A multi-coordinate system conversion method

Abstract

The present application belongs to the field of space coordinate control, and mainly relates to a kind of multiple coordinate system conversion method, which is around the unified conversion demand of satellite orbit state between inertial coordinate system, geostationary coordinate system, local orbit coordinate system, load coordinate system and relative encounter coordinate system, first form reference orbit state, then with reference orbit state as unified reference synchronous derivation multi-coordinate system state;Further, through cross-coordinate system link, extraction closed-loop difference and conversion difference, form correction residual set, and feed back correction residual to orbit state and attitude parameter;On this basis, regenerate multi-coordinate system result, and combine quality determination output target conversion result;The method can reduce the influence of inconsistent phase reference, geometric relationship drift and inconsistent result expression in multiple conversion process, improve the consistency, stability and availability of multi-coordinate system result.

Description

Technical Field

[0001] This invention belongs to the field of aerospace coordinate control and mainly relates to a method for transforming multiple coordinate systems. Background Technology

[0002] In low-Earth orbit (LEO) satellite multi-mission collaborative operation scenarios, such as LEO satellite constellations performing collision avoidance and early warning, inter-satellite link maintenance, ground payload pointing, and ground coverage analysis tasks, the orbital state of the same satellite at multiple forecast epochs needs to be repeatedly switched between the inertial coordinate system, the ground-fixed coordinate system, the local orbit coordinate system, the payload coordinate system, and the relative encounter coordinate system. This scenario is characterized by two main aspects: firstly, it requires simultaneous processing of long-term orbit forecasts, ground projection representation, payload line-of-sight control, and relative encounter geometry, resulting in frequent coordinate system switching; secondly, J2 perturbations continuously cause long-term drift in the ascending node and perigee directions, causing the geometric relationships in different coordinate systems to change over time. In this scenario, if there is a lack of a unified state reference among multiple coordinate systems, problems such as inconsistent orbital phase references for the same forecast state in different coordinate systems, accumulated local orbital axial drift, and inconsistent encounter geometry and pointing results can easily arise. This makes it difficult to maintain a consistent long-term representation of the collision avoidance decision window, beam pointing angle, ground landing point position, and relative approach direction.

[0003] To address the aforementioned issues, existing technologies typically employ a phased coordinate transformation approach. This involves: first, based on the determined orbit or initial state vector, orbit prediction is performed using the J2 perturbation model in the inertial coordinate system; then, based on Earth's rotation and orientation parameters, the prediction results are converted to the Earth-fixed coordinate system; subsequently, local orbital coordinate systems such as LVLH and RTN are constructed based on the satellite's position and velocity vectors; and when payload pointing analysis or relative encounter analysis is required, the results are further converted to the celestial coordinate system, payload coordinate system, or relative encounter coordinate system using attitude quaternions or installation matrices, thus completing the basic orbital state conversion under each target coordinate system.

[0004] However, existing technologies have certain drawbacks: the J2 long-term drift calculation and multi-coordinate system transformation are separated into independent links, resulting in a lack of unified phase reference among the inertial frame, the Earth-fixed frame, and the local orbital frame. With increased prediction duration or multiple transformations, residuals are easily generated due to orbital plane rotation, Earth rotation compensation, and asynchronous local orbital axis construction. Furthermore, even if deviations in the outputs of different coordinate systems are detected, existing technologies only smooth or locally correct the results in the target coordinate system, failing to feed back geometric inconsistencies in the local orbital frame, load frame, and relative encounter frame to the common orbital state. Therefore, it is difficult to simultaneously and stably output the minimum collision avoidance distance direction, the load line-of-sight pointing angle, the ground projection point position, and the approach direction in the relative encounter plane. To address these issues, this invention proposes a multi-coordinate system transformation method. Summary of the Invention

[0005] To address the problems in existing technologies that treat J2 long-term drift calculation and multi-coordinate system transformation as independent links, which lack a unified phase reference and fail to feed back geometrically inconsistent information from the local orbit system, load system, and relative encounter system to the common orbit state, resulting in residual accumulation and difficulty in stably outputting the minimum collision avoidance distance direction, load line-of-sight pointing angle, ground projection point position, and approach direction in the relative encounter plane, this invention provides a multi-coordinate system transformation method to solve the problems mentioned in the background technology.

[0006] To solve the above problems, the present invention employs the following technology: A method for transforming multiple coordinate systems, including: Collect raw orbital data of the target satellite, and preprocess the raw orbital data to obtain orbital state data; Feature extraction is performed on the orbital state data to obtain a reference orbit feature set; long-term term propagation analysis is performed on the reference orbit feature set to obtain a reference orbit state set; multi-coordinate system transformation is performed based on the reference orbit state set to obtain a multi-coordinate system synchronous state set; closed-loop link construction is performed based on the multi-coordinate system synchronous state set to obtain a cross-coordinate system link; correction analysis is performed based on the cross-coordinate system link to obtain a corrected residual set; feedback correction is performed based on the corrected residual set to obtain a consistency correction result. Based on the consistency correction result, a re-output is performed to obtain a re-output result record; based on the re-output result record, a quality judgment process is performed to obtain the target conversion result.

[0007] The beneficial effects of this invention are: 1. Based on the same orbital state, generate multiple coordinate system states synchronously, construct a closed-loop link to compare residuals and reverse correct orbit and attitude parameters, eliminating the phase inconsistency problem between coordinate systems; 2. Long-term residual suppression and compensation: In the propagation of the long-term term of J2, the drift of the right ascension of the ascending node and the argument of the perigee are compensated. The residuals at multiple times are aligned, averaged, and then superimposed to the orbital state to suppress long-term cumulative deviation. 3. Based on the corrected orbital state and attitude parameters, the results of each coordinate system are re-output in a unified manner, which effectively avoids the problem of time base offset and state mismatch caused by independent calculation in traditional step-by-step transformation. Attached Figure Description

[0008] Figure 1 This is an exemplary method flowchart of a multi-coordinate system transformation method provided in an embodiment of the present invention; Figure 2 This is a comparison diagram showing the effects of a multi-coordinate system transformation method provided by an embodiment of the present invention compared to traditional inventions. Detailed Implementation

[0009] To make the technical means, creative features, and achieved objectives and effects of this invention easier to understand, the invention is further described below with reference to specific embodiments. However, the following embodiments are merely preferred embodiments of this invention and not all of them. Other embodiments obtained by those skilled in the art based on the embodiments described herein without creative effort are all within the protection scope of this invention. Unless otherwise specified, the experimental methods in the following embodiments are conventional methods, and the materials and reagents used in the following embodiments are commercially available unless otherwise specified.

