Real-time Spatial-temporal Traversability Assessment via Feature-based Sparse Gaussian Process

2025-03-06

AI Summary

To address the challenge of real-time terrain traversability assessment for ground mobile robots in unstructured outdoor environments, this paper proposes a spatiotemporal traversability reasoning method that jointly incorporates geometric perception and uncertainty modeling. We extract geometric features—including point cloud curvature, gradient, and elevation—to construct a feature-based Sparse Gaussian Process (SGP) model. A novel spatial-temporal Bayesian Gaussian Kernel (BGK) is designed to enable dynamic, incremental inference of traversability scores while explicitly characterizing environmental uncertainty. The method integrates GPU-accelerated point cloud processing with an autonomous navigation framework. In simulation, it surpasses state-of-the-art approaches in both accuracy and computational efficiency. Real-world validation on a differential-drive robot demonstrates robust 10 Hz real-time navigation in challenging野外 scenarios. The implementation is open-sourced.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Intelligent Robots: Localization, Mapping, and Navigation
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Terrain analysis is critical for the practical application of ground mobile robots in real-world tasks, especially in outdoor unstructured environments. In this paper, we propose a novel spatial-temporal traversability assessment method, which aims to enable autonomous robots to effectively navigate through complex terrains. Our approach utilizes sparse Gaussian processes (SGP) to extract geometric features (curvature, gradient, elevation, etc.) directly from point cloud scans. These features are then used to construct a high-resolution local traversability map. Then, we design a spatial-temporal Bayesian Gaussian kernel (BGK) inference method to dynamically evaluate traversability scores, integrating historical and real-time data while considering factors such as slope, flatness, gradient, and uncertainty metrics. GPU acceleration is applied in the feature extraction step, and the system achieves real-time performance. Extensive simulation experiments across diverse terrain scenarios demonstrate that our method outperforms SOTA approaches in both accuracy and computational efficiency. Additionally, we develop an autonomous navigation framework integrated with the traversability map and validate it with a differential driven vehicle in complex outdoor environments. Our code will be open-source for further research and development by the community, https://github.com/ZJU-FAST-Lab/FSGP_BGK.

Problem

• Real-time terrain traversability assessment for mobile robots.
• Dynamic evaluation using spatial-temporal Bayesian Gaussian kernel.
• Integration of historical and real-time data for navigation.

Innovation

• Sparse Gaussian Processes for feature extraction
• Spatial-temporal Bayesian Gaussian kernel inference
• GPU acceleration for real-time performance