A multi-platform tracking method based on high and low orbit satellites

Abstract

This invention discloses a multi-platform tracking method based on high-Earth orbit (HEO) and low-Earth orbit (LEO) satellites, belonging to the field of target tracking. The method includes: acquiring the raw observation values ​​of HEO and LEO satellites at their respective measurement times; calculating the target's position coordinates in a protocol-defined geocentric inertial coordinate system; establishing an IMM estimator and outputting a predicted state vector; constructing the satellite's measurement equation for the target and obtaining the target's corrected state vector using the predicted position coordinates; constructing the observation matrix of HEO and LEO satellites in the protocol-defined geocentric inertial coordinate system and calculating the instantaneous geometric precision factor (IGMP); calculating the fusion weights of the observation channels based on the IMM; fusing the target's corrected state vector using the fusion weights; outputting the target's future position coordinates; and tracking the target. This method can easily access observation data from more types of observation platforms, achieving a more powerful target tracking capability.

Description

Technical Field

[0001] This invention relates to the field of aerospace telemetry, tracking and control and target tracking, specifically to a multi-platform tracking method based on high and low orbit satellites. Background Technology

[0002] With the rapid development of aerospace technology, the demand for space situational awareness and ground-based key target surveillance is increasing. Single satellite observation platforms have inherent limitations: high-orbit satellites (such as geostationary orbit satellites) have wide coverage and long dwell times, enabling continuous surveillance of large areas, but their observation distances are long, resulting in lower target positioning accuracy and refresh rates; low-orbit satellites (such as reconnaissance satellites and remote sensing satellite constellations) are closer to targets, enabling the acquisition of high-precision, high-spatiotemporal resolution observation data, but their overpass time for specific areas is short, resulting in observation gaps and making it difficult to achieve continuous target tracking independently.

[0003] In existing technologies, multi-platform data fusion tracking is mostly concentrated on homogeneous platforms (such as multiple radars) or simple observation relay. It lacks the ability to coordinate heterogeneous and heterogeneous high and low orbit satellites. The observation data of high and low orbit satellites have significant differences in timestamps, coordinate systems, and accuracy levels. Direct fusion introduces large errors. The tracked target (such as aircraft, ships, and hypersonic vehicles) may maneuver, and a single motion model is difficult to adapt to its complex motion patterns. Summary of the Invention

[0004] To address the aforementioned shortcomings in existing technologies, this invention provides a multi-platform tracking method based on high and low orbit satellites, achieving unified spatiotemporal reference, adaptive target maneuvering, and multi-platform target tracking with fused weights.

[0005] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows:

[0006] A multi-platform tracking method based on high and low Earth orbit satellites is provided, which includes:

[0007] Step S1: Obtain the raw observation values ​​of high-orbit and low-orbit satellites at their respective measurement times, calculate the vector measurement value of the target position, and obtain the target's position in the agreed geocentric inertial orbit. Position coordinates in a coordinate system;

[0008] Step S2: Establish IMM estimators on both the high-orbit and low-orbit satellite observation channels, construct the target's state vector using the position coordinates, input it into the IMM estimator, and output the predicted state vector.

[0009] Step S3: Construct the satellite's measurement equation for the target, calculate the predicted position coordinates using the predicted state vector, and correct the predicted position coordinates using cosine similarity to obtain the target corrected state vector.

[0010] Step S4: Construct the geocentric inertial configuration of the high-orbit and low-orbit satellites based on the target-corrected state vector. The observation matrix in the coordinate system is used to calculate the position error covariance matrix of high-orbit and low-orbit satellites, and then the instantaneous geometric accuracy factor is obtained.

[0011] Step S5: Calculate the fusion weight of the observation channel based on the instantaneous geometric accuracy factors of the high-orbit and low-orbit satellites, use the fusion weight to fuse the target's corrected state vector, output the target's future position coordinates, and track the target.

[0012] Further, step S1 includes:

[0013] Step S11: Obtain the measurements of the high-orbit satellite h and the low-orbit satellite l at their respective measurement times. The raw observations, including azimuth angles Pitch angle and the slope distance R, and the respective measurement times. Unify to Coordinated Universal Time (UTC), then convert to continuous second-counting observation time t;

[0014] Step S12: Unify the raw observations and satellite platform status to the protocol geocentric inertial system. In the coordinate system, the satellite platform status includes position. and speed The vector measurement value of the target position is obtained. And based on vector measurements Output target in the protocol geocentric inertia Position coordinates in coordinate system ;

[0015] ;

[0016] in, This is the unit line-of-sight vector from the satellite to the target.