[0010] Example 1 Figure 1 As shown in the flowchart of the present invention, this embodiment provides a method for multiple coordinate system transformations, specifically including the following steps: Step S1: Collect the raw orbital data of the target satellite and preprocess the raw orbital data to obtain orbital state data; Specifically, the raw orbital data to be collected includes: initial position velocity vector, initial orbital elements, target prediction time, attitude installation parameters, Earth orientation parameters, and target coordinate system type set; The initial position-velocity vector is used to characterize the instantaneous orbital state of the target satellite at the corresponding sampling time, including the position and velocity components of the target satellite at the corresponding time. Initial orbital elements are used to characterize the orbital geometry and orientation of the target satellite at a given time, including semi-major axis, eccentricity, orbital inclination, right ascension of the ascending node, argument of perigee, and perigee or mean perigee. The target prediction time is used to define the time range covered by subsequent orbit propagation and multi-coordinate system transformation, including one or more time points after the start time where results need to be output; Attitude installation parameters are used to characterize the installation relationship between the star coordinate system and the payload coordinate system, or to characterize the alignment relationship between the payload installation direction and the satellite body reference axis, including one or more of the installation matrix, installation angle parameters, and attitude zero offset parameters; Earth orientation parameters are used to characterize the orientation relationship between an inertial reference frame and a fixed Earth reference frame, including one or more of the following: Earth rotation-related parameters, polar motion parameters, and nutation and precession-related parameters. The target coordinate system type set is used to limit which coordinate systems the orbital state results need to be output in this processing, including one or more of the following: inertial coordinate system, ground-fixed coordinate system, local orbital coordinate system, load coordinate system, and relative encounter coordinate system when necessary; In addition to the orbit and attitude data mentioned above, time reference data also needs to be acquired. This data includes one or more of the following: UTC time stamp, on-board time stamp, and orbit propagation time stamp. The methods for obtaining the aforementioned raw orbit data specifically include: directly reading the orbit element file output by the ground orbit determination system for the initial orbit elements, or using the initial position velocity vector as input and calculating the orbit elements according to the predetermined orbit element conversion rules; For target prediction time, read from mission plan file, collision avoidance analysis mission instructions, inter-satellite link maintenance mission instructions or ground pointing mission instructions; For attitude installation parameters, it is preferable to read the installation matrix or installation angle parameters from the whole satellite ground calibration database; if the satellite is already in orbit and has generated correction parameter records, the currently valid installation compensation parameters can also be read from the on-orbit parameter update table. For Earth orientation parameters, it is preferable to read them from the ground-based parameter table; when real-time parameters are unavailable, the parameter values ​​corresponding to the current valid interval can also be read from the historical fitting parameter file. The target coordinate system type set can be read from the configuration of this business process. For example, in the collision avoidance analysis task, the inertial coordinate system, local orbit coordinate system and relative encounter coordinate system can be read; in the ground coverage analysis task, the inertial coordinate system, ground-fixed coordinate system and load coordinate system can be read; in the inter-satellite link maintenance task, the local orbit coordinate system and load coordinate system can be read. After reading, the coordinate system identifiers are combined to form the target coordinate system type set. For time reference data, it is read from the UTC field in the orbit source data, the on-board time stamp field in the on-board time synchronization record, and the propagation start and end time field in the mission propagation configuration, respectively. After acquiring the raw orbit data, the raw orbit data is preprocessed, including time unification processing, orbit status alignment processing, parameter validity verification processing, forecast sequence processing, and target output processing. The time unification process specifically involves: reading the UTC time stamp, on-board time scale, and orbit propagation time stamp, and identifying the time caliber corresponding to each time stamp; then, converting the UTC time stamp, on-board time scale, and orbit propagation time stamp to the same time base according to a predetermined time conversion relationship to generate a unified time series; subsequently, selecting a unified reference time from the unified time series that corresponds to the initial position velocity vector and the initial orbital elements, determining this unified reference time as the starting epoch, and using this starting epoch as the starting time for subsequent analysis; The orbital state alignment process is as follows: Based on the initial epoch, the position and velocity components corresponding to the initial position and velocity vector are aligned with the orbital element records corresponding to the initial orbital elements. When the initial position and velocity vector and the initial orbital elements are inconsistent at the corresponding time, the orbital state record consistent with the initial epoch is retained as the current propagation input. When only the initial position and velocity vector is obtained, the initial orbital elements are calculated according to the Kepler orbital element conversion rules, using the initial position and velocity vector as input. When only the initial orbital elements are obtained, the initial orbital elements are registered as the orbital element record under the initial epoch, and the position and velocity state is marked as a state to be recovered. Recovery is then performed: Based on the semi-major axis, eccentricity, orbital inclination, right ascension of the ascending node, argument of perigee, and mean perigee angle corresponding to the initial orbital elements, the position and velocity vectors under the initial epoch are recovered according to the conversion rules from orbital elements to inertial coordinate system state vectors, and the corresponding position and velocity state records are generated. The parameter validity verification process specifically involves: checking whether the initial position velocity vector contains null values, missing components, or abnormal components; checking whether the initial orbital elements contain missing items or parameter values ​​exceeding the predetermined physical range; checking whether the attitude installation parameters and Earth orientation parameters are within the current valid range; and checking whether the coordinate system identifiers in the target coordinate system type set contain duplicate or invalid items. When the verification passes, the corresponding parameters are written into the pending data record. When the verification fails, the corresponding reason for the anomaly is recorded, and the current data is marked as unpropagable. The forecast sequence processing is as follows: read the forecast times of each target and map them to the same time base; then sort the forecast times of each target in chronological order to generate a target forecast time series; subsequently, starting from the initial epoch, record the time interval between each target forecast time and the initial epoch as the propagation time parameter for subsequent long-term term propagation analysis. The target output processing is specifically as follows: Based on the installation matrix, installation angle parameters, or attitude zero-bias parameters recorded in the attitude installation parameters, an attitude alignment record is generated according to the correspondence between the celestial coordinate system and the payload coordinate system; based on the parameter values ​​in the Earth orientation parameters and their corresponding valid time intervals, an orientation parameter record is generated; deduplication and sequential arrangement processing are performed on each coordinate system identifier in the target coordinate system type set to generate a coordinate system output configuration record; among which, the sequential arrangement processing is used to fix the generation order of subsequent multi-coordinate system synchronization state sets; Organize the preprocessing results to generate orbital state data, including: initial position and velocity records at the starting epoch, initial orbital element records at the starting epoch, unified time reference records corresponding to the starting epoch, target prediction time series arranged in chronological order, propagation time parameters corresponding to each target prediction time, attitude alignment records after validity verification, orientation parameter records after validity verification, and coordinate system output configuration records after deduplication and sequential arrangement.

