A space-based target orbit determination method with angle-only measurement and compensation for orbit position error

Abstract

The present invention discloses a method for determining a space-based target orbit using only angle measurement, which compensates for orbital position errors and belongs to the field of satellite application measurement technology. The method comprises: establishing an error model for introducing measurement errors into the empirical orbital dynamics model based on the target's true orbital dynamics model; training an initial neural network based on preset training samples to obtain an orbital position error estimation model; wherein the training samples are determined using historical orbital data and the empirical orbital dynamics model; and compensating the error model based on the orbital position error estimation model, processing the initial state of the target to be measured using the deviation-compensated least squares method, and obtaining an optimal estimate of the target's orbit to be measured. The present invention can effectively improve the accuracy of determining the target's orbit.

Description

Technical Field

[0001] The present invention relates to the field of satellite application measurement technology, and in particular to a space-based angle-only target orbit determination method for compensating for orbit position errors. Background Art

[0002] Space-based observation platforms, equipped with optical cameras and leveraging their low size, power consumption, and long detection range, can detect, locate, and track space targets from a distance, enhancing their ability to perceive space situations. When the observation platform is far from the target, it can only obtain relative angular data of the target. Therefore, research is needed to determine the orbit of the target using this relative angular data.

[0003] In related technologies, when estimating the target orbit, it is necessary to accumulate measurement data for a period of time before the target orbit determination result can converge. Random errors and systematic errors will occur in this process, and the systematic error will often introduce constant deviations into the orbit determination result, thereby resulting in a decrease in the accuracy of space target orbit determination.

[0004] Based on this, there is an urgent need for a space-based target orbit determination method that only measures angles and compensates for orbital position errors to solve the above technical problems. Summary of the Invention

[0005] This invention provides a space-based method for determining target orbits using only angle measurement, which compensates for orbital position errors. This method addresses the problem of compensating for orbital position estimation errors during long-term cumulative observations in related technologies. The technical solution is as follows:

[0006] In one aspect, a method for determining a space-based target orbit using only angle measurement and compensating for orbital position error is provided, the method comprising:

[0007] According to the real orbital dynamics model of the target, an error model is established to introduce measurement error into the empirical orbital dynamics model.

[0008] Training an initial neural network based on preset training samples to obtain an orbit position error estimation model; wherein the training samples are determined by historical orbit data and an empirical orbit dynamics model;

[0009] According to the orbit position error estimation model, the error model is compensated, and the initial state of the target to be measured is processed using the deviation compensation least square method to obtain the optimal estimation value of the orbit of the target to be measured.

[0010] On the other hand, a space-based target orbit determination device with only angle measurement and compensation for orbit position error is provided, the device comprising:

[0011] A modeling module is used to establish an error model for introducing measurement errors into the empirical orbital dynamics model based on the target's real orbital dynamics model;

[0012] A training module, configured to train an initial neural network based on preset training samples to obtain an orbital position error estimation model; wherein the training samples are determined by using historical orbital data and an empirical orbital dynamics model;

[0013] The processing module is used to compensate the error model according to the orbit position error estimation model, and process the initial state of the target to be measured using the deviation compensation least square method to obtain the optimal estimate of the orbit of the target to be measured.

[0014] On the other hand, a computer device is provided, which includes a memory and a processor, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the steps of the above-mentioned method for determining the orbit of a space-based target with only angle measurement and compensating for orbital position errors.

[0015] On the other hand, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned method for determining the orbit of a space-based target with only angle measurement and compensating for orbital position errors are implemented.

[0016] On the other hand, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for determining a space-based target orbit with only angle measurement and compensating for orbital position errors.

