CN-121385866-B - Space target multi-radar handover tracking method based on prediction error covariance transfer

CN121385866BCN 121385866 BCN121385866 BCN 121385866BCN-121385866-B

Abstract

The application belongs to the technical field of radar target tracking, and relates to a spatial target multi-radar handover tracking method based on prediction error covariance transfer, which comprises the following steps that a first radar carries out filtering tracking on a spatial target to obtain a first position, a first speed and first covariance information at the final observation moment; transforming to a geocentric inertial coordinate system, constructing initial forecasting state and initial forecasting error covariance information, forecasting and extrapolation to obtain the initial forecasting state and the initial forecasting error covariance information at the future moment, determining the time when a target enters a second radar coverage airspace, planning an interception position and an interception area by the second radar and tracking the target to obtain candidate state and candidate covariance information, constructing test statistics, and determining cross-radar association results by combining multi-moment association quality evaluation to finish space target handover tracking. The technical scheme of the application can improve the track consistency, error consistency and correlation reliability in space target handover.

Inventors

XIU JIANJUAN
TAN SHUNCHENG
SUN WEIWEI

Assignees

中国人民解放军海军航空大学

Dates

Publication Date: 20260505
Application Date: 20251224

Claims (10)

1. The spatial target multi-radar handover tracking method based on prediction error covariance transfer is characterized by comprising the following steps of: Filtering and tracking a space target under the northeast coordinate system of the first radar to obtain a first position, a first speed and first covariance information comprising a first position covariance, a first speed covariance and a first position speed cross covariance at the last observation time; transforming the first position, the first speed and the first covariance information to a geocentric inertial coordinate system to obtain a second position, a second speed and second covariance information; Constructing a forecast initial state and forecast error initial covariance information according to the second position, the second speed and the second covariance information, and carrying out forecast extrapolation to obtain a forecast state and forecast error covariance information at a future moment; Determining the entering time of the space target into the coverage airspace of the second radar according to the forecasting state and the forecasting error covariance information, and transmitting the forecasting state and the forecasting error covariance information corresponding to the entering time to the second radar; The second radar plans an interception position and an interception area based on the prediction state and the prediction error covariance information and tracks the interception position and the interception area to obtain candidate state and candidate covariance information; based on the forecast state, the forecast error covariance information, the candidate state and the candidate covariance information, a test statistic is constructed, and a cross-radar association result is determined by combining the multi-moment association quality evaluation, so that the cross-radar association tracking of the space target is completed.
2. The method for spatial target multi-radar cross-over tracking based on forecast error covariance transfer of claim 1, wherein the step of transforming the first position, the first velocity, and the first covariance information to a geocentric inertial coordinate system to obtain the second position, the second velocity, and the second covariance information comprises: Constructing a coordinate transformation matrix based on the geodetic longitude, the geodetic latitude, the geodetic elevation and the earth rotation angular velocity of the station where the first radar is located; And linearly transforming the first position, the first speed and the first covariance information by using the coordinate transformation matrix to obtain second position, second speed and second covariance information, wherein the second covariance information comprises second position covariance, second speed covariance and second position speed cross covariance.
3. The prediction error covariance transfer-based spatial target multi-radar cross-connection tracking method of claim 1, wherein the prediction extrapolation comprises calculating an acceleration vector of the spatial target in a geocentric inertial coordinate system according to the earth gravitational field parameters and a prediction initial state by using a spatial dynamics model, and updating the prediction state with the prediction initial state and the acceleration vector, wherein the spatial dynamics model comprises an earth non-spherical gravity term.
4. The method for spatial target multi-radar cross-connection tracking based on prediction error covariance transfer of claim 3, wherein the prediction extrapolation further comprises constructing a linearization error propagation relationship based on a jacobian matrix of a spatial dynamics model, propagating prediction error initial covariance information in a combination form of the jacobian matrix and a state transition matrix, and introducing process noise covariance information in a propagation process to obtain prediction error covariance information at a future time.
5. The method for tracking the cross of the space target with multiple radars based on the forecast error covariance transfer according to claim 1, wherein the step of determining the entering time of the space target into the coverage space of the second radar according to the forecast state and the forecast error covariance information comprises the steps of converting the forecast states of a plurality of future moments into coordinate representations based on the second radar and comparing the coordinate representations with the coverage space of the second radar, and selecting the earliest moment of the second radar detected the target from the forecast moments meeting the constraint of the coverage space as the entering time.
6. The method for tracking the cross of the space target multiple radars based on the forecast error covariance transfer according to claim 1, wherein the step of planning the interception position and the interception area by the second radar based on the forecast state and the forecast error covariance information comprises the steps of carrying out coordinate transformation on the forecast error covariance information in a measurement space of the second radar, constructing a three-dimensional envelope area surrounding the forecast state according to the eigenvectors and eigenvalues of the transformed covariance matrix, and executing interception search in the three-dimensional envelope area.
7. The method for spatial target multi-radar cross-connection tracking based on prediction error covariance transfer according to claim 1, wherein the step of constructing test statistics based on the prediction state, the prediction error covariance information, the candidate state and the candidate covariance information comprises the steps of forming a state difference vector by a difference value between the prediction state and the candidate state, forming a joint covariance matrix by adding the prediction error covariance information and the candidate covariance information, and performing matrix multiplication operation by using a transpose of the state difference vector, an inverse of the joint covariance matrix and the state difference vector to obtain the test statistics.
8. The method for spatial target multi-radar cross-connection tracking based on forecast error covariance transfer of claim 1, wherein the step of determining cross-radar association results in combination with multi-time association quality assessment includes calculating a test statistic once for each pair of forecast and candidate states, updating an association quality function based on a comparison between the test statistic and a preset limit, adding a predetermined increment to the association quality function when association is judged to be successful, keeping the association quality function unchanged when association is judged to be failed, normalizing the association quality function based on a quantitative relationship between the number of successful association times and the number of associated judgment times in a set of consecutive sampling times, and determining cross-radar association results based on the normalized results.
9. The prediction error covariance transfer-based spatial target multi-radar cross-connection tracking method according to claim 8, wherein when the correlation quality function is updated, a weight value related to the test statistic is allocated to the correlation success record according to the magnitude of the test statistic at the corresponding sampling time, the weight value is reduced along with the increase of the test statistic, and the weight value at the early sampling time is attenuated in the continuous sampling process, so that the correlation quality function sensitive to the recent correlation result is constructed.
10. The prediction error covariance transfer-based spatial target multi-radar cross-connection tracking method according to claim 8, wherein when the second radar generates a plurality of candidate states and candidate covariance information at the same sampling time, test statistics are calculated for each candidate state, a candidate state with the smallest test statistics among candidate states satisfying a preset limit is selected as an associated object at the current sampling time, and an associated quality function is updated based on the associated object.

