Satellite positioning precision improvement method and device based on robust adaptive factor optimization
Through the graph optimization model based on the optimization of the difference-resistant adaptive factor, the noise covariance matrix is adjusted to adapt to complex environments, and the problem of insufficient accuracy of satellite positioning method in complex environments is solved, achieving higher positioning accuracy and robustness.
Patent Information
- Application Number
- CN202510204316.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-24
- Publication Date
- 2025-07-04
AI Technical Summary
The existing satellite positioning methods have poor positioning accuracy and large errors in complex environments. The traditional least squares method and Kalman filtering methods cannot effectively improve positioning accuracy when the noise characteristics change. The graph optimization method assumes that the observation noise is fixed, resulting in insufficient positioning accuracy in complex environments.
The graph optimization model based on the optimization of the anti-difference adaptive factor is adopted. By adjusting the correlation between the observed noise covariance matrix and the observed residual, the system robustness and positioning accuracy are enhanced, and the anti-difference adaptive factor is used to adjust the noise covariance matrix size to adapt to noise changes in complex environments.
It improves the accuracy and robustness of satellite positioning, especially in complex environments, which can effectively reduce errors and enhances the system's positioning performance under multipath effect and atmospheric interference.
Smart Images
Figure CN120256839A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of satellite navigation and positioning, and particularly relates to a method and device for improving satellite positioning accuracy based on the optimization of a robust adaptive factor. Background Art
[0002] The global satellite navigation system includes four major systems: the Global Positioning System (GPS) of the United States, GLONASS of Russia, Galileo of the European Union, and the Beidou satellite navigation system. The satellite navigation system has been widely used in many fields. In the satellite navigation system, single-point positioning is a commonly used positioning method, and currently, most single-point positioning uses the least squares method or the Kalman filtering method for solution.
[0003] The core principle of the least squares method is to estimate the user's position by minimizing the sum of the squares of the errors between the observed values and the predicted values. However, in some complex observation environments such as areas with dense high-rise buildings and canyons, due to factors such as multipath effects and atmospheric interference during the propagation of satellite signals, there are large errors in satellite observed values at certain moments. The results directly obtained by the least squares method often have deviations and are difficult to meet the requirements of high-precision positioning.
[0004] Conventional Kalman filtering can use the estimated value of the previous state and the observed value of the current state to update the estimated value of the current state. When used for positioning solution, it mainly depends on the linear system assumption and the known noise statistical characteristics. However, it is difficult for the noise situation in the actual environment to fully meet this assumption condition, and it is difficult to accurately model the user's motion model. Therefore, it is also impossible to further improve the positioning accuracy.
[0005] In recent years, graph optimization methods have achieved remarkable results in the fields of computer vision, robot positioning and navigation, etc. The graph optimization method models the problem as a graph structure, where the nodes in the graph represent the variables to be optimized (such as position, attitude, etc.), and the edges represent the constraint relationships between the variables (such as observed data). By optimizing the edge weights and node positions in the graph, the optimal variable values can be solved. In satellite navigation and positioning, the graph optimization method can utilize the correlation between historical observation data to improve the positioning accuracy. However, traditional graph optimization methods usually assume that the observation noise is fixed, which often does not hold in practical applications, especially in complex environments where the noise characteristics change with time and space.
[0006] Therefore, the current satellite positioning methods have poor positioning accuracy and large errors. Summary of the Invention
[0007] The embodiments of the present invention provide a method and device for improving satellite positioning accuracy based on the optimization of a robust adaptive factor, which can solve the problems of poor positioning accuracy and large errors of the current satellite positioning methods.
[0008] In a first aspect, an improved satellite positioning accuracy method based on robust adaptive factor optimization provided by an embodiment of the present invention includes:
[0009] Based on a graph optimization model, globally optimize the user positioning data according to the rough estimate value and historical rough estimate value of the user positioning data at the current moment, and solve to obtain the refined estimate value of the user positioning data at the current moment;
[0010] Among them, the cost function of the graph optimization model is constructed based on a robust adaptive factor, and the robust adaptive factor is used to adjust the size of the observation noise covariance matrix in the cost function according to the size relationship between the observation residual and a preset observation residual threshold, so that the size of the observation noise covariance matrix is positively correlated with the size of the observation residual.
