Low earth orbit satellite precision orbit determination method, device and equipment based on machine learning

Abstract

The invention provides a low-orbit satellite precise orbit determination method, device and equipment based on machine learning, and relates to the technical field of earth observation and navigation, and the method comprises the steps: building a satellite-borne GNSS heterogeneous multimode particle swarm model, and carrying out the iterative screening of satellite-borne GNSS observation data, and obtaining the target satellite-borne GNSS observation data of each epoch participating in the low-orbit satellite precise orbit determination calculation; complementing the laser survey data through a deep learning neural network to obtain corresponding target laser survey data in an observable arc section of the laser survey ground station to the low-orbit satellite; estimating pseudo-random pulse parameters through an inverse reinforcement learning model; and performing precise orbit determination according to the target satellite-borne GNSS observation data, the target laser measurement data and the pseudo-random pulse parameters. According to the method, the problems of satellite-borne GNSS data redundancy and lack of laser survey data in rainy and snowy weather are effectively solved, and the uncertainty of pseudo-random pulse parameter selection is avoided, so that the orbit determination precision of the low-orbit satellite is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of earth observation and navigation technology, and in particular to a method, device and equipment for precise orbit determination of low-orbit satellites based on machine learning. Background Art

[0002] With the continuous development of low-orbit satellite communications, satellite remote sensing, low-orbit navigation signal enhancement and other technologies, low-orbit satellites are playing an increasingly prominent role in earth observation, communications, satellite navigation and other fields. High-precision orbit determination is a prerequisite for low-orbit satellites to carry out various tasks. Therefore, precise orbit determination of low-orbit satellites is very important.

[0003] At present, the precise orbit determination technology for low-orbit satellites mainly includes onboard GNSS (Global Navigation Satellite System) orbit determination technology and SLR (Satellite Laser Ranging) orbit determination technology. In terms of orbit determination technology: As multi-mode and multi-frequency observation data continue to increase, the redundancy brought by redundant observations in onboard GNSS orbit determination technology has caused uncertainty in orbit determination accuracy and greatly increased the amount of calculation. Laser satellite measurement has high measurement accuracy, but it will be affected by weather factors such as rain and snow, which will lead to data loss, thus affecting the orbit determination accuracy.

[0004] In addition, the precise orbit determination methods for low-orbit satellites mainly include dynamic orbit determination method, kinematic orbit determination method and simplified dynamic orbit determination method. In terms of orbit determination methods: The dynamic orbit determination method mainly constructs the satellite motion equation based on the mechanical model, and uses the observation data to continuously correct the orbit. This method is greatly affected by the accuracy of the mechanical model. The kinematic orbit determination method completely relies on the geometric information of the observation data for orbit determination. The simplified dynamic orbit determination method is between the dynamic method and the kinematic method. Pseudo-random pulse parameters are added during the orbit determination process to balance the dynamic information and the geometric information of the observation data. Therefore, it has higher orbit determination accuracy, but the setting of the pseudo-random pulse parameters generally adopts empirical values, which is greatly affected by subjective consciousness.

[0005] In summary, the orbit determination accuracy of existing low-orbit satellite precise orbit determination technologies / methods needs to be improved. Summary of the invention

[0006] In view of this, the purpose of the present invention is to provide a method, device and equipment for precise orbit determination of low-orbit satellites based on machine learning, which effectively solves the problems of redundant on-board GNSS data and missing laser satellite measurement data in rainy and snowy weather, and avoids the uncertainty of pseudo-random pulse parameter selection, thereby effectively improving the orbit determination accuracy of low-orbit satellites.

[0007] In a first aspect, the present invention provides a method for precise orbit determination of a low-orbit satellite based on machine learning, comprising: Obtain onboard GNSS observation data and laser satellite measurement data of low-orbit satellite observations; A spaceborne GNSS heterogeneous multi-mode particle swarm model is established for the spaceborne GNSS observation data, and the spaceborne GNSS observation data is iteratively screened through the spaceborne GNSS heterogeneous multi-mode particle swarm model to obtain the target spaceborne GNSS observation data participating in the precise orbit determination of low-orbit satellites at each epoch; Furthermore, the laser satellite measurement data is supplemented by a pre-trained deep learning neural network to obtain the target laser satellite measurement data corresponding to the observable arc of the laser satellite measurement ground station for the low-orbit satellite; and, constructing an action set based on the attribute data corresponding to the low-orbit satellite, and outputting a state set corresponding to the action set through a pre-trained inverse reinforcement learning model, wherein the state set includes pseudo-random pulse parameters corresponding to the low-orbit satellite; The precise orbit determination result corresponding to the low-orbit satellite is determined based on the target satellite-borne GNSS observation data, target laser satellite measurement data and pseudo-random pulse parameters.

[0008] In one implementation, a spaceborne GNSS heterogeneous multi-mode particle swarm model is established for spaceborne GNSS observation data, including: Determine the quality index parameters corresponding to the onboard GNSS observation satellite data, the quality index parameters include multipath index parameters, cycle slip index parameters and signal-to-noise ratio index parameters; Based on multipath index parameters, cycle slip index parameters and signal-to-noise ratio index parameters, a space-borne GNSS heterogeneous multi-mode particle swarm model is established.

[0009] In one implementation, the onboard GNSS observation data is iteratively screened by using the onboard GNSS heterogeneous multi-mode particle swarm model to obtain target onboard GNSS observation data participating in the low-orbit satellite precise orbit determination solution at each epoch, including: For each epoch, the following operations are performed on the onboard GNSS observation data obtained by the low-orbit satellite for multiple GNSS satellite observations: Based on the onboard GNSS observation data obtained by the low-orbit satellite for observing multiple GNSS satellites in the epoch, an initial population corresponding to the epoch is generated, where different particles in the initial population include the onboard GNSS observation data obtained by the low-orbit satellite for observing multiple different GNSS satellites and their corresponding quality indicator parameters; Through the space-borne GNSS heterogeneous multi-mode particle swarm model, the particles contained in the initial population corresponding to the epoch are iteratively screened to obtain the target space-borne GNSS observation data participating in the precise orbit determination of low-orbit satellites at this epoch.