[0017] The technical solution provided by the present invention can bring at least the following beneficial effects: first, an error model combining the empirical orbital dynamics model error and the measurement error is established, and an initial error estimation network based on deep learning is designed; then, the initial error estimation network is trained using preset training samples to obtain an orbital position error estimation model; finally, a hybrid dynamics model combining the orbital position error estimation model and the empirical orbital dynamics model is proposed, and a deviation-compensated least squares orbit determination method is designed to determine the optimal estimate of the target orbit. This method addresses the problem of errors between the empirical orbital dynamics model of the target and the target's true motion model, which "deviates" the target's initial state estimate. It proposes calculating the target orbit using a hybrid dynamics model combining an empirical orbital dynamics model and a neural network error estimation. This solves the problem of compensating for orbital position estimation errors under long-term cumulative observations when determining the target orbit using relative angle measurement data, thereby improving the accuracy of target orbit determination. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is a flow chart of a method for determining a target orbit by measuring only angles and compensating for orbital position errors provided by one embodiment of the present invention;

[0020] Figure 2 This is a schematic diagram of error-induced trajectory estimation provided by an embodiment of the present invention;

[0021] Figure 3 is a schematic diagram of the initial neural network structure provided by one embodiment of the present invention;

[0022] Figure 4 This is a flow chart of a least squares trajectory determination method with deviation compensation provided by one embodiment of the present invention;

[0023] Figure 5 This is a schematic diagram of the composition of a digital simulation environment provided by one embodiment of the present invention;

[0024] Figure 6 This is a structural diagram of a space-based target orbit determination device that only measures angles and compensates for orbit position errors, provided by one embodiment of the present invention;

[0025] Figure 7 This is a hardware architecture diagram of a computer device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0027] As mentioned above, when using relative angle measurement data to determine the orbital position of the target, it is necessary to accumulate measurement data for a period of time before the target orbit determination result converges. In this process, random errors and systematic errors will occur, and it is difficult to establish an accurate mathematical model for the systematic error.

[0028] Based on this, the concept of the present invention is to replace the empirical orbit dynamics model with a hybrid dynamics model of "empirical orbit dynamics model + neural network error compensation model" to achieve system error compensation, thereby improving the accuracy of space target orbit determination.

[0029] The specific implementation of the above concept is described below.

[0030] Please refer to Figure 1 An embodiment of the present invention provides a method for determining a space-based target orbit by only measuring angles and compensating for orbital position errors. The method includes:

[0031] Step 100 , based on the real orbital dynamics model of the target, establish an error model for introducing measurement errors into the empirical orbital dynamics model error;

[0032] Step 102: Training the initial neural network based on preset training samples to obtain an orbit position error estimation model; wherein the training samples are determined by historical orbit data and an empirical orbit dynamics model;

[0033] Step 104 : Compensate the error model according to the orbit position error estimation model, process the initial state of the target to be measured using the deviation compensation least square method, and obtain the optimal estimated value of the orbit of the target to be measured.

[0034] In this embodiment of the present invention, an error model combining the empirical orbital dynamics model error and measurement error is first established, and an initial error estimation network based on deep learning is designed. The initial error estimation network is then trained using preset training samples to obtain an orbital position error estimation model. Finally, a hybrid dynamics model combining the orbital position error estimation model and the empirical orbital dynamics model is proposed, and a bias-compensated least squares orbit determination method is designed to determine the optimal estimate of the target orbit. This method addresses the problem of errors between the empirical orbital dynamics model and the target's true motion model, which can "distort" the target's initial state estimate. By using this hybrid dynamics model, the target orbit is calculated using an empirical orbital dynamics model combined with a neural network error estimation algorithm, improving the accuracy of target orbit determination.

[0035] Described below Figure 1 How to perform the steps shown.

[0036] First, for step 100, an error model for introducing measurement errors into the empirical orbital dynamics model is established based on the real orbital dynamics model of the target.

[0037] When using the relative angle measurement data of the target from the space-based observation platform to determine the true state x0 of the target orbit, it is necessary to use the target t0~t N Angle measurement data during the observation period.N The internal target has a real orbit motion, and accordingly there is real angle measurement data of the target. For the optimal state estimate of the initial real state x0 By extrapolating the empirical orbital dynamics model and combining it with the measurement model, the estimated measurement value at the corresponding observation time can be calculated. The error between the empirical orbital dynamics model and the target's true orbital motion gradually increases with the observation time. At this time, a constant deviation will appear in the estimated measurement value, thereby "pull the optimal state estimate" like Figure 2 shown.