[0017] Further, step S2 includes:

[0018] Step S21: Establish an independent IMM estimator for the observation channel of high-orbit satellite h and low-orbit satellite l respectively. The IMM estimator contains a uniform velocity model CV and a uniform acceleration model CA.

[0019] Step S22: Using position coordinates Constructing the target in the protocol's geocentric inertia State vector in coordinate system ,in, For the target in the protocol's centripetal inertia velocity components in the coordinate system For the target in the protocol's centripetal inertia Acceleration components in the coordinate system, where k is the observation time index;

[0020] Step S23: Construct the state transition matrix using the uniform velocity model (CV) and the uniform acceleration model (CA) respectively. State transition matrix The output is based on the previous observation time. State vector Predicted state vector .

[0021] Further, step S3 includes:

[0022] Step S31: Construct the satellite's measurement equations for the target: H is the measurement matrix. For observation time The target is in the agreement's geocentric inertia Position coordinates in coordinate system , The observation noise matrix;

[0023] Step S32: The predicted state vector Input the coordinates into the measurement equations respectively, and output the position coordinates predicted by the uniform velocity model (CV) and the uniform acceleration model (CA). And calculate the predicted position coordinates respectively. Coordinates of the observed position Cosine similarity between ;

[0024] Step S33: Utilize cosine similarity Calculate the prediction accuracy weights for the uniform velocity model (CV) and the uniform acceleration model (CA). ;

[0025] Step S34: Utilize prediction accuracy weights Calculate the observation time of high-orbit satellite h and low-orbit satellite l. Predicted location coordinates .

[0026] Step S35: Predict the location coordinates Input them into the measurement equations respectively to calculate the observation time of the high-orbit satellite h and the low-orbit satellite l. Target modified state vector .

[0027] Further, step S4 includes:

[0028] Step S41: Correct the state vector based on the target Constructing high-orbit satellite h and low-orbit satellite l in a protocol-based geocentric inertial system Observation matrix in coordinate system ;

[0029] ;

[0030] Step S42: Using the observation matrix Calculate the position error covariance matrix of high-orbit satellite h and low-orbit satellite l. ;

[0031] ;

[0032] in, These are the original observation error covariance matrices for high-orbit satellite h and low-orbit satellite l, respectively.

[0033] Step S43: Utilize the position error covariance matrix Calculate the instantaneous geometric precision factor for high-orbit satellite h and low-orbit satellite l. ;

[0034] ;

[0035] in, Let be the trace function of the matrix.

[0036] Further, step S5 includes:

[0037] Step S51: Based on the instantaneous geometric precision factor Calculate the fusion weights of the observation channels of high-orbit satellite h and low-orbit satellite l. ;

[0038] Step S52: The observation channels for high-orbit satellite h and low-orbit satellite l output the observation time using the current IMM estimator. Corresponding target corrected state vector and utilize fusion weights Correct the target state vector By merging the data, the observation time can be obtained. Fusion state vector ;

[0039] Step S16: Fuse the state vector Input measurement equation, output target observation time In the agreement's geocentric inertia Position coordinates in a coordinate system are used to track the target.

[0040] The beneficial effects of this invention are as follows: This invention provides a continuous and stable tracking baseline and coarse-precision status through high-orbit satellites, effectively compensating for the observation gaps of low-orbit satellites; low-orbit satellites provide high-precision observations during overhead transit, significantly improving the instantaneous accuracy of tracking. The IMM algorithm is used to construct a target motion model, achieving adaptive target state and position correction, effectively addressing the target's maneuvering behavior. The adaptive weighted fusion strategy based on IGDOP quantifies the instantaneous contribution of each platform from the geometric root of observation, which is more scientific than traditional fixed-weight or observation noise variance-based weighted strategies, significantly improving the accuracy and stability of the fused trajectory. This method can easily access observation data from more types of observation platforms, achieving more powerful target tracking capabilities. Attached Figure Description

[0041] Figure 1 This is a flowchart of a multi-platform tracking method based on high and low orbit satellites. Detailed Implementation

[0042] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0043] like Figure 1 As shown, a multi-platform tracking method based on high and low Earth orbit satellites includes:

