A Multi-view Landmark Representation Approach with Application to GNSS-Visual-Inertial Odometry

2025-08-07

AI Summary

In vision-aided GNSS-inertial odometry, the iterated extended Kalman filter (IEKF) suffers from excessive computational burden due to joint optimization of camera poses and landmark points. Method: This paper proposes a multi-view tightly coupled pose estimation algorithm. Its core innovation is a pose-driven visual measurement model: landmarks are explicitly parameterized in terms of multi-camera poses and observations, ensuring that the null space of the measurement Jacobian is strictly independent of the estimated poses. This preserves tight coupling accuracy while eliminating landmark states from the filter’s state dimension and computational complexity. A lightweight feature management strategy is further introduced to enable pose-only efficient updates. Results: Simulations and real-world experiments demonstrate that the proposed method achieves ~40% higher computational efficiency and 15–25% improved positioning accuracy over conventional joint-optimization approaches, significantly enhancing real-time performance and robustness.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Invariant Extended Kalman Filter (IEKF) has been a significant technique in vision-aided sensor fusion. However, it usually suffers from high computational burden when jointly optimizing camera poses and the landmarks. To improve its efficiency and applicability for multi-sensor fusion, we present a multi-view pose-only estimation approach with its application to GNSS-Visual-Inertial Odometry (GVIO) in this paper. Our main contribution is deriving a visual measurement model which directly associates landmark representation with multiple camera poses and observations. Such a pose-only measurement is proven to be tightly-coupled between landmarks and poses, and maintain a perfect null space that is independent of estimated poses. Finally, we apply the proposed approach to a filter based GVIO with a novel feature management strategy. Both simulation tests and real-world experiments are conducted to demonstrate the superiority of the proposed method in terms of efficiency and accuracy.

Problem

• Reducing computational burden in landmark-pose joint optimization
• Improving efficiency for multi-sensor fusion in odometry
• Enhancing accuracy in GNSS-Visual-Inertial Odometry systems

Innovation

• Multi-view pose-only estimation approach
• Visual measurement model for landmarks
• Filter-based GVIO with feature management