[0038] In an embodiment of the present invention, the process of establishing an error model includes: establishing a real orbital dynamics model of the target based on a first acceleration of the target determined by an empirical orbital dynamics model and a second acceleration determined by the empirical orbital dynamics model and the real orbital motion of the target; establishing a first equation and a second equation of the target at the target moment based on the state transfer matrices of the real orbital dynamics model and the empirical orbital dynamics model, as well as real angle measurement data and measurement equations; and approximating the difference between the first equation and the second equation and substituting it into the first equation to establish the error model.

[0039] Specifically, it is assumed that the target's true orbital dynamics model is:

[0040]

[0041] Where r is the orbital position vector in the target inertial system, and the two points represent its second-order derivative; a0 is the acceleration that can be accurately modeled and is used in the empirical orbital dynamics model; a μ is the unknown unmodelable acceleration, which is the error between the empirical orbital dynamics model and the true orbital motion of the target.

[0042] The state transfer matrix corresponding to this model Satisfies the following differential equation:

[0043]

[0044] Where I represents the identity matrix, 0 represents the zero matrix, and t k-1 ≤t≤t k-1 +τ, the state transfer matrix satisfies Where Φ(t k ,t k-1 ) is the state transition matrix corresponding to a0, Φ μ (t k ,t k-1 ) is a μ The corresponding state transition matrix.

[0045] Target time t kThe true state is x k , including the position r k and speed v k , denoted as x k =[r k ;v k ], we get the first equation between the initial state deviation and the measurement deviation:

[0046]

[0047] Where z k is the actual angle value, h(x k ) is based on the state x k The calculated angle value, According to the state x k Calculate the Jacobian matrix of the measurement equation, From time t0 to t k Time state transfer matrix The abbreviation of is the random error vector that ensures the equality, Δx0 is the initial state deviation, and state x k is a six-dimensional vector, including the target position and target velocity.

[0048] Similarly, an estimate of the initial state The t extrapolated by the empirical orbital dynamics model k The status at the moment is Include Location and speed Record Similarly, the second equation between the initial state deviation and the measurement deviation can be obtained:

[0049]

[0050] Where, According to the status Calculated estimated measurement value, H k According to the status The Jacobian matrix of the measurement equation is calculated, Φ is the time from t0 to t k The state transfer matrix Φ(t k ,t0), ε k is the random error vector that ensures the equality.

[0051] Subtract the first equation from the second equation and substitute into the relationship get:

[0052]

[0053] Among them, Φ μ From time t0 to t kTime state transfer matrix Φ μ (t k ,t0)

[0054] because General Δx k For a small amount, The above formula can be simplified as:

[0055]

[0056] Substituting this formula into the first equation yields:

[0057]

[0058] remember s k =H k Φ(I-Φ μ )Δx0, where s k is the constant deviation introduced into the measurement by the deviation between the empirical orbit dynamics model and the target's true orbit motion, i.e., the error model, which will introduce a constant deviation into the orbit determination result. N When N sets of measurement values ​​are accumulated within a time period, the orbit state estimation equation containing the error model can be obtained, in which the forms of vectors Y, H, S, and V are defined respectively.

[0059]

[0060] Because S is unknown and cannot be accurately modeled, this embodiment will subsequently use a neural network to approximate S with the help of the target's historical orbital motion data.

[0061] Then, for step 102, the initial neural network is trained according to the preset training samples to obtain a track position error estimation model.

[0062] In an embodiment of the present invention, the orbital position error estimation model is trained by: establishing an initial neural network based on deep learning, whose input is the orbital state estimate and epoch time, and whose output is the orbital position error; wherein the input layer of the initial neural network is a linear transfer purelin function, and the hidden layer activation functions are a linear transfer purelin function and a linear transfer satlin function;

[0063] According to the number of training samples and the output of the initial neural network, a training loss function L is established:

[0064]

[0065] Where M is the number of training samples, Δr i and The expected output and actual output of the neural network respectively;

[0066] The training samples are input into the initial neural network for training according to the pre-divided training set, verification set and test set to obtain the track position error estimation model.

