Multi-Session Ground Texture SLAM in Low-Dynamic Environments

2026-05-19

AI Summary

This work addresses the challenge of insufficient robustness in loop closure detection for multi-session SLAM in environments where only ground textures are available and exhibit low dynamic changes—such as wear, weather, or seasonal variations. To enhance the reliability of loop closure detection across sessions, we propose, for the first time, the incorporation of Kullback-Leibler (KL) divergence as a similarity metric within ground texture-based SLAM. Experimental results demonstrate that this approach significantly improves trajectory estimation accuracy. Furthermore, we introduce the first publicly available dataset comprising multi-session ground texture imagery alongside high-precision ground-truth poses, establishing a new benchmark for visual SLAM research in low-dynamic environments.

Abstract

The simultaneous localization and mapping community has introduced a growing number of systems adapted for multi-session operations where the operational environment features low-dynamic changes that impact mapping, such as surface wear, weather phenomena, or seasonal change. These systems allow for lifelong operations by a robot within these environments. There is also growing interest in operations in environments where the unique ground texture is the only mapping feature available for use. These ground texture systems are not yet targeted for multi-session low-dynamic-change environments though. This work explores the impact of three different techniques on trajectory estimation accuracy in these multi-session low-dynamic ground texture environments. Of the three, the use of Kullback-Leibler Divergence, as a similarity score and a bias influencing loop closure confidence, is found to have the most success. We show an analysis of all three methods and a deeper exploration of the impact of Kullback-Leibler Divergence. We also introduce a dataset for use by the robotics community that contains multi-session images where the ground changes between sessions and also high-accuracy pose information for use in evaluation.

Problem

• multi-session
• ground texture
• low-dynamic environments
• SLAM
• trajectory estimation

Innovation

• ground texture SLAM
• multi-session
• low-dynamic environment
• Kullback-Leibler Divergence
• loop closure