[0010] In one implementation, the particles contained in the initial population corresponding to the epoch are iteratively screened through a space-borne GNSS heterogeneous multi-mode particle swarm model to obtain the target space-borne GNSS observation data for the epoch participating in the low-orbit satellite precise orbit determination solution, including: Determine the fitness value corresponding to the particle based on the quality index parameter corresponding to the onboard GNSS observation data obtained by the low-orbit satellite included in the particle for each GNSS satellite observation; If the fitness value corresponding to the particle is better than the preset individual optimal value and the global optimal value, the individual optimal value and the global optimal value are updated using the fitness value corresponding to the particle; According to the updated individual optimal value and global optimal value, the particle velocity and particle position corresponding to the particle are updated; The next particle is determined according to the updated particle velocity and particle position, and the fitness value corresponding to the next particle is determined based on the quality index parameters corresponding to the onboard GNSS observation data obtained by observing each GNSS satellite of the low-orbit satellite included in the next particle, until the preset iteration stop condition is met, and the target onboard GNSS observation data participating in the low-orbit satellite precise orbit determination solution at this epoch is obtained; wherein the iteration stop condition is that the global optimal value is less than the preset threshold.

[0011] In one embodiment, determining the fitness value corresponding to the particle based on the quality indicator parameter corresponding to the onboard GNSS observation data obtained by observing each low-orbit satellite included in the particle includes: Determine the fitness value corresponding to the particle according to the following formula: ; in, For the Particles The corresponding fitness value is is the weighting factor, For particles Medium and low orbit satellites are aimed at The multipath indicator parameters corresponding to the onboard GNSS observation data obtained by observing GNSS satellites are: For particles Medium and low orbit satellites are aimed at The cycle slip index parameters corresponding to the onboard GNSS observation data obtained by observing GNSS satellites are: For particles Medium and low orbit satellites are aimed at The signal-to-noise ratio index parameter corresponding to the onboard GNSS observation data obtained by observing GNSS satellites.

[0012] In one implementation, the laser satellite measurement data is supplemented by a pre-trained deep learning neural network to obtain target laser satellite measurement data corresponding to the observable arc of the laser satellite measurement ground station for the low-orbit satellite, including: Determine the observable arc of the laser satellite ground station to the low-orbit satellite; Normalizing the laser satellite measurement data to obtain normalized laser satellite measurement data; Through the pre-trained deep learning neural network, based on the normalized laser satellite measurement data, the missing laser satellite measurement data in the observable arc segment is supplemented to obtain the target laser satellite measurement data corresponding to the observable arc segment of the low-orbit satellite by the laser satellite measurement ground station.

[0013] In one embodiment, the action set is constructed based on the low-orbit satellite orbit height, low-orbit satellite mass, low-orbit satellite solar sail area and orbit solution time contained in the attribute data corresponding to the low-orbit satellite; before outputting the state set corresponding to the action set through the pre-trained inverse reinforcement learning model, the method also includes: Output the historical state set corresponding to the historical action set through the inverse reinforcement learning model; Generate expert strategy trajectory based on historical action set and historical state set; The activation function is constructed according to the expert strategy trajectory to train the inverse reinforcement learning model using the activation function. The trained inverse reinforcement learning model is used to output the state set corresponding to the action set. The expression of the activation function is as follows: ; in, is the activation function, For the Expert strategy trajectories, For the The expert strategy trajectory Actions, , For the The expert strategy trajectory , status, is the activation function weight, is a constant, is the state feature, is the state transfer function.

[0014] In one implementation, determining a precise orbit determination result corresponding to a low-orbit satellite based on target satellite-borne GNSS observation data, target laser satellite measurement data, and pseudo-random pulse parameters includes: Constructing the onboard GNSS carrier phase observation equation corresponding to the low-orbit satellite based on the target onboard GNSS observation data; constructing the laser satellite measurement observation equation corresponding to the low-orbit satellite based on the target laser satellite measurement data; and constructing the satellite motion equation corresponding to the low-orbit satellite based on the pseudo-random pulse parameters; The onboard GNSS carrier phase observation equation, laser satellite observation equation and satellite motion equation are combined to determine the precise orbit determination results corresponding to the low-orbit satellite. The precise orbit determination results include the satellite position coordinates and satellite velocity.

[0015] In a second aspect, the present invention further provides a low-orbit satellite precise orbit determination device based on machine learning, comprising: The data acquisition module is used to obtain the onboard GNSS observation data and laser satellite measurement data of low-orbit satellite observations; The GNSS screening module is used to establish a satellite-borne GNSS heterogeneous multi-mode particle swarm model for the satellite-borne GNSS observation data, and iteratively screen the satellite-borne GNSS observation data through the satellite-borne GNSS heterogeneous multi-mode particle swarm model to obtain the target satellite-borne GNSS observation data participating in the precise orbit determination of low-orbit satellites at each epoch; The laser satellite measurement completion module is used to complete the laser satellite measurement data through the deep learning neural network obtained by pre-training, and obtain the corresponding target laser satellite measurement data within the observable arc segment of the laser satellite measurement ground station for the low-orbit satellite; A pseudo-random pulse generation module is used to construct an action set based on the attribute data corresponding to the low-orbit satellite, and output a state set corresponding to the action set through a pre-trained inverse reinforcement learning model, where the state set includes pseudo-random pulse parameters corresponding to the low-orbit satellite; The precise orbit determination module is used to determine the precise orbit determination results corresponding to the low-orbit satellite based on the target satellite-borne GNSS observation data, target laser satellite measurement data and pseudo-random pulse parameters.

[0016] In a third aspect, the present invention further provides an electronic device, comprising a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement any one of the methods provided in the first aspect.