[0067] Specifically, if Figure 3 As shown, Figure 3 The initial neural network structure diagram provided for this embodiment includes an input layer, two hidden layers, and an output layer. The input layer has 7 neurons, each hidden layer has 14 neurons, and the output layer has 3 neurons. The activation function of the input layer and one of the hidden layers is the linear transfer purelin function:

[0068] purelin(x)=x

[0069] The activation function of the other hidden layer is chosen to be a linear transfer satlin function with saturation characteristics:

[0070]

[0071] The 7 inputs of the neural network are epoch time t k and the state estimation at the corresponding time The three outputs of the neural network are position deviations Δr k .

[0072] In an embodiment of the present invention, training samples are determined using historical orbit data and an empirical orbital dynamics model, including: determining multiple initial states and multiple initial epochs of the target based on the historical orbit data; wherein a first interval of each initial epoch is at least 24 hours; calculating an orbital state estimate for each initial state within a first interval at a preset second interval based on the empirical orbital dynamics model; calculating a position error of the target based on the orbital state estimate and the historical orbital true state, and using the orbital state estimate and the corresponding position error as a set of data for the training sample.

[0073] For example, first give the target 20 initial states x 0,n (n=1,...,20), and give 15 initial epochs t at 24-hour intervals 0,m (m=1,…,15), from 2021-08-10 04:00:00.00(UTC) to 2021-08-24 04:00:00.00(UTC); then, for each initial state x 0,n and the initial epoch time t 0,m , based on the empirical orbital dynamics model, the orbital state is calculated at 3-minute intervals over a 24-hour period And according to the target's real historical orbital motion state x k Calculate the position deviation Δr k ,in Any initial state x 0,n and any initial epoch t 0,m The generated orbital state and position deviation sequence is a set of samples. A set of samples contains 480 samples, so a total of 20×15×480 samples are generated.

[0074] Furthermore, the determined samples are divided into training set, validation set and test set. Any group of samples with an initial epoch of 2021-08-24 04:00:00.00 (UTC) is set as the test set, and the remaining samples are randomly divided into training set and validation set in a ratio of 9:1. Then, the initial neural network is trained according to the loss function in the above process to obtain the orbital position error estimation model, which can estimate the vector parameter S calculated in the above process.

[0075] With respect to step 104 , the error model is compensated according to the orbit position error estimation model, and the initial state of the target to be measured is processed using the deviation compensation least squares method to obtain the optimal estimated value of the orbit of the target to be measured.

[0076] In the embodiment of the present invention, the optimal orbit estimate is determined by the following process: given a state estimate x0, the state of the target to be measured is calculated using the empirical orbit dynamics model to obtain a state estimate at the target time; the state estimate at the target time is input into the orbit position error estimation model, and the position error at the target time is output; the position error at the target time is input into the error model, and a constant deviation is output; the optimal orbit estimate is calculated based on the vector parameters and the constant deviation.

[0077]

[0078] Specifically, by combining the error model calculated in the above process with the orbit position error estimation model, a hybrid dynamics model of "empirical orbit dynamics + neural network error compensation" is obtained, and the following is proposed: Figure 4 The deviation-compensated least squares trajectory determination method shown is given by the initial state of the target As well as the initial epoch t0, t is calculated based on the empirical orbital dynamics model k Momentary status The vector parameters Y and H are further calculated.

[0079] Furthermore, the trained neural network is used to estimate t k Position deviation Δr at time k, then calculate s according to the measurement equation k , and substitute it into the above formula to calculate the target's orbit estimate. Finally, when the calculation result meets the preset judgment, it is determined to be the optimal orbit estimate.

