GeoFlow-SLAM: A Robust Tightly-Coupled RGBD-Inertial Fusion SLAM for Dynamic Legged Robotics

2025-03-18

AI Summary

This work addresses the robustness and accuracy bottlenecks of SLAM for legged robots operating in highly dynamic, low-texture environments. To this end, we propose a tightly coupled RGB-D–inertial SLAM framework. Methodologically: (1) we introduce the first joint optimization incorporating depth-map-based map constraints and GICP geometric constraints; (2) we design a multi-source robust pose initialization scheme integrating IMU/legged odometry, inter-frame PnP, and GICP; and (3) we propose a dual-stream optical flow method, GeoFlow, to enhance feature correspondence under rapid motion. The approach effectively mitigates feature matching failures during high-speed locomotion, initialization difficulties in texture-poor scenes, and long-term pose drift. Evaluated on both a newly collected legged-robot dataset and established public benchmarks, our method achieves state-of-the-art performance, demonstrating significant improvements in localization robustness and accuracy.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Search and Optimization: Sampling/Simulation-based Search
• Computer Vision: Low Level & Physics-based Vision

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

This paper presents GeoFlow-SLAM, a robust and effective Tightly-Coupled RGBD-inertial SLAM for legged robots operating in highly dynamic environments.By integrating geometric consistency, legged odometry constraints, and dual-stream optical flow (GeoFlow), our method addresses three critical challenges:feature matching and pose initialization failures during fast locomotion and visual feature scarcity in texture-less scenes.Specifically, in rapid motion scenarios, feature matching is notably enhanced by leveraging dual-stream optical flow, which combines prior map points and poses. Additionally, we propose a robust pose initialization method for fast locomotion and IMU error in legged robots, integrating IMU/Legged odometry, inter-frame Perspective-n-Point (PnP), and Generalized Iterative Closest Point (GICP). Furthermore, a novel optimization framework that tightly couples depth-to-map and GICP geometric constraints is first introduced to improve the robustness and accuracy in long-duration, visually texture-less environments. The proposed algorithms achieve state-of-the-art (SOTA) on collected legged robots and open-source datasets. To further promote research and development, the open-source datasets and code will be made publicly available at https://github.com/NSN-Hello/GeoFlow-SLAM

Problem

• Enhances feature matching in fast locomotion using dual-stream optical flow.
• Improves pose initialization for legged robots with IMU and odometry integration.
• Increases robustness in texture-less environments with depth-to-map and GICP constraints.

Innovation

• Dual-stream optical flow enhances feature matching.
• Robust pose initialization integrates IMU and odometry.
• Tightly-coupled depth-to-map and GICP improve accuracy.