A Joint Estimation Algorithm of Target State and System Error in Radar Network

Abstract

The invention provides a radar network target state and system error joint estimation algorithm, the steps are as follows, according to the airborne radar measurement principle, combined with the geographic location and attitude angle of the airborne platform, construct a system error, measurement noise and target state Each airborne radar measures the mathematical model; according to the target motion state, combined with the system error of each airborne radar, a mathematical model of the expanded target motion state containing process noise at each moment is constructed; according to the mathematical model constructed in the previous two steps Model, set the measurement noise, process noise, the initial value of the target motion state after dimension expansion and the initial value of the estimated error covariance matrix, and apply the CKF filtering method to realize the real-time simultaneous estimation of the target motion state and the error of each airborne radar system. The invention solves the problems existing in the traditional two-dimensional plane system error registration algorithm based on coordinate projection, such as data deformation, inability to estimate the pitch angle system error, and being unsuitable for long-distance system error registration.

Description

technical field [0001] The invention relates to the field of radar networking information fusion target tracking, in particular to the field of multi-platform cooperative tracking of airborne radar under the conditions of high measurement accuracy, high system dimension, and complex strong nonlinearity, and specifically relates to a radar networking target state Joint estimation algorithm with systematic error. Background technique [0002] The huge benefits that radar networking information fusion can bring have been recognized by countries all over the world. Although radar networking information fusion technology is still developing rapidly, due to the existence of radar detection target errors in the actual system, the guarantee of system real-time fusion effect has become a very difficult problem in the field of radar networking information fusion technology for a long time. Practical application shows that in multi-radar network tracking system, the existence of radar...

AI Technical Summary

Problems solved by technology

[0004] ① Although the spherical (polar) projection method uses high-order approximation to improve accuracy, since the earth is an ellipsoid rather than a sphere, it will introduce errors to the measurement during projection;

[0005] ②The spherical (polar) projection method will distort the data

Spherical (polar) conformal projection only guarantees that the azimuth angle will not be deformed, but it cannot guarantee that the slope distance will not be deformed, which will cause the system error to no longer be constant and related to measurement;

[0006] ③In the two-dimensional public coordinate system, only the systematic error of azimuth and radial distance can be estimated, but the systematic error of pitch angle cannot be estimated

However, EKF needs to calculate the Jacobian matrix. In the case of high-precision measurement, high-dimensional system, and complex and strong nonlinearity, it often cannot approximate the posterior probability density function of the system state very well, resulting in a greatly reduced filtering accuracy, and even causing filtering diverge

Although UKF avoids the shortcomings of EKF and improves the estimation accuracy to a certain extent, when UKF is applied to high-dimensional nonlinear systems, the weights of its deterministic sampling points are prone to negative values, which leads to the introduction of truncation errors in its moment integrals, which cannot Ensure numerical stability and semi-positive definiteness of the state covariance matrix, resulting in a decrease in filtering accuracy

Images