1992Unpublished venueRequires access

Identification of measurement and maneuver noise in α-β target tracking systems

Thomas A. Chmielewski, P. Kalata

Open publisher page 0 citations

Abstract

This paper addresses the problem of estimating the variances of maneuver and masurement noise in α-β Target Tracking Systems. A switching modeling process augments the original tracking states with the maneuver and measurement noise states which leads to noise tracks generated by the α-β target tracker. It is shown that noise variance estimates can be obtained by using noise track estimate variances and the corresponding error covariance terms. Since the optimal state and its error are orthogonal, it is easily shown that the noise track error covariance and the estimated noise track variance may be algebraically combined to provide estimates of the unkmown variance of the maneuver and measurement input noise processes. Closed form, steady state performances, i.e., the noise states error covariance terms, as well as solutions for the estimated state variances for an α-β tracker are obtained. Using these solutions, the concept of Noise Variance Observability is introduced. For the α-β tracker, it is possible to show the conditions under which the time sequence measurement or maneuver noise states can be directly estimated.

About this research paper

What this paper is about

This paper addresses the problem of estimating the variances of maneuver and masurement noise in α-β Target Tracking Systems. A switching modeling process augments the original tracking states with the maneuver and measurement noise states which leads to noise tracks generated by the α-β target tracker. It is shown that noise variance estimates can be obtained by using noise track estimate variances and the corresponding error covariance terms. Since the optimal state and its error are orthogonal, it is easily shown that the noise track error covariance and the estimated noise track variance may be algebraically combined to provide estimates of the unkmown variance of the maneuver and measurement input noise processes. Closed form, steady state performances, i.e., the noise states error covariance terms, as well as solutions for the estimated state variances for an α-β tracker are obtained. Using these solutions, the concept of Noise Variance Observability is introduced. For the α-β tracker, it is possible to show the conditions under which the time sequence measurement or maneuver noise states can be directly estimated.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

This paper addresses the problem of estimating the variances of maneuver and masurement noise in α-β Target Tracking Systems. A switching modeling process augments the original tracking states with the maneuver and measurement noise states which leads to noise tracks generated by the α-β target tracker. It is shown that noise variance estimates can be obtained by using noise track estimate variances and the corresponding error covariance terms. Since the optimal state and its error are orthogonal, it is easily shown that the noise track error covariance and the estimated noise track variance may be algebraically combined to provide estimates of the unkmown variance of the maneuver and measurement input noise processes. Closed form, steady state performances, i.e., the noise states error covariance terms, as well as solutions for the estimated state variances for an α-β tracker are obtained. Using these solutions, the concept of Noise Variance Observability is introduced. For the α-β tracker, it is possible to show the conditions under which the time sequence measurement or maneuver noise states can be directly estimated.

Key concepts: Noise (video), Observability, Covariance, Noise measurement, Control theory (sociology), Tracking (education), Gaussian noise, Variance (accounting)

Related papers

Back to paper searchBrowse research topicsOriginal source
Identification of measurement and maneuver noise in α-β target tracking systems — Research Paper | ScholarLens