[0011] Step S2: Perform long-term propagation analysis on the orbital state data to obtain a reference orbital state set; Specifically, the initial orbital element records at the starting epoch are obtained from the orbital state data and used as orbital element inputs for long-term term propagation analysis; unified time reference records are obtained to ensure consistency of time caliber for each propagation epoch; and target forecast time series and propagation time parameters are obtained to limit the epoch range and propagation duration of long-term term propagation analysis. Reference orbit elements are extracted from the initial orbital element records at the starting epoch. Specifically, the semi-major axis, eccentricity, orbital inclination, right ascension of the ascending node, argument of perigee, and mean perigee are read from the initial orbital element records. Then, based on the forecast duration and orbital parameter stability, the above orbital elements are subjected to long-term term propagation adaptation processing to generate the J2 reference orbital element set. A long-term propagation analysis was performed on the J2 reference orbit feature set, specifically as follows: First, the right ascension parameters of the ascending node and the argument of perigee are updated: Starting from the initial epoch, the propagation time parameters corresponding to the predicted times of each target are read. Based on the semi-major axis, eccentricity, and orbital inclination from the J2 reference orbit element set, combined with the Earth's gravitational field constant, the long-term drift rate of the right ascension of the ascending node and the long-term drift rate of the argument of perigee are calculated using the formulas for the long-term drift rate of the right ascension of the ascending node and the long-term drift rate of the argument of perigee. Next, the propagation time parameters corresponding to the predicted times of each target are multiplied by the long-term drift rates of the right ascension of the ascending node and the long-term drift rates of the argument of perigee, respectively, to obtain the drift amount of the right ascension of the ascending node and the drift amount of the argument of perigee corresponding to the predicted times of each target. Subsequently, the right ascension of the ascending node and the argument of perigee at the initial epoch are added to the drift amount of the right ascension of the ascending node and the drift amount of the argument of perigee at the corresponding predicted times of the target, respectively, to obtain the propagated right ascension parameters of the ascending node and the propagated argument of perigee corresponding to the predicted times of the target. Time advance is performed on the mean apogee angle: the mean apogee angle at the initial epoch is read, and the average angular velocity at the current reference orbit is calculated according to the mean angular velocity formula based on the semi-major axis of the J2 reference orbit element set; then the propagation time parameter corresponding to each target prediction time is multiplied by the average angular velocity to obtain the mean apogee angle increment corresponding to each target prediction time; subsequently, the mean apogee angle at the initial epoch is added to the mean apogee angle increment at the corresponding target prediction time to obtain the propagated mean apogee angle corresponding to each target prediction time; After the propagation of the right ascension parameters of the ascending node, the propagation of the argument of perigee, and the propagation of the mean aperimeter, the semi-major axis, eccentricity, orbital inclination, the propagation of the right ascension parameters of the ascending node, the propagation of the argument of perigee, and the propagation of the mean aperimeter are written into the orbital element records of the corresponding target prediction time, thereby generating the propagation orbital element records corresponding to each target prediction time. Finally, the propagation orbital element records are processed to restore the reference state: each propagation orbital element record corresponding to the predicted time of each target is read one by one, and the semi-major axis, eccentricity, orbital inclination, right ascension parameter of the ascending node after propagation, argument of perigee after propagation, and mean perigee angle after propagation are read from each propagation orbital element record; according to the conversion rules from orbital elements to the state vector of the reference inertial frame, the position and velocity components in the orbital plane are generated; then, according to the coordinate rotation order corresponding to the orbital inclination, right ascension parameter of the ascending node after propagation, and argument of perigee after propagation, the position and velocity components in the orbital plane are rotated to obtain the position and velocity vectors in the reference inertial frame; then, the position and velocity vectors in the reference inertial frame are associated with the corresponding predicted time of the target to generate the reference orbital state record corresponding to the predicted time of each target; After generating the reference orbit state records corresponding to the predicted times of each target, the reference orbit state records are organized to obtain a reference orbit state set. The reference orbit state includes: the reference position vector, the reference velocity vector, and the time identifier corresponding to the reference position vector and the reference velocity vector for each target predicted time.

[0012] Step S3: Perform multi-coordinate system transformation based on the reference orbit state set to obtain a multi-coordinate system synchronous state set; construct a closed-loop link based on the multi-coordinate system synchronous state set to obtain a cross-coordinate system link; Specifically, based on the reference orbit state set, the reference position vector, reference velocity vector, and corresponding time identifier are read from each reference orbit state record; Earth orientation parameters, Earth rotation parameters, attitude installation parameters, and target coordinate system type set are read from the orbit state data; using the reference position vector and reference velocity vector as the basic state input for multi-coordinate system transformation, multi-coordinate system transformation is performed according to the coordinate system state type to be generated as defined in the target coordinate system type set, including inertial coordinate system state generation, ground-fixed coordinate system state generation, local orbit coordinate system state generation, load coordinate system state generation, and relative encounter coordinate system state generation; Specifically, the inertial coordinate system state generation involves directly setting the reference position vector and reference velocity vector in the reference orbit state record to the inertial coordinate system position and velocity corresponding to the current target prediction time, thereby generating the inertial coordinate system state record corresponding to the current target prediction time. Since the reference orbit state itself is the state vector recovered under the reference inertial frame, this process does not perform any additional transformations, but only performs state record and time association writing. The specific steps for generating the Earth-fixed coordinate system state are as follows: First, read the time stamp corresponding to the current target forecast time. Then, based on this time stamp, call the Earth's rotation and orientation parameters to generate a rotation matrix from the inertial coordinate system to the Earth-fixed coordinate system at the current target forecast time. Next, read the time stamp corresponding to the current target forecast time. Then, based on this time stamp, call the Earth's rotation and orientation parameters to generate a position rotation matrix from the inertial coordinate system to the Earth-fixed coordinate system at the current target forecast time. Finally, based on the position rotation matrix, perform coordinate rotation on the reference position vector to obtain the Earth-fixed coordinate system position vector corresponding to the current target forecast time. Then, based on the velocity transformation rules from the inertial coordinate system to the Earth-fixed coordinate system, perform velocity transformation on the reference velocity vector to obtain the Earth-fixed coordinate system velocity vector corresponding to the current target forecast time. Finally, associate the Earth-fixed coordinate system position vector with the corresponding time stamp to generate the Earth-fixed coordinate system state record corresponding to the current target forecast time. The local orbit coordinate system state generation process is as follows: The reference position vector and reference velocity vector corresponding to the current target prediction time are read, and the local orbit coordinate axes at the current target prediction time are constructed based on these vectors. Preferably, the local orbit coordinate system is an RTN coordinate system or an LVLH coordinate system, and the local orbit coordinate axes are generated based on the reference position vector and reference velocity vector. After constructing the local orbit coordinate axes, the reference position vector, reference velocity vector, or relative position state is projected onto the local orbit coordinate axes to obtain the local orbit coordinate system state variables corresponding to the current target prediction time. Subsequently, the local orbit coordinate system state variables are associated with the corresponding time identifier to generate the local orbit coordinate system state record corresponding to the current target prediction time. The load coordinate system state generation process is as follows: After determining the local orbit coordinate system state, the installation matrix, installation angle parameters, or attitude zero-bias parameters in the attitude installation parameters are read. Based on the installation relationship between the local orbit coordinate system and the load coordinate system, a load coordinate system transformation matrix is ​​generated for the current target prediction time. Then, load coordinate system transformation is performed on the local orbit coordinate axes corresponding to the current target prediction time to obtain the load coordinate system axial state and load line-of-sight pointing state corresponding to the current target prediction time. Subsequently, the load coordinate system axial state and load line-of-sight pointing state are associated and recorded with the corresponding time markers to generate the load coordinate system state record corresponding to the current target prediction time. When the target coordinate system type set includes a relative encounter coordinate system, the relative encounter coordinate system state generation is performed. Specifically, the reference orbit state and the relative reference state of the corresponding target object at the current target forecast time are read. Based on the position difference and velocity difference between the reference orbit state and the relative reference state, the relative position state and approach direction state within the relative encounter plane are generated according to the construction rules of the relative encounter plane. Subsequently, the relative position state, approach direction state and corresponding time identifier are associated and recorded to generate the relative encounter coordinate system state record corresponding to the current target forecast time. After generating the state records of the inertial coordinate system, the ground-fixed coordinate system, the local orbit coordinate system, the load coordinate system, and the relative encounter coordinate system, the state records of each coordinate system generated at the same target prediction time are collected and organized to generate a multi-coordinate system synchronous state record corresponding to the current target prediction time. The coordinate system results in the multi-coordinate system synchronous state record all come from the same reference orbit state record and share the same time identifier. After generating the multi-coordinate system synchronization state records corresponding to the predicted time of each target, the multi-coordinate system synchronization state records are organized to obtain the multi-coordinate system synchronization state set. The multi-coordinate system synchronization state set includes: the position and velocity state of the inertial coordinate system, the position state of the ground-fixed coordinate system, the state quantity of the local orbit coordinate system, the axial state or line-of-sight pointing state of the load coordinate system, and the relative position state and approach direction state generated when the target coordinate system type set includes the relative encounter coordinate system.