Description

Space target multi-radar handover tracking method based on prediction error covariance transfer Technical Field The invention belongs to the technical field of radar target tracking, and particularly relates to a spatial target multi-radar handover tracking method based on prediction error covariance transfer. Background In the prior art, multi-radar space target monitoring generally relies on the front and rear radars each independently completing tracking, and the handover requirements are met by extrapolating the target state near the coverage boundary. However, existing handoff approaches suffer from significant drawbacks in terms of trajectory consistency between different dynamics models, consistent expression of cross-coordinate system error statistics, and stability of cross-radar track correlations. In the prior art, the prediction process is mostly based on ground fixed coordinates or measurement coordinates, covariance information is generally limited to the inside of a single radar, an error statistical structure among radars is weak, meanwhile, an extrapolation model is not uniform in the processing of the perturbation force, so that the track continuity in the handover stage is poor, and the prediction performance entering a coverage airspace is easy to be influenced by the change of the perturbation force to fluctuate. In addition, the traditional association method is independent of single comparison of position difference or speed difference, the statistical support of the association basis is weak, and under the condition of weak measurement or error accumulation, the stability of association judgment is low, so that the problem of unstable association quality change is easy to occur. Therefore, the problems of poor continuity of the prediction state, weak consistency of error statistics, low correlation reliability across radars and the like exist in the prior art. This is a disadvantage of the prior art. In view of the foregoing, it is desirable to provide a spatial target multi-radar cross-connection tracking method based on prediction error covariance transfer, so as to solve the above-mentioned drawbacks in the prior art. Disclosure of Invention The invention aims to provide a space target multi-radar handover tracking method based on prediction error covariance transfer, aiming at the defects of poor continuity of prediction states, weak error statistics consistency and low cross-radar correlation reliability in the prior art, so as to solve the technical problems. In order to achieve the above purpose, the present invention provides the following technical solutions: A space target multi-radar handover tracking method based on forecast error covariance transfer comprises the following steps: Filtering and tracking a space target under the northeast coordinate system of the first radar to obtain a first position, a first speed and first covariance information comprising a first position covariance, a first speed covariance and a first position speed cross covariance at the last observation time; transforming the first position, the first speed and the first covariance information to a geocentric inertial coordinate system to obtain a second position, a second speed and second covariance information; Constructing a forecast initial state and forecast error initial covariance information according to the second position, the second speed and the second covariance information, and carrying out forecast extrapolation to obtain a forecast state and forecast error covariance information at a future moment; Determining the entering time of the space target into the coverage airspace of the second radar according to the forecasting state and the forecasting error covariance information, and transmitting the forecasting state and the forecasting error covariance information corresponding to the entering time to the second radar; The second radar plans an interception position and an interception area based on the prediction state and the prediction error covariance information and tracks the interception position and the interception area to obtain candidate state and candidate covariance information; based on the forecast state, the forecast error covariance information, the candidate state and the candidate covariance information, a test statistic is constructed, and a cross-radar association result is determined by combining the multi-moment association quality evaluation, so that the cross-radar association tracking of the space target is completed. By adopting the technical scheme, the continuous state deduction and error expression link is constructed among multiple radars, so that the consistent processing of state and covariance information under a cross-coordinate system is realized, the prediction track can be enabled to have higher consistency, the error statistics can be kept to be stronger in uniformity, the association judgment has better stability, and the requirements of high