Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GPS-VIO Fusion with Online Rotational Calibration

Junlin Song, Pedro J. Sanchez-Cuevas, Antoine Richard, Raj Thilak Rajan, Miguel Olivares-Mendez

TL;DR

A novel GPS-VIO system that is able to significantly benefit from the online calibration of the rotational extrinsic parameter between the GPS reference frame and the VIO reference frame is presented.

Abstract

Accurate global localization is crucial for autonomous navigation and planning. To this end, various GPS-aided Visual-Inertial Odometry (GPS-VIO) fusion algorithms are proposed in the literature. This paper presents a novel GPS-VIO system that is able to significantly benefit from the online calibration of the rotational extrinsic parameter between the GPS reference frame and the VIO reference frame. The behind reason is this parameter is observable. This paper provides novel proof through nonlinear observability analysis. We also evaluate the proposed algorithm extensively on diverse platforms, including flying UAV and driving vehicle. The experimental results support the observability analysis and show increased localization accuracy in comparison to state-of-the-art (SOTA) tightly-coupled algorithms.

GPS-VIO Fusion with Online Rotational Calibration

TL;DR

A novel GPS-VIO system that is able to significantly benefit from the online calibration of the rotational extrinsic parameter between the GPS reference frame and the VIO reference frame is presented.

Abstract

Accurate global localization is crucial for autonomous navigation and planning. To this end, various GPS-aided Visual-Inertial Odometry (GPS-VIO) fusion algorithms are proposed in the literature. This paper presents a novel GPS-VIO system that is able to significantly benefit from the online calibration of the rotational extrinsic parameter between the GPS reference frame and the VIO reference frame. The behind reason is this parameter is observable. This paper provides novel proof through nonlinear observability analysis. We also evaluate the proposed algorithm extensively on diverse platforms, including flying UAV and driving vehicle. The experimental results support the observability analysis and show increased localization accuracy in comparison to state-of-the-art (SOTA) tightly-coupled algorithms.
Paper Structure (17 sections, 25 equations, 4 figures, 2 tables)

This paper contains 17 sections, 25 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Problem Formulation
  4. Reference frames and Notation
  5. Classical MSCKF-based VIO structure
  6. GPS Measurement Update
  7. Observability Analysis
  8. Comments on Linear Observability Analysis
  9. Nonlinear Observability Analysis
  10. Zeroth-Order Lie Derivatives
  11. First-Order Lie Derivatives
  12. Observability analysis
  13. Results
  14. Validation of the Observability Analysis
  15. EuRoC dataset
  16. ...and 2 more sections

Figures (4)

  • Figure 1: Coordinate systems, similar as Fig. 1a in song2023gps.
  • Figure 2: Top: $\psi$ convergence over time respect to different initial guesses. Bottom: One standard deviation (1 $\sigma$) of $\psi$.
  • Figure 3: (a) $\psi$ convergence over time. (b) Horizontal view of aligned trajectory with different level of GPS noise.
  • Figure 4: (a) Top: $\left( {\psi - {\psi _0}} \right)$ convergence over time. Bottom: Calibration results of the time offset between GPS and IMU. (b) $\psi$ convergence over time respect to different initial values. The labels of legend represent different perturbation values.