A Unified Filter for Fusion of Multiple Inertial Measurement Units

Yaakov Libero,Itzik Klein

DOI: https://doi.org/10.48550/arXiv.2208.14524

IF: 4.729

2022-08-30

Signal Processing

Abstract:Navigation plays a vital role in the ability of autonomous surface and underwater platforms to complete their tasks. Most navigation systems apply a fusion between inertial sensors and other external sensors, such as global navigation satellite systems, when available, or a Doppler velocity log. In recent years, there has been increased interest in using multiple inertial measurement units to improve navigation accuracy and robustness. State of the art examples include the virtual inertial measurement unit (VIMU) and the federated extended Kalman filter (FEKF). However, each of those approaches has its drawbacks. The VIMU does not improve the sensor biases, which constitute a significant source of error, especially in low-cost inertial sensors. While the FEKF does improve accuracy, it models uncertainty propagation empirically. If not modeled correctly, this can cause the global solution to diverge. To cope with those shortcomings, we propose and derive a new filter structure for multiple inertial sensors data fusion: the unified extended Kalman filter (UEKF). In addition, to cope with the multiple equal bias variance estimation problem we offer the bias variance redistribution algorithm. Our filter design enables bias estimation for each of the inertial sensors in the system, improving its accuracy and allowing the use of a varying number of inertial sensors. We show that our UEKF performs better than other state of the art, multiple inertial sensor filters using real data recorded during sea experiments.