GeVI-SLAM: Gravity-Enhanced Stereo Visua Inertial SLAM for Underwater Robots

2025-10-28

AI Summary

Underwater visual-inertial SLAM (VI-SLAM) suffers from severe visual degradation and insufficient IMU excitation, leading to unstable pose estimation. Method: This paper proposes a gravity-enhanced stereo visual-inertial SLAM system. Key components include: (1) leveraging precise gravity direction to decouple pitch and roll angles, enabling unbiased 4-DOF PnP formulation and solution for consistent pose estimation; (2) an adaptive gravity-prior weighting mechanism to improve robustness in dynamic underwater environments; and (3) integration of stereo direct depth estimation, gravity-aided IMU initialization, minimal three-point PnP, RANSAC-based outlier rejection, and joint IMU covariance estimation. Results: Evaluated on both synthetic and real underwater datasets, the method achieves superior localization accuracy and stability over state-of-the-art approaches—particularly under low-acceleration and weak-excitation conditions.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Biometrics, Face, Gesture & Pose
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Accurate visual inertial simultaneous localization and mapping (VI SLAM) for underwater robots remains a significant challenge due to frequent visual degeneracy and insufficient inertial measurement unit (IMU) motion excitation. In this paper, we present GeVI-SLAM, a gravity-enhanced stereo VI SLAM system designed to address these issues. By leveraging the stereo camera's direct depth estimation ability, we eliminate the need to estimate scale during IMU initialization, enabling stable operation even under low acceleration dynamics. With precise gravity initialization, we decouple the pitch and roll from the pose estimation and solve a 4 degrees of freedom (DOF) Perspective-n-Point (PnP) problem for pose tracking. This allows the use of a minimal 3-point solver, which significantly reduces computational time to reject outliers within a Random Sample Consensus framework. We further propose a bias-eliminated 4-DOF PnP estimator with provable consistency, ensuring the relative pose converges to the true value as the feature number increases. To handle dynamic motion, we refine the full 6-DOF pose while jointly estimating the IMU covariance, enabling adaptive weighting of the gravity prior. Extensive experiments on simulated and real-world data demonstrate that GeVI-SLAM achieves higher accuracy and greater stability compared to state-of-the-art methods.

Problem

• Addresses visual degeneracy and insufficient IMU excitation in underwater SLAM
• Enables stable scale estimation through stereo depth and gravity initialization
• Reduces computational complexity via 4-DOF PnP with minimal point solver

Innovation

• Uses stereo camera for direct depth estimation
• Solves 4-DOF PnP problem with minimal solver
• Proposes bias-eliminated estimator with provable consistency