[0017] The present invention provides a method, device and equipment for precise orbit determination of low-orbit satellites based on machine learning. First, the onboard GNSS observation data and laser satellite measurement data of low-orbit satellite observation are obtained; then a onboard GNSS heterogeneous multi-mode particle swarm model is established for the onboard GNSS observation data, and the onboard GNSS observation data is iteratively screened through the onboard GNSS heterogeneous multi-mode particle swarm model to obtain target onboard GNSS observation data participating in the precise orbit determination solution of low-orbit satellites at each epoch; and, the laser satellite measurement data is complemented through a deep learning neural network obtained by pre-training to obtain target laser satellite measurement data corresponding to the observable arc segment of the laser satellite measurement ground station for the low-orbit satellite; and, based on the attribute data corresponding to the low-orbit satellite, an action set is constructed, and a state set corresponding to the action set is output through a pre-trained inverse reinforcement learning model, and the state set includes pseudo-random pulse parameters corresponding to the low-orbit satellite; finally, the precise orbit determination result corresponding to the low-orbit satellite is determined according to the target onboard GNSS observation data, the target laser satellite measurement data and the pseudo-random pulse parameters. The above method uses the onboard GNSS heterogeneous multi-mode particle swarm model to iteratively screen the onboard GNSS observation data, constructs a deep learning neural network model to complete the laser satellite measurement data, and constructs an inverse reinforcement learning model to estimate the pseudo-random pulse parameters. On this basis, the target onboard GNSS observation data and the target laser satellite measurement data are fused, and the pseudo-random pulse parameters are introduced into the satellite motion equation for precise orbit determination of low-orbit satellites. This effectively solves the problems of redundant onboard GNSS data and missing laser satellite measurement data in rainy and snowy weather, and avoids the uncertainty of pseudo-random pulse parameter selection, thereby effectively improving the orbit determination accuracy of low-orbit satellites.

[0018] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0019] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1A schematic diagram of a process flow of a method for precise orbit determination of a low-orbit satellite based on machine learning provided in an embodiment of the present invention; Figure 2 A schematic diagram of the overall process of a method for precise orbit determination of a low-orbit satellite based on machine learning provided in an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a low-orbit satellite precise orbit determination device based on machine learning provided in an embodiment of the present invention; Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described in combination with the embodiments below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] At present, the orbit determination accuracy of existing low-orbit satellite precise orbit determination technologies / methods needs to be improved. Based on this, the present invention provides a method, device and equipment for low-orbit satellite precise orbit determination based on machine learning, which effectively solves the problems of redundant onboard GNSS data and missing laser satellite measurement data in rainy and snowy weather, and avoids the uncertainty of pseudo-random pulse parameter selection, thereby effectively improving the orbit determination accuracy of low-orbit satellites.

[0024] To facilitate understanding of this embodiment, firstly, a method for precise orbit determination of a low-orbit satellite based on machine learning disclosed in an embodiment of the present invention is described in detail. Figure 1 The flowchart of a method for precise orbit determination of a low-orbit satellite based on machine learning is shown, and the method mainly includes the following steps S102 to S110: Step S102, obtaining onboard GNSS observation data and laser satellite measurement data of low-orbit satellite observations.

[0025] Step S104, a satellite-borne GNSS heterogeneous multi-mode particle swarm model is established for the satellite-borne GNSS observation data, and the satellite-borne GNSS heterogeneous multi-mode particle swarm model is used to iteratively screen the satellite-borne GNSS observation data to obtain the target satellite-borne GNSS observation data participating in the low-orbit satellite precise orbit determination solution at each epoch.

[0026] In one example, corresponding quality index parameters, including multipath index parameters, cycle slip index parameters and signal-to-noise ratio index parameters, can be determined based on the onboard GNSS observation data. On this basis, a onboard GNSS heterogeneous multimode particle swarm model is constructed. Through the onboard GNSS heterogeneous multimode particle swarm model, based on the multipath index parameters, cycle slip index parameters and signal-to-noise ratio index parameters, the onboard GNSS observation data obtained by observing multiple GNSS satellites by a low-orbit satellite in each epoch are iteratively screened to obtain the target onboard GNSS observation data participating in the precise orbit determination of the low-orbit satellite in each epoch.

[0027] Step S106, the laser satellite measurement data is supplemented by the pre-trained deep learning neural network to obtain the target laser satellite measurement data corresponding to the observable arc segment of the laser satellite measurement ground station for the low-orbit satellite.

[0028] In one example, the observable arc segment of the low-orbit satellite by the laser satellite ground station is determined. After the laser satellite data is normalized, the pre-trained deep learning neural network is used to complete the missing laser satellite data in the observable arc segment based on the normalized laser satellite data. The corresponding target laser satellite data within the observable arc segment of the low-orbit satellite by the laser satellite ground station can be obtained.

[0029] Step S108, constructing an action set based on the attribute data corresponding to the low-orbit satellite, and outputting a state set corresponding to the action set through a pre-trained inverse reinforcement learning model, wherein the state set includes pseudo-random pulse parameters corresponding to the low-orbit satellite.

[0030] Among them, the attribute data include the low-orbit satellite orbital height, low-orbit satellite mass, low-orbit satellite solar panel area and orbital solution time, that is, the action set is constructed based on the low-orbit satellite orbital height, low-orbit satellite mass, low-orbit satellite solar panel area and orbital solution time contained in the attribute data corresponding to the low-orbit satellite. The action set is input into the pre-trained inverse reinforcement learning model so that the inverse reinforcement learning model outputs a state set corresponding to the action set to realize the estimation of pseudo-random pulse parameters.

[0031] Step S110, determining the precise orbit determination result corresponding to the low-orbit satellite according to the target satellite-borne GNSS observation data, the target laser satellite measurement data and the pseudo-random pulse parameters.

[0032] Among them, the precise orbit determination results include the satellite position coordinates and satellite velocity of the low-orbit satellite. In one example, based on the target satellite-borne GNSS observation data, the target laser satellite measurement data and the pseudo-random pulse parameters, the satellite-borne GNSS carrier phase observation equation, the laser satellite measurement observation equation and the satellite motion equation are constructed respectively. The precise orbit determination results corresponding to the low-orbit satellite can be obtained by combining the above three equations.

[0033] The method for precise orbit determination of low-orbit satellites based on machine learning provided in an embodiment of the present invention utilizes a satellite-borne GNSS heterogeneous multi-mode particle swarm model to iteratively screen satellite-borne GNSS observation data, constructs a deep learning neural network model to complete laser satellite measurement data, constructs an inverse reinforcement learning model to estimate pseudo-random pulse parameters, and on this basis, fuses the target satellite-borne GNSS observation data and the target laser satellite measurement data, introduces pseudo-random pulse parameters into the satellite motion equation to perform precise orbit determination of low-orbit satellites, effectively solves the problems of redundant satellite-borne GNSS data and missing laser satellite measurement data in rainy and snowy weather, and avoids the uncertainty of pseudo-random pulse parameter selection, thereby effectively improving the orbit determination accuracy of low-orbit satellites.