[0013] Step S4: Perform correction analysis based on cross-coordinate system links to obtain the correction residual set; use the correction residual set to perform feedback correction on the multi-coordinate system synchronization state set to obtain the consistency correction result; Specifically, a consistency analysis is performed based on the states of each coordinate system in the multi-coordinate system synchronization state set, as well as the corresponding time stamps, attitude installation parameter records, and earth orientation parameter records. The consistency analysis includes: Closed-loop link construction: Based on the coordinate system states already generated at the current target prediction time, construct cross-coordinate system links, including a first closed-loop link and a second transformation link. The first closed-loop link refers to: from the inertial coordinate system to the Earth-fixed coordinate system, from the Earth-fixed coordinate system to the local orbit coordinate system, and from the local orbit coordinate system back to the inertial coordinate system. The second transformation link refers to: from the inertial coordinate system to the local orbit coordinate system, from the local orbit coordinate system to the payload coordinate system, and from the payload coordinate system to the relative encounter coordinate system. When the target coordinate system type set does not include the relative encounter coordinate system, only the first closed-loop link at the current target prediction time is generated. The link status of the constructed cross-coordinate system link is compared, specifically as follows: For the first closed-loop link, the inertial coordinate system state record is taken as the starting state. According to the link sequence, the corresponding coordinate system transformation is performed sequentially according to the aerospace coordinate system transformation standard system to obtain the inertial coordinate system return state. The inertial coordinate system return state is compared with the starting state to generate the closed-loop difference value corresponding to the first closed-loop link. The closed-loop difference value includes one or more of the position difference value and the direction difference value. When a second conversion link exists, for the second conversion link, the inertial coordinate system state record is used as the starting state. According to the corresponding link sequence, the link consistency comparison is performed on the coordinate system states at the same target prediction time. The relative encounter coordinate system state is compared with the end state formed by the step-by-step transfer of the inertial coordinate system state to obtain the conversion difference corresponding to the second conversion link, including one or more of the relative position difference, approach direction difference and axial deviation value. The obtained closed-loop difference is recorded in the closed-loop residual record as the first residual item. When the transformation difference exists, it is recorded in the closed-loop residual record as the second residual item. At the same time, the link type, difference type and target prediction time corresponding to the first residual item and the second residual item are recorded to form the consistent residual set corresponding to the current target prediction time. Perform categorized residual sorting: extract all position and direction differences in the first residual term, and all axial deviations, proximity direction differences, and relative position differences in the second residual term to form a categorized difference set; Based on the type of the difference set, determine the object to be corrected: Set the correction object of the set of position residuals, orientation residuals and relative position residuals as the reference orbit state; The correction objects corresponding to the axial residual set and the approach direction residual set are determined as attitude installation parameters; Based on the identified correction object, the object content residuals are merged, including merging the residual groups corresponding to the reference orbit state and merging the residual groups corresponding to the attitude installation parameters. The process of merging the residual groups corresponding to the reference orbit state includes: merging the position residual group and the relative position residual group as a single group; aligning the residual values ​​within the merged group with the same epoch and averaging them to generate the position correction residual; and aligning the residual values ​​in the direction residual group with the same epoch and averaging them to generate the direction correction residual. The position correction residual and the orientation correction residual are recorded as the track state correction residual; Merge the residual groups corresponding to the attitude installation parameters, including: aligning each residual value in the axial residual group and each residual value in the approach direction residual group with the same epoch and averaging them respectively to obtain the axial correction difference and the approach direction correction difference, which are recorded as attitude parameter correction residuals. The orbital state correction residual and the attitude parameter correction residual are recorded as a correction residual set.

[0014] Feedback corrections are performed based on the obtained orbital state correction residuals and attitude parameter correction residuals, including: First, perform orbit state write-back on the orbit state correction residuals: read the position correction residuals and orientation correction residuals corresponding to the predicted time of each target one by one, and read the reference orbit state record corresponding to the predicted time of the target; superimpose the position correction residuals onto the reference position vector to obtain the corrected reference position vector; adjust the direction of the reference velocity vector according to the orientation correction residuals to obtain the corrected reference velocity vector; associate the corrected reference position vector, the corrected reference velocity vector with the corresponding time identifier and record them to generate the corrected orbit state record; Then, the attitude parameter correction residuals are written back: the axial correction difference and approach direction correction difference corresponding to each target prediction time are read one by one, and the currently valid attitude installation parameter record is read; when the attitude installation parameters are recorded using the installation matrix, the matrix elements in the corresponding axial and approach directions in the installation matrix are incrementally corrected according to the axial correction difference and approach direction correction difference to generate the corrected installation matrix; when the attitude installation parameters are recorded using installation angle parameters or attitude zero bias parameters, the corresponding angle parameters are compensated according to the axial correction difference and approach direction correction difference to generate the corrected attitude installation parameter record; the corrected installation matrix or the corrected attitude installation parameter record is associated with the corresponding time identifier and written to generate the attitude parameter correction record; The corrected orbital state record and attitude parameter correction record are stored as a consistency correction result.