[0080] The feasibility of the above method is verified by an embodiment below:

[0081] The digital simulation method is used to verify the above steps. Figure 5 The digital simulation environment shown verifies the effectiveness of the present invention. The digital simulation environment includes a system parameter initialization module, an empirical orbit dynamics model, a line of sight measurement simulation module, an error estimation module, a space target orbit determination module, a historical motion data excitation module, and an estimated performance analysis module. The system parameter initialization module sets the key system parameters and the observation platform and target orbit parameters; the empirical orbit dynamics model is used to recursively deduce the orbits of the target and the observation platform; the line of sight measurement simulation module is used to simulate the line of sight measurement of the observation platform to the target; the error estimation module uses a trained neural network to estimate the error of the empirical orbit dynamics model; the space target orbit determination module completes the target orbit estimation according to the orbit determination method in step 5; the historical motion data excitation module is used to provide the target's real on-orbit motion data, complete the neural network training and be used for algorithm performance evaluation; the estimated performance analysis module is used to evaluate the performance of the method of the present invention.

[0082] The orbital parameters of the observation platform are shown in Table 1. Table 1 shows the orbital parameters of the observation platform;

[0083] Table 1

[0084]

[0085] The target orbit parameters are shown in Table 2. Table 2 shows the target orbit parameters;

[0086] Table 2

[0087]

[0088] The starting time is 2021-08-26 00:00:00.00 (UTC), the sampling period is 60s, the total observation time is 20h, and the mean absolute value of the three-axis position estimation error during 200 Monte Carlo target shooting simulations is M e To evaluate the track estimation performance, it is defined as follows:

[0089]

[0090] Where r0=[x,y,z] is the target's true position, is the estimated value of the target simulation position for the i-th time.

[0091] By comparing the performance of the traditional least squares method and the method of the present invention for target position estimation, the absolute values ​​of the three-axis position estimation errors of the two methods are shown in Table 3. Table 3 compares the three-axis position estimation errors of the target by the two methods. It can be seen that the effectiveness of the method of the present invention is better to compensate for the error of the empirical orbit dynamics model and improve the accuracy of target orbit determination.

[0092] Table 3

[0093]

[0094] Please refer to Figure 6 An embodiment of the present invention provides a space-based target orbit determination device that compensates for orbit position errors by only measuring angles, the device comprising:

[0095] A modeling module 600 is used to establish an error model for introducing measurement errors into the empirical orbital dynamics model based on the real orbital dynamics model of the target;

[0096] A training module 602 is configured to train an initial neural network based on preset training samples to obtain an orbit position error estimation model; wherein the training samples are determined using historical orbit data and an empirical orbit dynamics model;

[0097] The processing module 604 is used to compensate the error model according to the orbit position error estimation model, process the initial state of the target to be measured using the deviation compensation least square method, and obtain the optimal estimate of the orbit of the target to be measured.

[0098] In the embodiment of the present invention, when the modeling module 600 establishes an error model for introducing measurement errors into the empirical orbital dynamics model based on the real orbital dynamics model of the target to be measured, it is specifically configured to perform the following operations:

[0099] A real orbital dynamics model of the target is established based on a first acceleration of the target determined by an empirical orbital dynamics model and a second acceleration determined by the empirical orbital dynamics model and the real orbital motion of the target; a first equation and a second equation of the target at the target moment are established based on the state transfer matrices of the real orbital dynamics model and the empirical orbital dynamics model, as well as real angle measurement data and measurement equations; and the difference between the first equation and the second equation is approximated and substituted into the first equation to establish the error model.

[0100] In the embodiment of the present invention, the error model is established by the following formula:

[0101]

[0102] Where Y, H, S, and V are all vector parameters of the error model; is the error between the true angle measurement value and the estimated angle measurement value; N is the number of measurement data sets in the target time period; s is the constant deviation introduced into the measurement by the deviation between the empirical orbit dynamics model and the target true orbit motion; H N The Jacobian matrix of the measurement equation is determined according to the trajectory state; Φ is from t0 to t N The state transfer matrix at the moment; Δx0 is the initial state deviation; ε N is the random error vector that ensures the equality.

