GSAT: Geometric Traversability Estimation using Self-supervised Learning with Anomaly Detection for Diverse Terrains
This work aims to improve the reliability of traversability estimation for autonomous navigation systems by addressing the limitations of human-defined thresholds and positive-only learning in self-supervised approaches.
This paper proposes GSAT, a self-supervised learning framework that estimates terrain traversability by constructing a positive hypersphere in latent space to classify traversable regions through anomaly detection. It avoids the need for explicit negative supervision or additional prototypes, and it employs joint learning of anomaly classification and traversability prediction.
Safe autonomous navigation requires reliable estimation of environmental traversability. Traditional methods have relied on semantic or geometry-based approaches with human-defined thresholds, but these methods often yield unreliable predictions due to the inherent subjectivity of human supervision. While self-supervised approaches enable robots to learn from their own experience, they still face a fundamental challenge: the positive-only learning problem. To address these limitations, recent studies have employed Positive-Unlabeled (PU) learning, where the core challenge is identifying positive samples without explicit negative supervision. In this work, we propose GSAT, which addresses these limitations by constructing a positive hypersphere in latent space to classify traversable regions through anomaly detection without requiring additional prototypes (e.g., unlabeled or negative). Furthermore, our approach employs joint learning of anomaly classification and traversability prediction to more efficiently utilize robot experience. We comprehensively evaluate the proposed framework through ablation studies, validation on heterogeneous real-world robotic platforms, and autonomous navigation demonstrations in simulation environments.