[0015] Step S5: Re-output based on the consistency correction result to obtain the re-output result record; perform quality judgment processing based on the re-output result record to obtain the target transformation result; Specifically, the process involves reading the corrected orbital state record, attitude parameter correction record, corresponding time identifier, and coordinate system output configuration record from the consistency correction results; using the corrected reference position vector and corrected reference velocity vector from the corrected orbital state record as the re-output base state; and using the corrected installation matrix, corrected installation angle parameter, or corrected attitude zero bias parameter from the attitude parameter correction record as the attitude alignment input. Based on the coordinate system output order in the target coordinate system output configuration record, the process regenerates the multi-coordinate system output results corresponding to the prediction time of each target, including: For the output of the inertial coordinate system, read the corrected reference position vector and the corrected reference velocity vector corresponding to the predicted time of each target one by one, and write the corrected reference position vector and the corrected reference velocity vector directly into the inertial coordinate system result record under the corresponding time identifier to generate the inertial coordinate system position state and velocity state. For the Earth-fixed coordinate system output, the corrected reference position vector, corrected reference velocity vector, corresponding time stamp, and Earth orientation parameter record corresponding to each target's predicted time are read one by one. Based on the corresponding time stamp, the currently valid Earth rotation parameters and Earth orientation parameters are retrieved to generate the position rotation matrix and velocity transformation parameters from the inertial coordinate system to the Earth-fixed coordinate system. Then, coordinate rotation is performed on the corrected reference position vector to obtain the Earth-fixed coordinate system position vector, and velocity transformation is performed on the corrected reference velocity vector to obtain the Earth-fixed coordinate system velocity vector. After generating the Earth-fixed coordinate system position vector, the corresponding longitude, latitude, and altitude information, or the corresponding ground projection point position, are further calculated based on the Earth-fixed coordinate system position vector. Subsequently, the Earth-fixed coordinate system position vector, Earth-fixed coordinate system velocity vector, and longitude, latitude, and altitude information or ground projection point position are associated with the corresponding time stamp and recorded to generate the Earth-fixed coordinate system result record. For the local orbit coordinate system output, the corrected reference position vector and corrected reference velocity vector corresponding to each target prediction time are read one by one; the local orbit coordinate axes at the current target prediction time are reconstructed based on the corrected reference position vector and corrected reference velocity vector; after the local orbit coordinate axes are constructed, the corrected reference position vector, corrected reference velocity vector, or relative position state are projected onto the local orbit coordinate axes to obtain the local orbit coordinate system state variables; subsequently, the local orbit coordinate system state variables are associated with the corresponding time identifiers and recorded to generate the local orbit coordinate system result record; For the load coordinate system output, the local orbit coordinate axes corresponding to each target prediction time and the corrected installation matrix, corrected installation angle parameters, or corrected attitude zero bias parameters in the attitude parameter correction record are read one by one; based on the installation relationship between the local orbit coordinate system and the load coordinate system, the load coordinate system transformation matrix at the current target prediction time is regenerated; then the load coordinate system transformation is performed on the local orbit coordinate axes to obtain the load coordinate system axial state and load line-of-sight pointing state; subsequently, the load coordinate system axial state and load line-of-sight pointing state are associated and recorded with the corresponding time markers to generate the load coordinate system result record; When the target coordinate system output configuration record includes a relative encounter coordinate system, for the relative encounter coordinate system output, the corrected orbital state record corresponding to each target prediction time and the relative reference state of the corresponding target object are read one by one; based on the position difference and velocity difference between the corrected orbital state record and the relative reference state, the relative position state and approach direction state in the relative encounter plane are regenerated according to the construction rules of the relative encounter plane; then, the relative position state, approach direction state and corresponding time identifier are associated and recorded to generate the relative encounter coordinate system result record; After generating the inertial coordinate system result records, ground-fixed coordinate system result records, local orbit coordinate system result records, load coordinate system result records, and relative encounter coordinate system result records corresponding to each target prediction time, the coordinate system result records generated at the same target prediction time are collected and organized to generate the re-output result record corresponding to the current target prediction time. Among them, the results of each coordinate system are all derived from the same corrected orbit state record and share the same time identifier, thereby maintaining the consistency of the re-output results at the same epoch. After generating the re-output result record, a quality assessment is performed on the re-output result record. Specifically, the quality assessment involves: based on the re-output result record corresponding to the current target prediction time, reading the inertial coordinate system results, ground-fixed coordinate system results, local orbital coordinate system results, load coordinate system results, and relative encounter coordinate system results; and calculating the coordinate transformation consistency quality factor corresponding to the current target prediction time. The calculation formula is as follows: ; in, For coordinate transformation consistency quality factor, To achieve overall consistency of residuals, The consistency reference residual threshold; The consistency reference residual threshold is used to set a normalization scale for the comprehensive consistency residual. Its value is determined based on the statistical value of the residuals of historical qualified samples in the current application scenario, the upper limit of the ground calibration residual, or the upper limit of the task allowable deviation. For example, if thirty consecutive pass status samples are selected in the same business scenario, the comprehensive consistency residual corresponding to each sample is calculated, and the average of the thirty comprehensive consistency residuals is calculated. The average value is determined as the consistency reference residual threshold. The threshold is set to 0.08 when the business scenario is collision avoidance warning; 0.12 when the business scenario is inter-satellite link maintenance; 0.15 when the business scenario is payload pointing; and 0.2 when the business scenario is ground coverage analysis. Preset quality judgment threshold This is used to determine the release of coordinate transformation consistency quality factors; its value is determined based on the task risk level. The quality judgment threshold is used to perform output judgment on the coordinate transformation consistency quality factor. The quality judgment threshold is determined according to the task risk level. Initially, the quality judgment threshold is set to 0.92 when the business scenario is collision avoidance warning; 0.88 when the business scenario is inter-satellite link maintenance; 0.85 when the business scenario is payload pointing; and 0.8 when ground coverage analysis is performed. The threshold is adjusted according to actual operation, specifically: after the system enters actual operation, based on the results of the most recent period, through recording and review... The kernel record adjusts the quality judgment threshold by counting the number of target forecasts marked as "passing", the number of target forecasts marked as "low confidence", and the number of misjudged results confirmed by manual review or high-precision supplementary propagation in the most recent period. When the proportion of misjudged results in the most recent period reaches a preset correction condition, the quality judgment threshold is adjusted incrementally. Specifically, when the number of low confidence results confirmed as usable after review exceeds 10%, the quality judgment threshold is lowered by a preset step size; when the number of passing results confirmed as unusable after review exceeds a preset proportion, the quality judgment threshold is raised by a preset step size. The preset step size is set to 0.02. The coordinate transformation consistency quality factor is compared with the quality judgment threshold to determine the confidence level of the corresponding re-output result at the current target prediction time: when When, the re-output result corresponding to the current target prediction time is determined as a high-confidence result; when and When, the re-output result corresponding to the current target prediction time is judged as a low-confidence result; when When this happens, the re-output result corresponding to the current target forecast time will be judged as a failed result; When a result is determined to be highly reliable, the inertial coordinate system result, ground-fixed coordinate system result, local orbit coordinate system result, load coordinate system result, and relative encounter coordinate system result when a relative encounter coordinate system exists, corresponding to the current target prediction time, along with the corresponding time identifier and highly reliable state identifier, are written into the target transformation result, and this result is used as the official output result of the current target prediction time. When a result is determined to be low confidence, the coordinate system results corresponding to the current target forecast time, along with the corresponding time identifier and low confidence status identifier, are written into the target transformation result. At the same time, the comprehensive consistency residual, failure reason code, and residual over-limit type corresponding to the current target forecast time are recorded. Supplementary propagation processing is triggered, which is used to recalculate the orbital state corresponding to the current target forecast time. Before the supplementary propagation result is returned, the current low confidence result is retained as a temporary output result. When a result is determined to be invalid, the re-output result corresponding to the current target forecast time is not written into the target conversion result as the formal output result. Instead, the high-confidence result corresponding to the previous target forecast time is read as the backoff reference result. The backoff reference result, the failure status identifier corresponding to the current target forecast time, and the failure reason code are associated and written into the target conversion result. At the same time, it is sent to a human for review. After determining the credibility level for each target forecast time, the high-credibility results, low-credibility results, and failure results are collected and organized to generate target transformation results. The target transformation results include at least: the multi-coordinate system state results for each target forecast time, the corresponding time identifier, the credibility level identifier, the residual record, and the supplementary propagation state record or backtracking result record in the case of low credibility or failure.