[0034] For ease of understanding, the present invention provides a specific implementation of a method for precise orbit determination of a low-orbit satellite based on machine learning, see Figure 2 The overall process diagram of a method for precise orbit determination of low-orbit satellites based on machine learning is shown, including: (i) obtaining onboard GNSS observation data and laser satellite measurement data of low-orbit satellites; (ii) screening multi-mode and multi-frequency onboard GNSS observation data based on a heterogeneous multi-mode particle swarm algorithm to obtain target onboard GNSS observation data for orbit determination at each epoch; (iii) constructing a deep learning neural network to complete the missing laser satellite measurement data to obtain the target laser satellite measurement data corresponding to the observable arc segment of the laser satellite measurement ground station for the low-orbit satellite; (iv) constructing an inverse reinforcement learning model to estimate pseudo-random pulse parameters; (v) using a simplified dynamics method to perform precise orbit determination of low-orbit satellites based on the target onboard GNSS observation data, target laser satellite measurement data and pseudo-random pulse parameters.

[0035] 1. Obtaining low-orbit satellite-borne GNSS observation data and laser satellite measurement data.

[0036] (ii) Based on the heterogeneous multi-mode particle swarm algorithm, the multi-mode and multi-frequency satellite-borne GNSS observation data are screened to obtain the target satellite-borne GNSS observation data for orbit determination in each epoch.

[0037] (2.1) Determine the quality index parameters corresponding to the onboard GNSS observation satellite data. The quality index parameters include multipath index parameters, cycle slip index parameters and signal-to-noise ratio index parameters.

[0038] In one example, the calculation process of the multipath indicator parameter is as follows: ; In the formula, is the multipath index parameter, is the GNSS satellite PRN number; is the satellite-borne GNSS pseudorange observation value; , They are and The carrier wavelength of the frequency point; is the satellite-borne GNSS carrier observation value; , , They are and The carrier frequency of the frequency point.

[0039] In one example, the calculation process of the cycle slip indicator parameters is as follows: ; ; ; In the formula, Cycle slip mark, 1 means a cycle slip occurs, 0 means no cycle slip occurs, is the GNSS satellite PRN number; There is no geometric observation value; is the MW combined observation value; is the satellite-borne GNSS carrier observation value; is the satellite-borne GNSS pseudorange observation value; , They are and The carrier wavelength of the frequency point; , They are and The carrier frequency of the frequency point; , are two adjacent epoch moments.

[0040] In one example, the calculation process of the signal-to-noise ratio indicator parameter is as follows: ; In the formula, is the signal-to-noise ratio value; is the carrier power density; is the noise power density.

[0041] (2.2) Based on the multipath index parameter, cycle slip index parameter and signal-to-noise ratio index parameter, a satellite-borne GNSS heterogeneous multimode particle swarm model is established. The satellite-borne GNSS observation data are iteratively screened through the satellite-borne GNSS heterogeneous multimode particle swarm model to obtain the target satellite-borne GNSS observation data participating in the precise orbit determination of low-orbit satellites at each epoch.

[0042] For the onboard GNSS observation data obtained by the low-orbit satellite for multiple GNSS satellites in each epoch, the following steps a to b are performed: Step a, based on the onboard GNSS observation data obtained by the low-orbit satellite observing multiple GNSS satellites in the epoch, generate an initial population corresponding to the epoch, and the different particles in the initial population include the onboard GNSS observation data and their corresponding quality index parameters obtained by the low-orbit satellite observing multiple different GNSS satellites.

[0043] In one example, the epoch A total of GNSS satellites, from which 6 GNSS satellite observation data are selected, a scale of The initial population , the initial population Middle The particles are: ; In the formula, , , …, Indicates The data obtained by the LEO satellites for the 1st, 2nd, ..., 6th GNSS satellites (including the onboard GNSS observation data and its corresponding quality index parameters), , , …, Indicates The multipath index parameters corresponding to the onboard GNSS observation data obtained by the low-orbit satellite in each particle for the 1st, 2nd, ..., 6th GNSS satellites are: , , …, Indicates The cycle slip index parameters corresponding to the onboard GNSS observation data obtained by the low-orbit satellite in each particle for the 1st, 2nd, ..., 6th GNSS satellites are: , , …, Indicates The signal-to-noise ratio index parameters corresponding to the onboard GNSS observation data obtained by the low-orbit satellite in each particle for the 1st, 2nd, ..., 6th GNSS satellites.

[0044] In addition, initialize the individual optimal value and the global optimum , and initialize the particle velocity and instance position.

[0045] Step b, through the onboard GNSS heterogeneous multi-mode particle swarm model, iteratively screen the particles contained in the initial population corresponding to the epoch, and obtain the target onboard GNSS observation data participating in the low-orbit satellite precise orbit determination solution for the epoch. Specifically, it includes the following steps b1 to b4: Step b1, determining the fitness value corresponding to the particle based on the quality indicator parameters corresponding to the onboard GNSS observation data obtained by the low-orbit satellite included in the particle for each GNSS satellite observation.

[0046] Determine the fitness value corresponding to the particle according to the following formula: ; in, For the Particles The corresponding fitness value is is a weighting factor, whose value is the PDOP (Position Dilution of Precision) value of the six GNSS satellites that make up the particle. For particles Medium and low orbit satellites are aimed at The multipath indicator parameters corresponding to the onboard GNSS observation data obtained by observing GNSS satellites are: For particles Medium and low orbit satellites are aimed at The cycle slip index parameters corresponding to the onboard GNSS observation data obtained by observing GNSS satellites are: For particles Medium and low orbit satellites are aimed at The signal-to-noise ratio index parameter corresponding to the onboard GNSS observation data obtained by observing GNSS satellites.

[0047] Step b2: If the fitness value corresponding to the particle is better than the preset individual optimal value and the global optimal value, the individual optimal value and the global optimal value are updated using the fitness value corresponding to the particle.

[0048] In one example, the particle fitness value is compared with the individual optimal value. If the particle fitness value is better than the individual optimal value, the particle fitness value is updated to the individual optimal value; the particle fitness value is compared with the global optimal value. If the particle fitness value is better than the global optimal value, the particle fitness value is updated to the global optimal value.

[0049] Step b3: update the particle speed and particle position corresponding to the particle according to the updated individual optimal value and the global optimal value.