[0044] Step S1: Obtain the raw observation values ​​of high-orbit and low-orbit satellites at their respective measurement times, calculate the vector measurement value of the target position, and obtain the target's position in the agreed geocentric inertial orbit. Position coordinates in the coordinate system. Step S1 specifically includes:

[0045] Step S11: Obtain the measurements of the high-orbit satellite h and the low-orbit satellite l at their respective measurement times. The raw observations, including azimuth angles Pitch angle and the slope distance R, and the respective measurement times. Unify to Coordinated Universal Time (UTC), then convert to continuous second-counting observation time t;

[0046] Step S12: Unify the raw observations and satellite platform status to the protocol geocentric inertial system. In the coordinate system, the satellite platform status includes position. and speed The vector measurement value of the target position is obtained. And based on vector measurements Output target in the protocol geocentric inertia Position coordinates in coordinate system ;

[0047] ;

[0048] in, This is the unit line-of-sight vector from the satellite to the target.

[0049] Location of this embodiment and speed Calculated from precise ephemeris, the slant range R represents the measured slant range from the satellite to the target, and the azimuth angle is... This indicates the azimuth angle (based on true north, increasing clockwise) and elevation angle from the satellite towards the target. This indicates the elevation angle (relative to the local horizontal plane) from the satellite pointing towards the target.

[0050] Step S1 transforms the original spherical coordinate observations, which have different physical meanings, into "position measurements" and their accuracy descriptions that can be directly used by the state estimator, laying a unified data foundation for subsequent fusion tracking.

[0051] The IMM algorithm requires accurate observation times to calculate the correct time intervals (for the state transition matrix and process noise). Step S1 provides a uniform observation time, ensuring the accuracy of subsequent target motion model extrapolation. Step S1 resolves the spatial reference problem for step S2, ensuring all observations and states are in a agreed-upon geocentric inertial environment. The coordinate system representation allows the state vector (the target's position and velocity in the ECI coordinate system) to be directly compared and corrected with the observed position.

[0052] Step S2: Establish IMM estimators on both the high-orbit and low-orbit satellite observation channels. Construct the target's state vector using the position coordinates and input it into the IMM estimator, outputting the predicted state vector. Step S2 specifically includes:

[0053] Step S21: Establish an independent IMM estimator for the observation channel of high-orbit satellite h and low-orbit satellite l respectively. The IMM estimator contains a uniform velocity model CV and a uniform acceleration model CA.

[0054] Step S22: Using position coordinates Constructing the target in the protocol's geocentric inertia State vector in coordinate system ,in, For the target in the protocol's centripetal inertia velocity components in the coordinate system For the target in the protocol's centripetal inertia Acceleration components in the coordinate system, where k is the observation time index;

[0055] Step S23: Construct the state transition matrix using the uniform velocity model (CV) and the uniform acceleration model (CA) respectively. State transition matrix The output is based on the previous observation time. State vector Predicted state vector ;

[0056] ;

[0057] in, It is a 3-order identity matrix. It is a 3rd order zero matrix. To predict the update cycle, These are the process noise covariance matrices for the uniform velocity model (CV) and the uniform acceleration model (CA), respectively. These are the process noise intensity coefficients for the uniform velocity model (CV) and the uniform acceleration model (CA), respectively. These are the state vectors predicted by the uniform velocity model (CV) and the uniform acceleration model (CA), respectively.

[0058] Step S3: Construct the satellite's measurement equations for the target, calculate the predicted position coordinates using the predicted state vector, and correct the predicted position coordinates using cosine similarity to obtain the corrected target state vector. Step S3 specifically includes:

[0059] Step S31: Construct the satellite's measurement equations for the target: H is the measurement matrix. For observation time The target is in the agreement's geocentric inertia Position coordinates in coordinate system , The observation noise matrix;

[0060] The measurement matrix H establishes a bridge between the internal state and external observation, and the position coordinates observed by the sensor are directly equal to the position coordinate components in the true state of the target.

[0061] Step S32: The predicted state vector Input the coordinates into the measurement equations respectively, and output the position coordinates predicted by the uniform velocity model (CV) and the uniform acceleration model (CA). And calculate the predicted position coordinates respectively. Coordinates of the observed position Cosine similarity between ;

[0062] ;

[0063] Step S33: Utilize cosine similarity Calculate the prediction accuracy weights for the uniform velocity model (CV) and the uniform acceleration model (CA). ;

[0064] ;

[0065] Step S34: Utilize prediction accuracy weights Calculate the observation time of high-orbit satellite h and low-orbit satellite l. Predicted location coordinates ;

[0066] ;

[0067] Step S35: Predict the location coordinates Input them into the measurement equations respectively to calculate the observation time of the high-orbit satellite h and the low-orbit satellite l. Target modified state vector .

