SyNeT: Synthetic Negatives for Traversability Learning

2026-01-31

AI Summary

Existing self-supervised traversability learning methods struggle to effectively identify diverse non-traversable regions due to the absence of explicit negative samples. This work proposes, for the first time, a synthetic negative sample generation mechanism that constructs seemingly plausible but actually non-traversable regions to enhance the model’s discriminative capability. The approach is compatible with both positive-unlabeled (PU) and positive-negative (PN) learning frameworks and introduces an object-oriented false positive rate (FPR) metric to quantify model consistency in identifying non-traversable areas without requiring additional human annotations. Experiments demonstrate that the proposed method significantly improves model robustness and cross-environment generalization across multiple public and self-collected datasets. Code and demonstration videos are publicly released.

Technology Category

• Search and Optimization: Learning to Search
• Intelligent Robots: Localization, Mapping, and Navigation
• Natural Language Processing: Safety and Robustness

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Reliable traversability estimation is crucial for autonomous robots to navigate complex outdoor environments safely. Existing self-supervised learning frameworks primarily rely on positive and unlabeled data; however, the lack of explicit negative data remains a critical limitation, hindering the model's ability to accurately identify diverse non-traversable regions. To address this issue, we introduce a method to explicitly construct synthetic negatives, representing plausible but non-traversable, and integrate them into vision-based traversability learning. Our approach is formulated as a training strategy that can be seamlessly integrated into both Positive-Unlabeled (PU) and Positive-Negative (PN) frameworks without modifying inference architectures. Complementing standard pixel-wise metrics, we introduce an object-centric FPR evaluation approach that analyzes predictions in regions where synthetic negatives are inserted. This evaluation provides an indirect measure of the model's ability to consistently identify non-traversable regions without additional manual labeling. Extensive experiments on both public and self-collected datasets demonstrate that our approach significantly enhances robustness and generalization across diverse environments. The source code and demonstration videos will be publicly available.

Problem

• traversability estimation
• synthetic negatives
• self-supervised learning
• non-traversable regions
• autonomous navigation

Innovation

• synthetic negatives
• traversability learning
• self-supervised learning
• object-centric evaluation
• positive-unlabeled learning