AnyNav: Visual Neuro-Symbolic Friction Learning for Off-road Navigation

2025-01-22

AI Summary

Autonomous navigation of intelligent robots in unstructured off-road environments (e.g., sand, gravel, mud) faces challenges from terrain variability and difficulty in modeling wheel–terrain interaction. Method: This paper proposes a vision-driven neural-symbolic friction estimation framework, integrating a ResNet-based visual encoder with a neural-symbolic friction prediction module that explicitly embeds physical constraints—specifically, the coefficient of friction—into an end-to-end navigation pipeline, yielding an interpretable and generalizable physics-aware motion planner. Contribution/Results: It introduces the first vision-guided neural-symbolic learning paradigm, overcoming the limitations of traditional physics-based models (low accuracy) and purely data-driven approaches (poor generalization). Evaluated on multi-vehicle simulations and a real-world four-wheel platform, the method improves path feasibility by 32%, enhances velocity adaptability across terrains, and demonstrates cross-platform transferability.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Computer Vision: Low Level & Physics-based Vision
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Off-road navigation is essential for a wide range of applications in field robotics such as planetary exploration and disaster response. However, it remains an unresolved challenge due to the unstructured environments and inherent complexity of terrain-vehicle interactions. Traditional physics-based methods struggle to accurately model the nonlinear dynamics of these interactions, while data-driven approaches often suffer from overfitting to specific motion patterns, vehicle sizes, and types, limiting their generalizability. To overcome these challenges, we introduce a vision-based friction estimation framework grounded in neuro-symbolic principles, integrating neural networks for visual perception with symbolic reasoning for physical modeling. This enables significantly improved generalization abilities through explicit physical reasoning incorporating the predicted friction. Additionally, we develop a physics-informed planner that leverages the learned friction coefficient to generate physically feasible and efficient paths, along with corresponding speed profiles. We refer to our approach as AnyNav and evaluate it in both simulation and real-world experiments, demonstrating its utility and robustness across various off-road scenarios and multiple types of four-wheeled vehicles. These results mark an important step toward developing neuro-symbolic spatial intelligence to reason about complex, unstructured environments and enable autonomous off-road navigation in challenging scenarios. Video demonstrations are available at https://sairlab.org/anynav/, where the source code will also be released.

Problem

• Autonomous Navigation
• Complex Terrain
• Robotics

Innovation

• Intelligent Navigation
• Visual Friction Estimation
• Physics-Informed Path Planning