[0103] In the embodiment of the present invention, the training module 602 is specifically configured to perform the following operations when training the initial neural network according to the preset training samples to obtain the track position error estimation model:

[0104] Based on deep learning, an initial neural network is established, whose input is the orbit state estimate and epoch time, and whose output is the orbit position error; wherein the input layer of the initial neural network is a linear transmission purelin function, and the hidden layer activation function is a linear transmission purelin function and a linear transmission satlin function;

[0105] According to the number of training samples and the output of the initial neural network, a training loss function L is established:

[0106]

[0107] Where M is the number of training samples, Δr i and The expected output and actual output of the neural network respectively;

[0108] The training samples are input into the initial neural network for training according to the pre-divided training set, verification set and test set to obtain the track position error estimation model.

[0109] In an embodiment of the present invention, the training samples are determined using historical orbit data and an empirical orbital dynamics model, including: determining multiple initial states and multiple initial epochs of the target based on the historical orbit data; wherein a first interval of each initial epoch is at least 24 hours; calculating an orbital state estimate for each initial state within a first interval at a preset second interval based on the empirical orbital dynamics model; calculating a position error of the target based on the orbital state estimate and the actual historical orbital state, and using the orbital state estimate and the corresponding position error as a set of data for the training samples.

[0110] In an embodiment of the present invention, the processing module 604 is specifically used to perform the following operations when executing the orbit position error estimation model, compensating the error model, processing the initial state of the target to be measured using the deviation compensation least squares method, and obtaining the optimal estimated value of the orbit of the target to be measured: according to the preset state estimation value, using the empirical orbit dynamics model to calculate the state of the target to be measured, and obtaining the state estimation at the target moment; inputting the state estimation at the target moment into the orbit position error estimation model, and outputting the position error at the target moment; inputting the position error at the target moment into the error model, and outputting the constant deviation; calculating the optimal estimated value of the orbit based on the vector parameters and the constant deviation.

[0111]

[0112] It should be noted that the above-mentioned embodiments of the space-based target orbit determination device with only angle measurement and compensation for orbital position error are illustrated only by the division of the aforementioned functional modules. In actual applications, the aforementioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the space-based target orbit determination device with only angle measurement and compensation for orbital position error provided in the above-mentioned embodiments and the space-based target orbit determination method with only angle measurement and compensation for orbital position error are based on the same concept. The specific implementation process is detailed in the method embodiments and will not be repeated here.

[0113] The embodiment of the present application also provides a computer device, please refer to Figure 7 The computer device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set, or instruction set, and the at least one instruction, at least one program, code set, or instruction set is loaded and executed by the processor to implement the space-based angle-only target orbit determination method with orbit position error compensation provided by the above-mentioned method embodiments.

[0114] An embodiment of the present application also provides a computer-readable storage medium, on which is stored at least one instruction, at least one program, code set, or instruction set, which is loaded and executed by a processor to implement the space-based angle-only target orbit determination method for compensating for orbital position errors provided in the above-mentioned method embodiments.

[0115] An embodiment of the present application also provides a computer program product, which includes a computer program. A processor of a computer device reads the computer program from a computer-readable storage medium, and the processor executes the computer program, so that the computer device executes the space-based angle-only target orbit determination method for compensating for orbital position errors as described in any of the above embodiments.

[0116] For the convenience of description, the above systems or devices are described as being divided into various modules or units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0117] Through the description of the above embodiments, it can be seen that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present application or certain parts of the embodiments.