[0068] Step S4: Construct the geocentric inertial configuration of the high-orbit and low-orbit satellites based on the target-corrected state vector. The observation matrix in the coordinate system is used to calculate the position error covariance matrix of high-orbit and low-orbit satellites, thereby obtaining the instantaneous geometric accuracy factor. Step S4 specifically includes:

[0069] Step S41: Correct the state vector based on the target Constructing high-orbit satellite h and low-orbit satellite l in a protocol-based geocentric inertial system Observation matrix in coordinate system ;

[0070] ;

[0071] Step S42: Using the observation matrix Calculate the position error covariance matrix of high-orbit satellite h and low-orbit satellite l. ;

[0072] ;

[0073] in, These are the original observation error covariance matrices for high-orbit satellite h and low-orbit satellite l, respectively.

[0074] Step S43: Utilize the position error covariance matrix Calculate the instantaneous geometric precision factor for high-orbit satellite h and low-orbit satellite l. ;

[0075] ;

[0076] in, Let be the trace function of the matrix.

[0077] Position error covariance matrix It includes both the satellite's original measurement error and the influence of observation geometry. The smaller the instantaneous geometric accuracy factor, the more favorable the satellite's observation for positioning is under the current geometry.

[0078] Step S5: Calculate the fusion weights of the observation channels based on the instantaneous geometric accuracy factors of the high-orbit and low-orbit satellites. Use these fusion weights to fuse the target's corrected state vector, outputting the target's future position coordinates for target tracking. Step S5 specifically includes:

[0079] Step S51: Based on the instantaneous geometric precision factor Calculate the fusion weights of the observation channels of high-orbit satellite h and low-orbit satellite l. ;

[0080] ;

[0081] in, As a correction, this embodiment takes... The value should be a very small positive number to prevent the denominator from being 0;

[0082] Step S52: The observation channels for high-orbit satellite h and low-orbit satellite l output the observation time using the current IMM estimator. Corresponding target corrected state vector and utilize fusion weights Correct the target state vector By merging the data, the observation time can be obtained. Fusion state vector ;

[0083] This invention introduces fusion weights, which dynamically reflect the amplification or reduction effect of the current observation geometry on the error through the instantaneous geometric accuracy factor. For example, although a low-orbit satellite has high original accuracy, if its line of sight is nearly parallel to the target's motion direction (poor geometry), its instantaneous geometric accuracy factor will increase sharply, and its weight will decrease accordingly; conversely, a high-orbit satellite may also obtain a high fusion weight under certain geometry.

[0084] Step S16: Fuse the state vector Input measurement equation, output target observation time In the agreement's geocentric inertia Position coordinates in a coordinate system are used to track the target.

[0085] This invention provides a continuous and stable tracking baseline and coarse-precision status through high-orbit satellites, effectively compensating for the observation gaps of low-orbit satellites; low-orbit satellites, in turn, provide high-precision observations during overhead transits, significantly improving the instantaneous accuracy of tracking. An IMM algorithm is used to construct a target motion model, enabling adaptive target state and position correction, effectively addressing the target's maneuvering behavior. An adaptive weighted fusion strategy based on IGDOP quantifies the instantaneous contribution of each platform from the geometric root of observation, which is more scientific than traditional fixed-weight or observation noise variance-based weighting strategies, significantly improving the accuracy and stability of the fused trajectory. This method can easily integrate observation data from more types of observation platforms, achieving more powerful target tracking capabilities.