[0050] In one example, the particle velocity and particle position can be updated according to the following function: ; In the formula, is the number of iterations, Indicates GNSS satellites; is the weight factor; is the particle velocity; is the particle position; and is a random factor with a value between between.

[0051] Step b4, determine the next particle according to the updated particle velocity and particle position, repeat steps b1 to b3 until the preset iteration stop condition is met, and obtain the target satellite-borne GNSS observation data participating in the low-orbit satellite precise orbit determination solution at this epoch.

[0052] The iteration stop condition is that the global optimal value is less than the preset threshold. For example, the iteration condition is set to .

[0053] (III) Construct a deep learning neural network to complete the missing laser satellite measurement data and obtain the target laser satellite measurement data corresponding to the observable arc of the laser satellite measurement ground station for the low-orbit satellite.

[0054] (3.1) Determine the observable arc of the laser satellite ground station for the low-orbit satellite.

[0055] In one example, the LEO satellite orbit determination solution is performed based on the laser satellite measurement data to obtain the LEO satellite position coordinates; the LEO satellite altitude angle relative to the laser satellite measurement ground station is calculated based on the laser satellite measurement ground station coordinates and the LEO satellite position coordinates, and the cut-off altitude angle is set to 10 degrees, thereby obtaining the observable arc of the laser satellite measurement ground station for the LEO satellite. .

[0056] (3.2) The laser satellite measurement data is normalized to obtain normalized laser satellite measurement data.

[0057] In one method, the laser satellite data is normalized according to the following formula: ; In the formula is the normalized laser satellite data time series, ; is the original sequence, ; Expressing expectation.

[0058] (3.3) The pre-trained deep learning neural network is used to complete the missing laser satellite data within the observable arc based on the normalized laser satellite data, and the corresponding target laser satellite data within the observable arc of the laser satellite ground station for the low-orbit satellite is obtained.

[0059] In one example, a single-layer deep learning neural network can be constructed to complete laser satellite data: ; In the formula, is the predicted value; , , All are weight matrices; For memory variables; is the deviation value; is the neuron activation function, and the Relu function is used.

[0060] Furthermore, before using the above single-layer deep learning neural network to complete the laser satellite data, it is necessary to conduct pre-training. In the above neural network training, there is a deviation value, which is represented by the mean square error: ; In the formula, is the number of samples for training; is the true value of laser satellite measurement; is the predicted value of laser detection.

[0061] The deviation value is used as the training end constraint, and its threshold is set to 0.999. If the deviation value is less than or equal to the threshold, it is considered to have converged and the training ends.

[0062] (iv) Construct an inverse reinforcement learning model to estimate the parameters of pseudo-random pulses.

[0063] First, explain the state set and action set: State Collection : , that is, the state set By pseudo-random pulse parameters composition; Action Collection : . Elements of the action set It is determined by the low-orbit satellite orbital height, satellite mass, satellite solar sail area, and orbit solution time. The calculation formula is: ; ; In the formula, The mass of a low-orbit satellite; is the orbital height of the low-orbit satellite; is the solar sail area of ​​low-orbit satellite; is a constant coefficient determined by the orbit solution time; is the orbit solution time, expressed in days per year.

[0064] In the above state set and action set Based on this, an embodiment of the present invention provides a specific process for constructing an inverse reinforcement learning model, including: (4.1) The inverse reinforcement learning model outputs the set of historical states corresponding to the set of historical actions.

[0065] (4.2) Generate the expert strategy trajectory based on the historical action set and the historical state set. The expression of the expert strategy trajectory is as follows: ; In the formula, is the expert strategy trajectory.

[0066] (4.3) Construct an activation function based on the expert strategy trajectory to use the activation function to train the inverse reinforcement learning model. The trained inverse reinforcement learning model is used to output the state set corresponding to the action set.

[0067] The expression of the activation function is as follows: ; in, is the activation function, For the Expert strategy trajectories, For the The expert strategy trajectory Actions, , For the The expert strategy trajectory , status, is the activation function weight, is a constant, is the state feature, is the state transfer function.

[0068] The final excitation function is obtained by iterating all expert trajectories. According to the excitation function, a state set consisting of pseudo-random pulse parameters can be output for the input action set, thereby estimating the pseudo-random pulse parameters required for trajectory solution. .

[0069] (V) Based on the target satellite-borne GNSS observation data, target laser satellite measurement data and pseudo-random pulse parameters, a simplified dynamic method is used to perform precise orbit determination of low-orbit satellites.

[0070] (5.1) Construct the onboard GNSS carrier phase observation equation corresponding to the low-orbit satellite based on the target onboard GNSS observation data; and construct the laser satellite observation equation corresponding to the low-orbit satellite based on the target laser satellite data; and construct the satellite motion equation corresponding to the low-orbit satellite based on the pseudo-random pulse parameters.

[0071] Construct the onboard GNSS carrier phase observation equation: ; In the formula, is the GNSS carrier phase observation value onboard the low-orbit satellite; is the geometric distance from the low-orbit satellite to the GNSS satellite; is the tropospheric mapping function; is the tropospheric zenith wet delay; is the ionospheric delay; is the carrier wavelength; is the GNSS phase ambiguity onboard the low-orbit satellite; is the onboard GNSS carrier phase noise; is the speed of light in vacuum; is the GNSS satellite clock error; is the clock error of the GNSS receiver onboard the low-orbit satellite.

[0072] Construct the laser satellite observation equation: ; In the formula, It is the laser satellite observation value of low-orbit satellite; is the geometric distance between the laser reflection prism of the low-orbit satellite and the laser satellite measurement ground station; Tropospheric delay correction; It is the correction value between the mass center of the low-orbit satellite and the geometric center of the laser reflecting prism; Correction for eccentricity of laser satellite ground station; To measure noise.

[0073] Construct the satellite's equations of motion: ; In the formula, is the acceleration of the low-orbit satellite in the inertial system; is the position of the low-orbit satellite in the inertial system; is the Earth's gravitational constant; is the perturbed acceleration.

[0074] Furthermore, the pseudo-random pulse parameter is added to the above satellite motion equation , in order to absorb the disturbance acceleration caused by the non-conservative perturbation force.

[0075] (5.2) Combine the onboard GNSS carrier phase observation equation, the laser satellite observation equation and the satellite motion equation to determine the precise orbit determination results corresponding to the low-orbit satellite. The precise orbit determination results include the satellite position coordinates and satellite velocity.

