Exactly Decoupled Kalman Filtering for Multitarget State Estimation with Sensor Bias

TL;DR

This paper introduces an exactly decoupled Kalman filtering method for multitarget state estimation with sensor biases, offering a computationally efficient alternative to the classical augmented approach while maintaining equivalent estimation accuracy.

Contribution

The paper proposes a novel decoupled Kalman filtering algorithm that simplifies multitarget estimation with sensor biases, proven to be equivalent to the traditional augmented Kalman filter.

Findings

Decoupled Kalman filter achieves the same results as the augmented filter.

The method is validated with numerical simulations.

Field experiments confirm practical effectiveness.

Abstract

The problem of multisensor multitarget state estimation in the presence of constant but unknown sensor biases is investigated. The classical approach to this problem is to augment the state vector to include the states of all the targets and the sensor biases, and then implement an augmented state Kalman filter (ASKF). In this paper, we propose a novel decoupled Kalman filtering algorithm. The decoupled Kalman filtering first processes each target in a separate branch, namely the single-target Kalman filtering branch, where the single-target states and the sensor biases are estimated. Then the bias estimate is refined by fusing the former bias estimates across all the single-target Kalman filtering branches. Finally, the refined bias estimate is fed back to each single-target Kalman filtering branch to improve the target state estimation. We prove that the proposed decoupled Kalman filtering is exactly equivalent to the ASKF in terms of the estimation results under a usual initial condition. The equivalence is also confirmed via the numerical example. Moreover, we further validate the proposed algorithm using the field experimental data of a multistatic passive radar.