[0016] like Figure 2 The comparison diagram of the effects of this invention and traditional methods is shown in the figure. The black bars in the figure represent the effects of this invention, and the gray bars represent the effects of traditional methods. By synchronously generating multiple coordinate system states based on the same reference orbit state, the consistency of multi-coordinate system result transformation is improved compared with traditional technology. By constructing a first closed-loop link and a second transformation link, comparing the residuals of the starting point and the return state and feeding them back to the orbit state and attitude parameters, the cross-coordinate system closed-loop correction capability is improved compared with traditional technology. By synchronously compensating for the long-term drift of the right ascension of the ascending node and the argument of the perigee during the propagation of the J2 long-term term, and combining the average of the residuals at multiple times with the same epoch and superimposing them on the orbit state, the long-term forecast phase drift compensation capability is improved.

[0017] Example 2: Taking the multi-coordinate system transformation processing task when a low-orbit remote sensing satellite performs an Earth imaging mission as an example, in this scenario, the processing focus is on maintaining the consistency between the position of the ground projection point in the Earth-fixed coordinate system, the axial state in the local orbit coordinate system, and the line-of-sight pointing state in the payload coordinate system. In this example, the target coordinate system type set includes the inertial coordinate system, the Earth-fixed coordinate system, the local orbit coordinate system, and the payload coordinate system, but does not include the relative encounter coordinate system. In this embodiment, a low-Earth orbit remote sensing satellite operating in a sun-synchronous orbit is selected as the target satellite; the starting epoch is set to March 15, 2026, at 10:00:00; the initial orbital elements at the starting epoch are read, wherein the semi-major axis is set to 7078 km, the eccentricity is set to 0.0012, the orbital inclination is set to 98.2°, the right ascension of the ascending node is set to 35.6°, the argument of perigee is set to 12.4°, and the mean perigee angle is set to 78.5°; the initial position-velocity vector corresponding to the starting epoch is read, wherein the three components of the position vector are set to -1265.432. The distances to the target are 5987.214 km and 3512.876 km, respectively, and the three components of the velocity vector are set to -6.241 km / s, -1.958 km / s, and 3.826 km / s. The target prediction time sequence is read, with a prediction duration of 20 minutes and a sampling interval of 10 seconds, resulting in 121 target prediction times from 10:00:00 on March 15, 2026 to 10:20:00 on March 15, 2026. Attitude installation parameters are read, where the installation matrix of the load coordinate system relative to the local orbit coordinate system is set as follows: The first row of data contains 0.999996, -0.002094, and 0.001396; The second row of data contains 0.002096, 0.999997, and -0.000872. The data in the third row is -0.001394, 0.000875, and 0.999998; Read the Earth orientation parameters and Earth rotation parameters. The polar motion parameters are taken as 0.081 arcseconds and 0.267 arcseconds respectively at the initial epoch, and the Earth rotation time difference is taken as 0.124s. Fix the target coordinate system type set as inertial coordinate system, Earth-fixed coordinate system, local orbit coordinate system and load coordinate system. First, the raw orbital data is preprocessed to generate orbital state data: the Coordinated Universal Time (UTC) identifier, on-board time scale, and orbital propagation time identifier are unified, and the starting epoch is determined to be March 15, 2026, 10:00:00. The initial position velocity vector and initial orbital elements are aligned to the same epoch. The 121 target prediction times are sorted to generate 121 propagation time parameters, with propagation time parameters ranging from 0s, 10s, 20s, 30s, up to 1200s. The attitude installation parameters, Earth orientation parameters, and target coordinate system type sets are organized to generate attitude alignment records, orientation parameter records, and coordinate system output configuration records. The coordinate system output order is fixed as inertial coordinate system, Earth-fixed coordinate system, local orbit coordinate system, and payload coordinate system. Then, long-term propagation analysis is performed on the orbital state data to obtain a reference orbital state set: the semi-major axis, eccentricity, orbital inclination, right ascension of the ascending node, argument of perigee, and mean perigee angle are extracted from the initial orbital element record at the initial epoch to generate a reference orbital element set; taking the propagation time parameter of 150s corresponding to the 15th target prediction time as an example, based on the semi-major axis, eccentricity, and orbital inclination in the reference orbital element set, combined with the Earth's gravitational field constant, the propagated right ascension of the ascending node is obtained as 35.5989°, the propagated argument of perigee as 12.4016°, and the propagated mean perigee angle as 79.4112°; then, the propagated orbital elements are restored to the position and velocity vectors in the reference inertial frame to obtain the reference orbital state record corresponding to the 15th target prediction time, where the position vector has three components. The positions are -2184.615km, 5843.227km, and 3338.904km, respectively, with three velocity vector components of -5.987km / s, -1.246km / s, and 4.128km / s. Taking the propagation time parameter of 600s corresponding to the 60th target prediction time as an example, the right ascension of the ascending node after propagation is 35.5958°, the argument of perigee after propagation is 12.4063°, and the mean perigee angle after propagation is 82.1447°. The three components of the recovered position vector are -4321.557km, 4735.682km, and 2548.337km, respectively, with three velocity vector components of -4.761km / s, -2.618km / s, and 5.342km / s, respectively. This generates 121 reference orbit state records. After generating the reference orbit state set, a multi-coordinate system transformation is performed: for each target prediction time, the inertial coordinate system state, Earth-fixed coordinate system state, local orbit coordinate system state, and load coordinate system state are simultaneously generated based on the same reference orbit state record; taking the 15th target prediction time as an example, the reference position vector and reference velocity vector are directly written into the inertial coordinate system state record; based on the Earth's rotation parameters and Earth's orientation parameters, coordinate rotation and velocity transformation are performed on the reference position vector and reference velocity vector to obtain the Earth-fixed coordinate system state record, in which the three components of the position vector in the Earth-fixed coordinate system are 1036.482km, 6178.335km, and 3331.118km, respectively, which are further converted to longitude of 80.4762°, latitude of 28.2184°, and altitude of... 704.511km; Based on the reference position vector and reference velocity vector, a local orbital coordinate axis is constructed, and relevant state quantities are projected onto the local orbital coordinate axis to obtain the local orbital coordinate system state quantity record, in which the radial position quantity is 7074.801km, the tangential velocity quantity is 7.512km / s, and the normal velocity quantity is 0.021km / s; Based on the attitude installation parameters, the