[0076] In one example, the satellite motion equation, the onboard GNSS carrier phase observation equation, and the laser satellite observation equation are combined to estimate the position coordinates of the low-orbit satellite using the least squares method. and speed .

[0077] In summary, satellite-borne GNSS observation data is developing in the direction of multi-mode and multi-frequency. How to reduce the amount of calculation while ensuring accuracy is an issue to be considered; laser satellite measurement data has high-precision measurement accuracy, but data loss is prone to occur due to weather influences; the determination of pseudo-random pulse parameters in the simplified dynamics method is often based on empirical values ​​and cannot accurately adapt to each satellite. In response to the above problems, an embodiment of the present invention proposes a method for precise orbit determination of low-orbit satellites based on machine learning. First, the satellite-borne GNSS observation data is screened based on a heterogeneous multi-mode particle swarm algorithm to obtain the satellite-borne GNSS observation data involved in the calculation; then a deep learning neural network is constructed to supplement the missing laser satellite measurement data due to weather reasons; then an inverse reinforcement learning model is constructed to estimate the pseudo-random pulse parameters; finally, precise orbit determination of low-orbit satellites is achieved based on the simplified dynamics method, which can effectively solve the above problems.

[0078] Based on the above embodiments, the present invention provides a low-orbit satellite precise orbit determination device based on machine learning. Figure 3 The structure diagram of a low-orbit satellite precise orbit determination device based on machine learning is shown in FIG. The device mainly includes the following parts: The data acquisition module 302 is used to acquire the onboard GNSS observation data and laser satellite measurement data of low-orbit satellite observations; The GNSS screening module 304 is used to establish a satellite-borne GNSS heterogeneous multi-mode particle swarm model for the satellite-borne GNSS observation data, and iteratively screen the satellite-borne GNSS observation data through the satellite-borne GNSS heterogeneous multi-mode particle swarm model to obtain the target satellite-borne GNSS observation data participating in the low-orbit satellite precise orbit determination solution at each epoch; The laser satellite measurement and completion module 306 is used to complete the laser satellite measurement and completion data through the deep learning neural network obtained by pre-training, and obtain the target laser satellite measurement and completion data corresponding to the observable arc segment of the laser satellite measurement and completion ground station for the low-orbit satellite; A pseudo-random pulse generation module 308 is used to construct an action set based on the attribute data corresponding to the low-orbit satellite, and output a state set corresponding to the action set through a pre-trained inverse reinforcement learning model, where the state set includes pseudo-random pulse parameters corresponding to the low-orbit satellite; The precise orbit determination module 310 is used to determine the precise orbit determination result corresponding to the low-orbit satellite based on the target satellite-borne GNSS observation data, the target laser satellite measurement data and the pseudo-random pulse parameters.

[0079] The low-orbit satellite precise orbit determination device based on machine learning provided by the embodiment of the present invention uses the on-board GNSS heterogeneous multi-mode particle swarm model to iteratively screen the on-board GNSS observation data, constructs a deep learning neural network model to complete the laser satellite measurement data, constructs an inverse reinforcement learning model to estimate pseudo-random pulse parameters, and on this basis, fuses the target on-board GNSS observation data and the target laser satellite measurement data, introduces pseudo-random pulse parameters into the satellite motion equation to perform precise orbit determination of the low-orbit satellite, effectively solves the problems of redundant on-board GNSS data and missing laser satellite measurement data in rainy and snowy weather, and avoids the uncertainty of pseudo-random pulse parameter selection, thereby effectively improving the orbit determination accuracy of the low-orbit satellite.

[0080] In one embodiment, the GNSS screening module 304 includes a heterogeneous multi-modal particle swarm model unit for: Determine the quality index parameters corresponding to the onboard GNSS observation satellite data, the quality index parameters include multipath index parameters, cycle slip index parameters and signal-to-noise ratio index parameters; Based on multipath index parameters, cycle slip index parameters and signal-to-noise ratio index parameters, a space-borne GNSS heterogeneous multi-mode particle swarm model is established.

[0081] In one implementation, the GNSS screening module 304 includes a data iteration screening unit configured to: For each epoch, the following operations are performed on the onboard GNSS observation data obtained by the low-orbit satellite for multiple GNSS satellite observations: Based on the onboard GNSS observation data obtained by the low-orbit satellite for observing multiple GNSS satellites in the epoch, an initial population corresponding to the epoch is generated, where different particles in the initial population include the onboard GNSS observation data obtained by the low-orbit satellite for observing multiple different GNSS satellites and their corresponding quality indicator parameters; Through the space-borne GNSS heterogeneous multi-mode particle swarm model, the particles contained in the initial population corresponding to the epoch are iteratively screened to obtain the target space-borne GNSS observation data participating in the precise orbit determination of low-orbit satellites at this epoch.

[0082] In one embodiment, the data iteration screening unit includes: The fitness value determination subunit is used to: determine the fitness value corresponding to the particle based on the quality index parameter corresponding to the onboard GNSS observation data obtained by the low-orbit satellite included in the particle for each GNSS satellite observation; The optimal value updating subunit is used to update the individual optimal value and the global optimal value using the fitness value corresponding to the particle if the fitness value corresponding to the particle is better than the preset individual optimal value and the global optimal value; A speedometer position updating subunit, used for updating the particle speed and particle position corresponding to the particle according to the updated individual optimal value and the global optimal value; The iterative subunit is used to determine the next particle according to the updated particle velocity and particle position, and determine the fitness value corresponding to the next particle based on the quality indicator parameters corresponding to the onboard GNSS observation data obtained by the low-orbit satellite included in the next particle for each GNSS satellite observation, until the preset iteration stop condition is met, and the target onboard GNSS observation data participating in the low-orbit satellite precise orbit determination solution at this epoch is obtained; wherein the iteration stop condition is that the global optimal value is less than a preset threshold.

[0083] In one implementation, the fitness value determination subunit is specifically used for: Determine the fitness value corresponding to the particle according to the following formula: ; in, For the Particles The corresponding fitness value is is the weighting factor, For particles Medium and low orbit satellites are aimed at The multipath indicator parameters corresponding to the onboard GNSS observation data obtained by observing GNSS satellites are: For particles Medium and low orbit satellites are aimed at The cycle slip index parameters corresponding to the onboard GNSS observation data obtained by observing GNSS satellites are: For particles Medium and low orbit satellites are aimed at The signal-to-noise ratio index parameter corresponding to the onboard GNSS observation data obtained by observing GNSS satellites.