Claims

1. A multi-platform tracking method based on high and low Earth orbit satellites, characterized in that, include: Step S1: Obtain the raw observation values ​​of high-orbit and low-orbit satellites at their respective measurement times, calculate the vector measurement value of the target position, and obtain the target's position in the agreed geocentric inertial orbit. Position coordinates in a coordinate system; Step S2: Establish IMM estimators on both the high-orbit and low-orbit satellite observation channels, construct the target's state vector using the position coordinates, input it into the IMM estimator, and output the predicted state vector. Step S3: Construct the satellite's measurement equation for the target, calculate the predicted position coordinates using the predicted state vector, and correct the predicted position coordinates using cosine similarity to obtain the target corrected state vector. Step S4: Construct the geocentric inertial configuration of the high-orbit and low-orbit satellites based on the target-corrected state vector. The observation matrix in the coordinate system is used to calculate the position error covariance matrix of high-orbit and low-orbit satellites, and then the instantaneous geometric accuracy factor is obtained. Step S5: Calculate the fusion weight of the observation channel based on the instantaneous geometric accuracy factor of the high-orbit and low-orbit satellites, use the fusion weight to fuse the target's corrected state vector, output the target's future position coordinates, and track the target. Step S3 includes: Step S31: Construct the satellite's measurement equations for the target: H is the measurement matrix. For observation time The target is in the agreement's geocentric inertia Position coordinates in coordinate system , For the observation noise matrix, For the target in the agreement's geocentric inertia State vector in the coordinate system; Step S32: Convert the predicted state vector Input the coordinates into the measurement equations respectively, and output the position coordinates predicted by the uniform velocity model (CV) and the uniform acceleration model (CA). And calculate the predicted position coordinates respectively. Coordinates of the observed position Cosine similarity between ; ； Step S33: Utilize cosine similarity Calculate the prediction accuracy weights for the uniform velocity model (CV) and the uniform acceleration model (CA). ; ； Step S34: Utilize prediction accuracy weights Calculate the observation time of high-orbit satellite h and low-orbit satellite l. Predicted location coordinates ; ； Step S35: Predict the location coordinates Input them into the measurement equations respectively to calculate the observation time of the high-orbit satellite h and the low-orbit satellite l. Target modified state vector .

2. The multi-platform tracking method based on high and low orbit satellites according to claim 1, characterized in that, Step S1 includes: Step S11: Obtain the measurements of the high-orbit satellite h and the low-orbit satellite l at their respective measurement times. The raw observations, including azimuth angles Pitch angle and the slope distance R, and the respective measurement times. Unify to Coordinated Universal Time (UTC), then convert to continuous second-counting observation time t; Step S12: Unify the raw observations and satellite platform status to the protocol geocentric inertial system. In the coordinate system, the satellite platform status includes position. and speed The vector measurement value of the target location is obtained. And based on vector measurements Output target in the protocol geocentric inertia Position coordinates in coordinate system ; ； in, This is the unit line-of-sight vector from the satellite to the target.

3. The multi-platform tracking method based on high and low orbit satellites according to claim 2, characterized in that, Step S2 includes: Step S21: Establish an independent IMM estimator for the observation channel of high-orbit satellite h and low-orbit satellite l respectively. The IMM estimator contains a uniform velocity model CV and a uniform acceleration model CA. Step S22: Using position coordinates Constructing the target in the protocol's geocentric inertia State vector in coordinate system ,in, For the target in the agreement's geocentric inertia Velocity components in the coordinate system For the target in the agreement's geocentric inertia Acceleration components in the coordinate system, where k is the observation time index; Step S23: Construct the state transition matrix using the uniform velocity model (CV) and the uniform acceleration model (CA) respectively. State transition matrix The output is based on the previous observation time. State vector Predicted state vector .

4. The multi-platform tracking method based on high and low orbit satellites according to claim 3, characterized in that, Step S4 includes: Step S41: Correct the state vector based on the target Constructing high-orbit satellite h and low-orbit satellite l in a protocol-based geocentric inertial system Observation matrix in coordinate system ; ； Step S42: Using the observation matrix Calculate the position error covariance matrix of high-orbit satellite h and low-orbit satellite l. ; ； in, These are the original observation error covariance matrices for high-orbit satellite h and low-orbit satellite l, respectively. Step S43: Utilize the position error covariance matrix Calculate the instantaneous geometrical accuracy factor for high-orbit satellite h and low-orbit satellite l. ; ； in, Let be the trace function of the matrix.

5. The multi-platform tracking method based on high and low orbit satellites according to claim 4, characterized in that, Step S5 includes: Step S51: Based on the instantaneous geometric precision factor Calculate the fusion weights of the observation channels of high-orbit satellite h and low-orbit satellite l. ; Step S52: The observation channels for high-orbit satellite h and low-orbit satellite l output the observation time using the current IMM estimator. Corresponding target corrected state vector and utilize fusion weights Correct the target state vector By merging the data, the observation time can be obtained. Fusion state vector ; Step S53: Fuse the state vector Input measurement equation, output target observation time In the agreement's geocentric inertia Position coordinates in a coordinate system are used to track the target.

Images