local orbital coordinate axis is transformed to the load coordinate system, generating the load coordinate system axial state and line-of-sight pointing state record, in which the lateral deflection angle of the load line-of-sight relative to the local orbital coordinate system is 0.31°, and the pitch deflection angle is -14.26°; Subsequently, the state records of each coordinate system generated at the same target prediction time are collected and organized to obtain the multi-coordinate system synchronous state record corresponding to the 15th target prediction time; Since the target coordinate system type set does not include relative encounter coordinate systems, only the first closed-loop link is constructed: for each target prediction time, a first closed-loop link is constructed from the inertial coordinate system to the Earth-fixed coordinate system, from the Earth-fixed coordinate system to the local orbit coordinate system, and from the local orbit coordinate system back to the inertial coordinate system; taking the 14th, 15th, and 16th target prediction times as examples, after performing coordinate transformations sequentially along the first closed-loop link, the comparison results between the returned state and the starting state of the inertial coordinate system are as follows: the position difference for the 14th target prediction time is 40m, and the direction difference is 0.18°; the position difference for the 15th target prediction time is 42m, and the direction difference is 0.19°; the position difference for the 16th target prediction time is 41m, and the direction difference is 0.17°; the above differences are recorded as the consistency residual set for the corresponding time. After forming a consistent residual set, the correction object determination and object content residual merging are performed. Since only the first closed-loop link is generated in this embodiment, the correction object is only the reference orbit state. Specifically, the position differences of 40m, 42m, and 41m at the 14th, 15th, and 16th target prediction times are taken as the position residual group. After aligning the position residual group with the same epoch, the average is calculated to obtain the position correction residual of 41m. The direction differences of 0.18°, 0.19°, and 0.17° at the 14th, 15th, and 16th target prediction times are taken as the direction residual group. After aligning the direction residual group with the same epoch, the average is calculated to obtain the direction correction residual of 0.18°. The position correction residual of 41m and the direction correction residual of 0.18° are recorded as the orbit state correction residual. Since no relative encounter coordinate system state is generated, no relative position residual, approach direction residual, or axial residual is formed. The attitude parameter correction residual is kept as a null value in this embodiment. After obtaining the orbital state correction residuals, feedback correction is performed: the position correction residuals and orientation correction residuals corresponding to each target's predicted time are read one by one, and the reference orbital state record corresponding to that target's predicted time is read; taking the 15th target's predicted time as an example, the 41m position correction residual is superimposed on the reference position vector, resulting in three components of the corrected reference position vector: -2184.587km, 5843.201km, and 3338.889km; the orientation correction residual is adjusted according to 0.18°, resulting in three components of the corrected reference velocity vector: -5.986km / s, -1.232km / s, and 4.124km / s. The corrected reference position vector and the corrected reference velocity vector are associated with the corresponding time markers and recorded to generate the corrected orbital state record; since no attitude parameter correction residuals are formed in this embodiment, the attitude installation parameter record remains unchanged and together with the corrected orbital state record, forms a consistent correction result; After obtaining the consistency correction results, the re-output is performed: the inertial coordinate system results, ground-fixed coordinate system results, and local orbital coordinate system results are regenerated based on the corrected orbital state record; the load coordinate system results are regenerated based on the original attitude installation parameter record; the regenerated inertial coordinate system results, ground-fixed coordinate system results, local orbital coordinate system results, and load coordinate system results at the same target prediction time are collected and organized to obtain the re-output result record. Taking the predicted time of the 15th target as an example, after re-output, the three components of the position vector in the Earth-fixed coordinate system become 1036.451km, 6178.304km, and 3331.096km, which, after conversion, yield longitude of 80.4757°, latitude of 28.2179°, and altitude of 704.486km; the radial position in the local orbit coordinate system becomes 7074.760km, and the tangential velocity becomes 7.509km / s; the lateral deflection angle of the load line of sight relative to the local orbit coordinate system becomes 0.12°, and the pitch deflection angle becomes -14.11°. After generating the re-output result record, the quality judgment process continues: read the re-output result record corresponding to the current target prediction time, and calculate the coordinate transformation consistency quality factor corresponding to the current target prediction time based on the inertial coordinate system result, the ground-fixed coordinate system result, the local orbit coordinate system result, and the load coordinate system result; where the consistency reference residual threshold is 0.15 and the quality judgment threshold is 0.85. At the 15th target prediction time, before feedback correction, the position difference corresponding to the first closed-loop link was 42m, the direction difference was 0.19°, the deviation of the ground projection point in the corresponding Earth-fixed coordinate system was 68m, the load line-of-sight deviation was 0.23°, and the coordinate transformation consistency quality factor was 0.79. After performing orbital state correction residual write-back, the position difference decreased to 11m, the direction difference decreased to 0.05°, the ground projection point deviation decreased to 18m, the load line-of-sight deviation decreased to 0.07°, and the coordinate transformation consistency quality factor increased to 0.88.

[0018] At the 60th target prediction time, before feedback correction, the position difference corresponding to the first closed-loop link was 39m, the direction difference was 0.16°, the ground projection point deviation in the corresponding Earth-fixed coordinate system was 57m, the load line-of-sight deviation was 0.21°, and the coordinate transformation consistency quality factor was 0.81. After performing orbital state correction residual write-back, the position difference decreased to 13m, the direction difference decreased to 0.06°, the ground projection point deviation decreased to 21m, the load line-of-sight deviation decreased to 0.08°, and the coordinate transformation consistency quality factor increased to 0.87. Since the coordinate transformation consistency quality factor corresponding to the 15th and 60th target prediction times is not less than 0.85 after feedback correction, the inertial coordinate system results, ground-fixed coordinate system results, local orbital coordinate system results, and load coordinate system results corresponding to these two target prediction times, along with the corresponding time identifiers and high-confidence state identifiers, are written into the target transformation results and used as the official output results for the corresponding target prediction times.