[0084] In one implementation, the laser detection and protection completion module 306 is specifically used to: Determine the observable arc of the laser satellite ground station to the low-orbit satellite; Normalizing the laser satellite measurement data to obtain normalized laser satellite measurement data; Through the pre-trained deep learning neural network, based on the normalized laser satellite measurement data, the missing laser satellite measurement data in the observable arc segment is supplemented to obtain the target laser satellite measurement data corresponding to the observable arc segment of the low-orbit satellite by the laser satellite measurement ground station.

[0085] In one embodiment, the action set is constructed based on the low-orbit satellite orbit height, low-orbit satellite mass, low-orbit satellite solar sail area and orbit solution time contained in the attribute data corresponding to the low-orbit satellite; and also includes an inverse reinforcement learning model training module for: Output the historical state set corresponding to the historical action set through the inverse reinforcement learning model; Generate expert strategy trajectory based on historical action set and historical state set; The activation function is constructed according to the expert strategy trajectory to train the inverse reinforcement learning model using the activation function. The trained inverse reinforcement learning model is used to output the state set corresponding to the action set. The expression of the activation function is as follows: ; in, is the activation function, For the Expert strategy trajectories, For the The expert strategy trajectory Actions, , For the The expert strategy trajectory , status, is the activation function weight, is a constant, is the state feature, is the state transfer function.

[0086] In one embodiment, the precise orbit determination module 310 is specifically used for: Constructing the onboard GNSS carrier phase observation equation corresponding to the low-orbit satellite based on the target onboard GNSS observation data; constructing the laser satellite measurement observation equation corresponding to the low-orbit satellite based on the target laser satellite measurement data; and constructing the satellite motion equation corresponding to the low-orbit satellite based on the pseudo-random pulse parameters; The onboard GNSS carrier phase observation equation, laser satellite observation equation and satellite motion equation are combined to determine the precise orbit determination results corresponding to the low-orbit satellite. The precise orbit determination results include the satellite position coordinates and satellite velocity.

[0087] The device provided in the embodiment of the present invention has the same implementation principle and technical effects as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference may be made to the corresponding contents in the aforementioned method embodiment.

[0088] An embodiment of the present invention provides an electronic device. Specifically, the electronic device includes a processor and a storage device. The storage device stores a computer program, and when the computer program is executed by the processor, it executes the method described in any one of the above-mentioned implementation methods.

[0089] Figure 4A structural diagram of an electronic device provided in an embodiment of the present invention, the electronic device 100 includes: a processor 40, a memory 41, a bus 42 and a communication interface 43, wherein the processor 40, the communication interface 43 and the memory 41 are connected via the bus 42; the processor 40 is used to execute an executable module stored in the memory 41, such as a computer program.

[0090] The memory 41 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 43 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0091] The bus 42 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0092] Among them, the memory 41 is used to store programs, and the processor 40 executes the program after receiving the execution instruction. The method executed by the device for flow process definition disclosed in any embodiment of the above-mentioned embodiment of the present invention can be applied to the processor 40 or implemented by the processor 40.

[0093] The processor 40 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 40. The above processor 40 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiment of the present invention can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 41, and the processor 40 reads the information in the memory 41 and completes the steps of the above method in combination with its hardware.

[0094] The computer program product of the readable storage medium provided in the embodiment of the present invention includes a computer-readable storage medium storing program code, and the instructions included in the program code can be used to execute the methods described in the previous method embodiments. The specific implementation can be referred to the previous method embodiments, which will not be repeated here.

[0095] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0096] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for precise orbit determination of low-orbit satellites based on machine learning, characterized in that: include: Obtain onboard GNSS observation data and laser satellite measurement data of low-orbit satellite observations; A satellite-borne GNSS heterogeneous multi-mode particle swarm model is established for the satellite-borne GNSS observation data, and the satellite-borne GNSS observation data is iteratively screened by the satellite-borne GNSS heterogeneous multi-mode particle swarm model to obtain target satellite-borne GNSS observation data participating in the precise orbit determination solution of the low-orbit satellite at each epoch; And, the laser satellite measurement data is supplemented by a pre-trained deep learning neural network to obtain the target laser satellite measurement data corresponding to the observable arc segment of the laser satellite measurement ground station for the low-orbit satellite; And, constructing an action set based on the attribute data corresponding to the low-orbit satellite, and outputting a state set corresponding to the action set through a pre-trained inverse reinforcement learning model, wherein the state set includes pseudo-random pulse parameters corresponding to the low-orbit satellite; The precise orbit determination result corresponding to the low-orbit satellite is determined according to the target satellite-borne GNSS observation data, the target laser satellite measurement data and the pseudo-random pulse parameters.

2. The method for precise orbit determination of low-orbit satellites based on machine learning according to claim 1, characterized in that: A spaceborne GNSS heterogeneous multi-mode particle swarm model is established for the spaceborne GNSS observation data, including: Determine quality index parameters corresponding to the onboard GNSS observation satellite data, wherein the quality index parameters include a multipath index parameter, a cycle slip index parameter, and a signal-to-noise ratio index parameter; Based on the multipath index parameter, the cycle slip index parameter and the signal-to-noise ratio index parameter, a space-borne GNSS heterogeneous multi-mode particle swarm model is established.

3. The method for precise orbit determination of low-orbit satellites based on machine learning according to claim 2, characterized in that: The onboard GNSS observation data are iteratively screened by the onboard GNSS heterogeneous multi-mode particle swarm model to obtain target onboard GNSS observation data participating in the low-orbit satellite precise orbit determination solution at each epoch, including: The following operations are performed for the onboard GNSS observation data obtained by observing multiple GNSS satellites by a low-orbit satellite in each epoch: Based on the onboard GNSS observation data obtained by the low-orbit satellite observing multiple GNSS satellites at the epoch, generating an initial population corresponding to the epoch, wherein different particles in the initial population include the onboard GNSS observation data obtained by the low-orbit satellite observing multiple different GNSS satellites and the corresponding quality indicator parameters; The particles contained in the initial population corresponding to the epoch are iteratively screened through the onboard GNSS heterogeneous multi-mode particle swarm model to obtain the target onboard GNSS observation data participating in the low-orbit satellite precise orbit determination solution at the epoch.

