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

TL;DR

This work tackles real-time traversability estimation for autonomous robots in unstructured outdoor terrains by addressing the limitations of existing single-frame SGP methods and their lack of historical data integration. It proposes a feature-based sparse Gaussian Process pipeline with GPU-accelerated curvature and gradient feature extraction, PCA decorrelation, and inducing points to efficiently model local terrain height and uncertainty. To incorporate historical observations and improve temporal consistency, a spatial-temporal Bayesian Gaussian Kernel (BGK) fusion is introduced, followed by Gaussian kernel smoothing to yield a robust traversability map. The framework is integrated into an autonomous navigation stack and validated through simulations and real-world experiments, with open-source code to foster reproducibility and further research.

Abstract

Terrain analysis is critical for the practical ap- plication of ground mobile robots in real-world tasks, espe- cially 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 outper- forms 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 envi- ronments. Our code will be open-source for further research and development by the community, https://github.com/ZJU-FAST-Lab/FSGP_BGK.

Figures (3)

• Figure 1: (a) Simulation results and the corresponding traversability map; (b) Real-world testing environment, where the yellow line indicates areas that are easy to traverse.
• Figure 2: Overview of the proposed terrain traversability mapping and navigation framework. From left to right, localization and LiDAR point cloud data are fed into the Feature Extraction module, where curvature and gradient features are computed for each point. These features are then processed by a Sparse Gaussian Process (SGP) model with induced points, yielding local height predictions, variance, and gradient information. Next, a spatial-temporal Bayesian Gaussian Kernel (BGK) fusion step integrates these predictions with historical maps to produce a refined traversability cost map. Finally, we employ A* for trajectory search, MINCOb7 for trajectory optimization, and a controller for trajectory tracking, thereby generating the necessary control commands for the autonomous vehicle to navigate uneven terrain safely and efficiently.
• Figure 3: Feature-based SGP results in an uneven environment. From left to right: We begin with the point cloud of the uneven terrain and extract SGP inducing feature points. The SGP then predicts local slope, curvature, gradient, and uncertainty layers, which are finally integrated into a traversability map.