A Robust and Efficient Visual-Inertial Initialization with Probabilistic Normal Epipolar Constraint

2024-10-25

arXiv.org

AI Summary

To address the poor robustness and low efficiency of Visual-Inertial Odometry (VIO) initialization under fast motion and degenerate conditions, this paper proposes a rotation-translation decoupled initialization framework. Our key contributions are: (1) the first gyroscope bias estimation algorithm based on probabilistic epipolar constraints in the normal plane; (2) a joint gravity-scale optimization module with post-initialization refinement to enhance stability under degeneracy; and (3) tight IMU-visual coupling with uncertainty-aware parameter estimation. Evaluated on the EuRoC dataset, our method reduces gyroscope bias, rotation, and gravity estimation errors by 16%, 4%, and 29%, respectively. On TUM-VI, gravity and scale errors are reduced by 14.2% and 5.7%. These improvements significantly enhance initialization accuracy and robustness across challenging motion and environmental conditions.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Motion & Tracking
• Constraint Satisfaction and Optimization: Other Foundations of Constraint Satisfaction

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

Accurate and robust initialization is essential for Visual-Inertial Odometry (VIO), as poor initialization can severely degrade pose accuracy. During initialization, it is crucial to estimate parameters such as accelerometer bias, gyroscope bias, initial velocity, gravity, etc. Most existing VIO initialization methods adopt Structure from Motion (SfM) to solve for gyroscope bias. However, SfM is not stable and efficient enough in fast-motion or degenerate scenes. To overcome these limitations, we extended the rotation-translation-decoupled framework by adding new uncertainty parameters and optimization modules. First, we adopt a gyroscope bias estimator that incorporates probabilistic normal epipolar constraints. Second, we fuse IMU and visual measurements to solve for velocity, gravity, and scale efficiently. Finally, we design an additional refinement module that effectively reduces gravity and scale errors. Extensive EuRoC dataset tests show that our method reduces gyroscope bias and rotation errors by 16% and 4% on average, and gravity error by 29% on average. On the TUM dataset, our method reduces the gravity error and scale error by 14.2% and 5.7% on average respectively. The source code is available at https://github.com/MUCS714/DRT-PNEC.git

Problem

• Improves Visual-Inertial Odometry initialization accuracy.
• Estimates biases and motion parameters robustly.
• Reduces errors in gyroscope and gravity measurements.

Innovation

• Probabilistic normal epipolar constraints
• IMU and visual measurements fusion
• Refinement module reduces gravity errors