4. The method for precise orbit determination of low-orbit satellites based on machine learning according to claim 3, characterized in that: The particles contained in the initial population corresponding to the epoch are iteratively screened by the onboard GNSS heterogeneous multi-mode particle swarm model to obtain the target onboard GNSS observation data participating in the low-orbit satellite precise orbit determination solution for the epoch, including: Determining an adaptation value corresponding to the particle based on the quality indicator parameter corresponding to the onboard GNSS observation data obtained by the low-orbit satellite included in the particle for each of the GNSS satellites; If the fitness value corresponding to the particle is better than the preset individual optimal value and the global optimal value, the individual optimal value and the global optimal value are updated using the fitness value corresponding to the particle; According to the updated individual optimal value and the global optimal value, updating the particle speed and particle position corresponding to the particle; The next particle is determined according to the updated particle velocity and the particle position, and the fitness value corresponding to the next particle is determined based on the quality indicator parameter corresponding to the onboard GNSS observation data obtained by the low-orbit satellite included in the next particle for each of the GNSS satellite observations, until a preset iteration stop condition is met, and the target onboard GNSS observation data participating in the low-orbit satellite precise orbit determination solution at this epoch is obtained; wherein the iteration stop condition is that the global optimal value is less than a preset threshold.

5. The method for precise orbit determination of low-orbit satellites based on machine learning according to claim 4, characterized in that: Determining the fitness value corresponding to the particle based on the quality indicator parameter corresponding to the onboard GNSS observation data obtained by observing each of the low-orbit satellites included in the particle, comprises: The fitness value corresponding to the particle is determined according to the following formula: ； in, For the Particles The corresponding fitness value is is the weighting factor, For particles The low-orbit satellite mentioned in the A multipath indicator parameter corresponding to the onboard GNSS observation data obtained by observing the GNSS satellites, For particles The low-orbit satellite mentioned in the A cycle slip index parameter corresponding to the onboard GNSS observation data obtained by observing the GNSS satellites, For particles The low-orbit satellite mentioned in the The signal-to-noise ratio index parameter corresponding to the onboard GNSS observation data obtained by observing the GNSS satellites.

6. The method for precise orbit determination of low-orbit satellites based on machine learning according to claim 1, characterized in that: The laser satellite measurement data is supplemented by a pre-trained deep learning neural network to obtain target laser satellite measurement data corresponding to the observable arc of the laser satellite measurement ground station for the low-orbit satellite, including: Determine the observable arc of the laser satellite ground station to the low-orbit satellite; Normalizing the laser satellite measurement data to obtain normalized laser satellite measurement data; Through the deep learning neural network obtained by pre-training, based on the normalized laser satellite measurement data, the missing laser satellite measurement data in the observable arc segment is supplemented to obtain the target laser satellite measurement data corresponding to the observable arc segment of the low-orbit satellite by the laser satellite measurement ground station.

7. The method for precise orbit determination of low-orbit satellites based on machine learning according to claim 1, characterized in that: The action set is constructed based on the low-orbit satellite orbit height, low-orbit satellite mass, low-orbit satellite solar sail panel area and orbit solution time contained in the attribute data corresponding to the low-orbit satellite; Before outputting the state set corresponding to the action set through the pre-trained inverse reinforcement learning model, the method further includes: Output the historical state set corresponding to the historical action set through the inverse reinforcement learning model; Generate an expert strategy trajectory based on the historical action set and the historical state set; An activation function is constructed according to the expert strategy trajectory to train the inverse reinforcement learning model using the activation function. The trained inverse reinforcement learning model is used to output a state set corresponding to the action set. The expression of the activation function is as follows: ； in, is the activation function, For the Expert strategy trajectories, For the The expert strategy trajectory Actions, , For the The expert strategy trajectory , status, is the activation function weight, is a constant, is the state feature, is the state transfer function.

8. The method for precise orbit determination of low-orbit satellites based on machine learning according to claim 1, characterized in that: Determining a precise orbit determination result corresponding to the low-orbit satellite according to the target onboard GNSS observation data, the target laser satellite measurement data and the pseudo-random pulse parameters, including: Constructing a satellite-borne GNSS carrier phase observation equation corresponding to the low-orbit satellite according to the target satellite-borne GNSS observation data; and constructing a laser satellite measurement observation equation corresponding to the low-orbit satellite according to the target laser satellite measurement data; and constructing a satellite motion equation corresponding to the low-orbit satellite according to the pseudo-random pulse parameters; The onboard GNSS carrier phase observation equation, the laser satellite measurement observation equation and the satellite motion equation are combined to determine the precise orbit determination result corresponding to the low-orbit satellite, wherein the precise orbit determination result includes the satellite position coordinates and the satellite velocity.

9. A low-orbit satellite precise orbit determination device based on machine learning, characterized in that: include: The data acquisition module is used to obtain the onboard GNSS observation data and laser satellite measurement data of low-orbit satellite observations; A GNSS screening module is used to establish a satellite-borne GNSS heterogeneous multi-mode particle swarm model for the satellite-borne GNSS observation data, and iteratively screen the satellite-borne GNSS observation data through the satellite-borne GNSS heterogeneous multi-mode particle swarm model to obtain target satellite-borne GNSS observation data participating in the precise orbit determination solution of low-orbit satellites at each epoch; A laser satellite measurement and completion module is used to complete the laser satellite measurement and completion data through a pre-trained deep learning neural network to obtain the target laser satellite measurement and completion data corresponding to the observable arc segment of the laser satellite measurement and completion ground station for the low-orbit satellite; A pseudo-random pulse generation module, used to construct an action set based on the attribute data corresponding to the low-orbit satellite, and output a state set corresponding to the action set through a pre-trained inverse reinforcement learning model, wherein the state set includes pseudo-random pulse parameters corresponding to the low-orbit satellite; The precise orbit determination module is used to determine the precise orbit determination result corresponding to the low-orbit satellite according to the target satellite-borne GNSS observation data, the target laser satellite measurement data and the pseudo-random pulse parameters.

10. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the method according to any one of claims 1 to 8.

Images