Iterated Invariant Extended Kalman Filter (IIEKF)

Sven Goffin,Axel Barrau,Silvère Bonnabel,Olivier Brüls,Pierre Sacré

2024-04-27

Abstract:In this paper, we introduce the Iterated Invariant Extended Kalman Filter (IIEKF), which is an invariant extended Kalman filter (IEKF) where the updated state in the light of the latest measurement is defined as a maximum a posteriori (MAP) estimate. Under some compatibility requirements on the output map, we prove strong mathematical guarantees which echo those of the Kalman filter in the linear case. We apply the technique to two problems: solving a system of equations on a Lie group, and a problem of engineering interest, namely ego-localization of the hook of a crane. The latter serves as a benchmarking example, where the IIEKF favorably compares to other filters.

Systems and Control

What problem does this paper attempt to address?

The paper attempts to address the problem of improving the performance of the Extended Kalman Filter (EKF) in nonlinear systems, especially when faced with noiseless measurements. Specifically, the authors propose a new method—the Iterative Invariant Extended Kalman Filter (IIEKF), aimed at solving the following issues: 1. **Accurate estimation in nonlinear systems**: Traditional EKF relies on linearization when dealing with nonlinear dynamic systems, which can lead to estimation errors. IIEKF improves the measurement update step by employing the Gauss-Newton method, thereby enhancing estimation accuracy. 2. **Preservation of geometric properties**: EKF typically ignores the geometric properties of the state space (such as Lie groups). IIEKF preserves these geometric properties by defining the Maximum A Posteriori (MAP) estimation in the Lie algebra. 3. **Guarantees under noiseless measurements**: In the case of noiseless measurements, IIEKF can recover some of the strong mathematical guarantees of the linear Kalman filter, namely that the updated state lies within the observation set and that the covariance matrix correctly encodes the information of reduced uncertainty. To validate these theoretical results, the authors apply IIEKF to two practical problems: solving systems of equations on Lie groups and the self-localization problem of a crane hook. Experimental results show that IIEKF performs excellently in these applications, outperforming other filtering methods.

Iterated Invariant Extended Kalman Filter (IIEKF)

Maneuvering Target Tracking with an Adaptive UKF Filter

Invariant Kalman Filtering with Noise-Free Pseudo-Measurements

Inverse Extended Kalman Filter -- Part I: Fundamentals

A Quick Guide for the Iterated Extended Kalman Filter on Manifolds

Invariant extended Kalman filter on matrix Lie groups

Inverse Extended Kalman Filter -- Part II: Highly Non-Linear and Uncertain Systems

Iterated Kalman Methodology For Inverse Problems

Notes on Kalman Filter (KF, EKF, ESKF, IEKF, IESKF)

Discrete‐time invariant extended Kalman filter on matrix Lie groups

Toward Consistent and Efficient Map-Based Visual-Inertial Localization: Theory Framework and Filter Design

Contact-Aided Invariant Extended Kalman Filtering for Robot State Estimation

Outlier-Robust Iterative Extended Kalman Filtering.

PIEKF-VIWO: Visual-Inertial-Wheel Odometry using Partial Invariant Extended Kalman Filter

Dynamically Iterated Filters: A unified framework for improved iterated filtering and smoothing

Iterated Filters for Nonlinear Transition Models

Generalized Iterated Kalman Filter and Its Performance Evaluation

Equivariant Filter Design for Inertial Navigation Systems with Input Measurement Biases

Projectile trajectory estimation: performance analysis of an Extended Kalman Filter and an Imperfect Invariant Extended Kalman Filter

Stirling's Interpolation Filtering and Extended Kalman Filtering for Orientation Estimation

Variational Bayesian Iteration-Based Invariant Kalman Filter for Attitude Estimation on Matrix Lie Groups