The Kalman filter requires knowledge of the noise statistics; however, the noise covariances are generally unknown . Although this problem has a long history, reliable algorithms for their estimation are scant, and necessary and sufficient conditions for identifiability of the covariances are in dispute. We address both of these issues in this paper. We first present the necessary and sufficient condition for unknown noise covariance estimation; these conditions are related to the rank of a matrix involving the auto and cross-covariances of a weighted sum of innovations, where the weights are the coefficients of the minimal polynomial of the closed-loop system transition matrix of a stable, but not necessarily optimal, Kalman filter. We present an optimization criterion and a novel six-step approach based on a successive approximation, coupled with a gradient algorithm with adaptive step sizes, to estimate the steady-state Kalman filter gain, the unknown noise covariance matrices, as well as the state prediction (and updated) error covariance matrix. Our approach enforces the structural assumptions on unknown noise covariances and ensures symmetry and positive definiteness of the estimated covariance matrices. We provide several approaches to estimate the unknown measurement noise covariance $R$ via post-fit residuals , an approach not yet exploited in the literature. The validation of the proposed method on five different test cases from the literature demonstrates that the proposed method significantly outperforms previous state-of-the-art methods. It also offers a number of novel machine learning motivated approaches, such as sequential (one sample at a time) and mini-batch-based methods, to speed up the computations.
12 January 2018 in IEEE Transactions on Aerospace and Electronic Systems
This paper presents algorithms that can be used to track the state of kinematic point targets, moving in two dimensions, that are capable of making very sharp heading maneuvers over short periods of time. Ground vehicles with continuous wheel tracks moving in an open field may perform such maneuvers. These targets are capable of 60°/s turn rates, while measurements are received at only 1 Hz. We describe how an Interacting Multiple Model Estimator (IMM) should be designed to accommodate such maneuvers and we introduce the Non-Zero Mean, White Noise Turn-Rate model (WNTR) to quickly detect sharp turns by modeling a multi-mode process noise probability density function. Results show that including such modes in an IMM achieves significant improvement in accuracy and consistency as compared to conventional coordinated-turn IMM filters
20 June 2018 in IEEE Transactions on Aerospace and Electronic Systems
Tracking with bi-static sonar or radar measurements is challenging due to the fact that the measurements are nonlinear functions of the Cartesian state. The performance of existing approaches, including the Extended Kalman Filter (EKF) and sigma point Kalman filters such as the Unscented Kalman Filter (UKF) and Cubature Kalman Filter (CKF), may not be acceptable in terms of mean square error or tracker consistency (i.e. the tracker's state estimation error covariance is not statistically consistent with the actual estimation errors). This work generalizes the sigma point Kalman filter (SPKF) as a converted measurement SPKF. The resulting CMSPKF is demonstrated to have improved performance over the conventional SPKF when employed in a bi-static tracking scenario
The maximum likelihood-probabilistic data association (MLPDA) target tracking algorithm is effective in tracking very low observable targets. A key limitation of MLPDA is that it is restricted to tracking a single target. We derive and implement a multiple target version of MLPDA called Joint MLPDA (JMLPDA). While the JMLPDA implementation presented in this paper is focused on a two-target case, this algorithm is extensible to any number of targets. The MLPDA and JMLPDA algorithms are combined to form a multi-target MLPDA tracking algorithm. Performance of the JMLPDA and the multi-target MLPDA algorithms are compared to a probabilistic multi-hypothesis tracker (PMHT) for two crossing targets, focusing on track management/update. Simulation results show that under conditions of heavy clutter, the multi-target MLPDA outperforms PMHT in terms of reduced track errors and longer track life.
31 January 2023 in IEEE Transactions on Aerospace and Electronic Systems
When tracking maneuvering targets with a nearly constant velocity (NCV) Kalman filter with white noise acceleration, the selection of the process noise variance is complicated by the fact that the process noise errors are modeled as white Gaussian in the Kalman filter, while target maneuvers are deterministic or highly correlated in time. In recent years, the deterministic maneuver index was introduced for NCV Kalman filters and used to develop a relationship between the anticipated maximum acceleration of the target and the process noise variance that minimizes the maximum mean squared error (MaxMSE) in position. Lower bounds on the process noise variance that prevent the position MSE from exceeding the measurement noise variance were also expressed in terms of the maximum acceleration and deterministic maneuver index. In this paper, those results are summarized and codified for practical application by the target tracking community. The design methods for NCV Kalman filters with discrete white noise acceleration (DWNA) and continuous white noise acceleration (CWNA) are presented for sustained and brief maneuvers. The application of the design methods to radar tracking of maneuvering targets is also addressed. The effectiveness of the design methods is illustrated via Monte Carlo simulations.