myopic behaviors in long-horizon scenarios. The use of metric maps built from LiDAR or cameras provides necessary local geometry and semantic information but is strictly limited by depth sensing range. By discarding data beyond the mapping horizon robots suffer from suboptimal, short-sighted decisions. To recover this lost information, we focus on extracting long-range traversability-aware frontiers directly from first-person-view (FPV) images. By leveraging satellite imagery, we compute the set of feasible navigation paths for a dataset of image/pose pairs and use them to supervise our network, reducing the need for extensive human demonstration data. We demonstrate that this approach improves performance in long-range off-road navigation over existing methods by more than 10% in various offline benchmarks and reduces the number of human interventions incurred in a set of real-world experiments. More details can be found at https://theairlab.org/ss_frontiers_iros .

Problem

• off-road navigation
• traversability prediction
• long-horizon planning
• image-based affordance
• frontier detection

Innovation

• traversability prediction
• image-based affordance
• satellite supervision
• off-road navigation
• frontier detection