[0019] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for transforming multiple coordinate systems, characterized in that, include: Collect raw orbital data of the target satellite, and preprocess the raw orbital data to obtain orbital state data; Feature extraction is performed on the orbital state data to obtain a reference orbital feature set; long-term term propagation analysis is performed on the reference orbital feature set to obtain a reference orbital state set. Based on the reference orbit state set, a multi-coordinate system transformation is performed to obtain a multi-coordinate system synchronous state set; based on the multi-coordinate system synchronous state set, a closed-loop link is constructed to obtain a cross-coordinate system link; Based on the cross-coordinate system link, a correction analysis is performed to obtain the correction residual set; Feedback correction is performed based on the corrected residual set to obtain a consistency correction result; Based on the consistency correction result, re-output is performed to obtain the re-output result record; Based on the re-output result record, a quality judgment process is performed to obtain the target conversion result.

2. The method according to claim 1, characterized in that, Long-term propagation analysis is performed on the reference orbit feature set to obtain the reference orbit state set, including: Based on the J2 reference orbit elements and the Earth's gravitational field constant, the long-term drift was calculated and parameters were updated to obtain the right ascension of the ascending node and the argument of perigee after propagation. Based on the perigee argument parameter, and combined with the semi-major axis and mean angular velocity formulas in the J2 reference orbit elements, time progression and mean perigee increment calculations are performed to obtain the propagated mean perigee angle. Perform the transformation and coordinate rotation of the orbital elements to the reference inertial frame state vector for all propagated ascending nodes and propagated mean anomalies to obtain the reference orbital state set.

3. The method according to claim 1, characterized in that, Based on the reference orbit state set, a multi-coordinate system transformation is performed to obtain a multi-coordinate system synchronization state set, including: Based on the reference orbit state record corresponding to the same target prediction time, the inertial coordinate system state, the ground-fixed coordinate system state, the local orbit coordinate system state, the load coordinate system state, and the relative encounter coordinate system state are generated synchronously to obtain a multi-coordinate system synchronous state set. The multi-coordinate system synchronous state set includes the inertial coordinate system state, the ground-fixed coordinate system state, the local orbit coordinate system state, the load coordinate system state, and the relative encounter coordinate system state.

4. The method according to claim 3, characterized in that, The process includes generating states for inertial coordinate systems, ground-fixed coordinate systems, local orbital coordinate systems, load coordinate systems, and relative encounter coordinate systems, including: Inertial coordinate system state generation: Based on the reference orbit state record, the state record and time stamp are associated and written. Earth-fixed coordinate system state generation: Based on the Earth's rotation parameters and Earth's orientation parameters, perform coordinate rotation and velocity transformation on the reference position vector and reference velocity vector; Local orbit coordinate system state generation: Based on the reference position vector and reference velocity vector, construct the local orbit coordinate axes and project the state variables; Load coordinate system state generation: Based on the attitude installation parameters and local orbit coordinate axes, load coordinate system transformation is performed; Relative encounter coordinate system state generation: Based on the reference orbit state and the relative reference state, construct the relative encounter plane and generate the relative state.

5. The method according to claim 1, characterized in that, Based on the multi-coordinate system synchronization state set, a closed-loop link is constructed to obtain a cross-coordinate system link, including: A cross-coordinate system link is constructed based on a multi-coordinate system synchronization state set, wherein the cross-coordinate system link includes a first closed-loop link and a second transformation link; The first closed-loop link is: from the inertial coordinate system to the ground-fixed coordinate system, from the ground-fixed coordinate system to the local orbit coordinate system, and from the local orbit coordinate system back to the inertial coordinate system; The second transformation link is: from the inertial coordinate system to the local orbit coordinate system, from the local orbit coordinate system to the load coordinate system, and from the load coordinate system to the relative encounter coordinate system.

6. The method according to claim 1, characterized in that, Based on the aforementioned cross-coordinate system link, a corrected residual set is obtained, including: Based on the cross-coordinate system link, the start and return states of the first closed-loop link are compared and the step-by-step transmission states and target states of the second transformation link are compared respectively. The closed-loop difference and transformation difference are calculated and summarized to form a consistent residual set. The consistency residual set is classified and organized to obtain the difference set; the correction object is determined based on the type of the difference set; the object content residual is merged based on the correction object to obtain the correction residual set.

7. The method according to claim 6, characterized in that, Residual merging of object content based on the corrected object includes: Based on the residual set corresponding to the reference orbit state, the position residual set and the relative position residual set are combined and aligned with the epoch, and then the average is calculated to generate the position correction residual. The direction residual set is aligned with the epoch, and then the average is calculated to generate the direction correction residual. The two are combined and recorded as the orbit state correction residual. Based on the residual groups corresponding to the attitude installation parameters, the axial residual groups are aligned and averaged to generate the axial correction difference. The approach direction residual groups are aligned with the same epoch and averaged to generate the approach direction correction difference. The two groups are then combined and recorded as attitude parameter correction residuals. The orbital state correction residual and the attitude parameter correction residual are associated and recorded as a correction residual set.

8. The method according to claim 1, characterized in that, Feedback correction is performed based on the corrected residual set to obtain a consistency correction result, including: Based on the track state correction residual and the reference track state record, position superposition and velocity direction adjustment are performed to generate the corrected track state record; Based on the attitude parameter correction residual and the current attitude installation parameter record, corresponding corrections are performed according to the axial correction difference and the approach direction correction difference to generate an attitude parameter correction record. The corrected orbital state record and attitude parameter correction record are stored as a consistency correction result.

9. The method according to claim 1, characterized in that, Based on the consistency correction result, a re-output is performed to obtain a re-output result record, including: Based on the corrected orbit state and attitude parameter correction records, the inertial coordinate system results, ground-fixed coordinate system results, local orbit coordinate system results, load coordinate system results, and relative encounter coordinate system results corresponding to the predicted time of each target are regenerated. The coordinate system result records under the same target prediction time are collected and organized to obtain the re-output result records.

10. The method according to claim 1, characterized in that, Based on the re-output result record, a quality assessment process is performed to obtain the target transformation result, including: A preset consistency reference residual threshold is set, and the coordinate transformation consistency quality factor is calculated based on the coordinate system results in the re-output results. A preset quality judgment threshold is used to determine the pass rate of the coordinate transformation consistency quality factor based on the quality judgment threshold. Based on the comparison results between the coordinate transformation consistency quality factor and the quality judgment threshold, the reliability level is divided. The re-output result records are processed in a hierarchical manner according to their credibility level to obtain the target transformation result output.

Images