[0118] Finally, it should be noted that, in this document, relational terms such as first, second, third, and fourth are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0119] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining a target orbit by measuring only angles in a space-based manner with compensation for orbital position error, characterized in that: The method comprises: According to the real orbital dynamics model of the target, an error model is established to introduce measurement error into the empirical orbital dynamics model. Training an initial neural network based on preset training samples to obtain an orbit position error estimation model; wherein the training samples are determined by historical orbit data and an empirical orbit dynamics model; According to the orbit position error estimation model, the error model is compensated, and the initial state of the target to be measured is processed using the deviation compensation least squares method to obtain the optimal estimate of the target orbit to be measured; The initial neural network is trained according to the preset training samples to obtain the track position error estimation model, including: Based on deep learning, an initial neural network is established whose input is the orbit state estimate and epoch time, and whose output is the orbit position error; wherein the input layer of the initial neural network is a linear transmission purelin function, and the hidden layer activation function is a linear transmission purelin function and a linear transmission satlin function; According to the number of training samples and the output of the initial neural network, a training loss function L is established: Where M is the number of training samples, Δr i and The expected output and actual output of the neural network respectively; Inputting the training samples into the initial neural network according to the pre-divided training set, validation set and test set to perform training, thereby obtaining the track position error estimation model; The method of compensating the error model according to the orbit position error estimation model and processing the initial state of the target to be measured by using the deviation compensation least squares method to obtain the optimal estimated value of the target orbit to be measured includes: According to the preset state estimation value, the state of the target to be measured is calculated using the empirical orbit dynamics model to obtain the state estimation at the target time; Inputting the state estimate of the target time into the orbit position error estimation model, and outputting the position error of the target time; Inputting the position error at the target moment into the error model and outputting a constant deviation; The optimal orbit estimate is calculated based on the vector parameters and the constant deviation. Wherein, Y, H, and S are all vector parameters of the error model.

2. The method according to claim 1, wherein The method of establishing an error model for introducing measurement errors into the empirical orbital dynamics model error based on the target's real orbital dynamics model includes: Establishing a real orbital dynamics model of the target to be measured based on a first acceleration of the target determined by the empirical orbital dynamics model and a second acceleration determined by the empirical orbital dynamics model and the real orbital motion of the target; Establishing a first equation and a second equation for the target at a target time based on the state transfer matrices of the real orbit dynamics model and the empirical orbit dynamics model, as well as the real angle measurement data and the measurement equation; The error model is established by performing approximate substitution processing on the difference between the first equation and the second equation and substituting the difference into the first equation.

3. The method according to claim 2, wherein The error model is established by the following formula: Where Y, H, S, and V are all vector parameters of the error model; is the error between the actual angle measurement value and the estimated angle measurement value; N is the number of measurement data sets within the target time period; s N is the constant deviation introduced into the measurement due to the deviation between the empirical orbital dynamics model and the target's true orbital motion; H N To determine the Jacobian matrix of the measurement equation according to the trajectory state; Φ(t N ,t0) is from t0 to t N The state transfer matrix at the moment; Δx0 is the initial state deviation; ε N is the random error vector that ensures the equality.

4. The method according to claim 3, wherein The training samples are determined by using historical orbit data and an empirical orbit dynamics model, including: Determining a plurality of initial states and a plurality of initial epochs of the target based on historical orbit data; wherein a first interval of each initial epoch is at least 24 hours; calculating an orbital state estimate for each initial state within a first interval at a predetermined second interval based on an empirical orbital dynamics model; The position error of the target is calculated based on the orbit state estimate and the historical orbit real state, and the orbit state estimate and the corresponding position error are used as a set of data for the training sample.

5. A space-based target orbit determination device that compensates for orbit position errors by measuring only angles, characterized in that: Applied to the method according to any one of claims 1 to 4, the device comprises: A modeling module is used to establish an error model for introducing measurement errors into the empirical orbital dynamics model based on the target's real orbital dynamics model; A training module, configured to train an initial neural network based on preset training samples to obtain an orbital position error estimation model; wherein the training samples are determined by using historical orbital data and an empirical orbital dynamics model; The processing module is used to compensate the error model according to the orbit position error estimation model, and process the initial state of the target to be measured using the deviation compensation least square method to obtain the optimal estimated value of the target orbit to be measured.

6. A computer device, characterized in that: The computer device includes a memory and a processor, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the steps of any one of the methods described in claims 1-4.

7. A computer-readable storage medium, characterized in that The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 4.

8. A computer program product, characterized in that The method comprises a computer program, which implements the steps of the method according to any one of claims 1 to 4 when the computer program is executed by a processor.

Images