[0011] In a second aspect, an embodiment of the present invention provides a device including a processing unit; the processing unit is configured to:
[0012] Based on a graph optimization model, globally optimize the user positioning data according to the rough estimate value and historical rough estimate value of the user positioning data at the current moment, and solve to obtain the refined estimate value of the user positioning data at the current moment;
[0013] Among them, the cost function of the graph optimization model is constructed based on a robust adaptive factor, and the robust adaptive factor is used to adjust the size of the observation noise covariance matrix in the cost function according to the size relationship between the observation residual and a preset observation residual threshold, so that the size of the observation noise covariance matrix is positively correlated with the size of the observation residual.
[0014] In a third aspect, an embodiment of the present invention provides an electronic device, including a processor and a memory. Among them, the memory is used to store a computer program; the processor can be used to execute the calculator program (instructions) stored in the memory to implement the method of the first aspect above.
[0015] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed, the method of the first aspect above can be implemented.
[0016] It can be understood that the beneficial effects of the second to fourth aspects above can refer to the relevant descriptions of the first aspect above, and will not be elaborated here.
[0017] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows: Since the cost function of the graph optimization model for calculating the user's position information and speed information in the present invention is constructed based on the robust adaptive factor, and the robust adaptive factor can adjust the size of the observation noise covariance matrix in the cost function to be proportional to the size of the observation residual. When there are large errors in the observation data, the robust factor is used to increase the value of the observation noise covariance matrix, thereby reducing the weight of the current observation in the calculation of the optimal state estimate and enhancing the robustness and positioning accuracy of the system in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 FIG. is a flowchart of an implementation method for constructing a cost function of a graph optimization model provided by an embodiment of the present invention;
[0019] Figure 2 FIG. is a flowchart of an implementation method for a method for improving satellite positioning accuracy optimized based on a robust adaptive factor provided by an embodiment of the present invention;
[0020] Figure 3 FIG. is a flowchart of an implementation method for a method for solving user positioning data based on the Levenberg-Marquardt method provided by an embodiment of the present invention;
[0021] Figure 4 FIG. is a schematic structural diagram of a device for improving satellite positioning accuracy optimized based on a robust adaptive factor provided by an embodiment of the present invention;
[0022] Figure 5 FIG. is a comparison schematic diagram of the positioning root mean square error provided by an embodiment of the present invention;
[0023] Figure 6 FIG. is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0024] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present invention. However, those skilled in the art should clearly understand that the present invention can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present invention.
[0025] It should be understood that when used in the specification and appended claims of the present invention, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.
[0026] It should also be understood that the term "and / or" as used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.
[0027] As used in the specification of the present invention and the appended claims, the term "if" can be construed, depending on the context, as "when" or "once" or "in response to determining" or "in response to detecting". Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be construed, depending on the context, to mean "once determined" or "in response to determining" or "once [the described condition or event] is detected" or "in response to detecting [the described condition or event]".
[0028] In addition, in the description of the specification of the present invention and the appended claims, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be construed as indicating or implying relative importance.
[0029] Reference to "one embodiment" or "some embodiments" or the like described in the specification of the present invention means that a specific feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of the present invention. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized.
[0030] The present invention will be further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.
[0031] Figure 1 The flowchart of the implementation of a method for constructing a cost function of a graph optimization model provided by an embodiment of the present invention is shown. By way of example and not limitation, the method may include steps S101 - S105, which will be described below.
[0032] S101, establish the state equations of the user's position and velocity.
[0033] In one example, the state equations of the user's position and velocity may satisfy the following formula:
[0034] X k = N k,k-1 X k-1 + α k (1.1)
[0035] Among them, X k is the refined estimated value of the user's positioning data at the k-th moment, and N k,k-1 is the refined estimated value of the user's positioning data at the (k - 1)-th moment, X k-1 to X k is the state transition matrix, and α k is the state noise vector at the k-th moment, which can usually be assumed to be zero-mean Gaussian noise.
[0036] Among them:
[0037]
[0038] Among them, Δt is the positioning time interval, that is, the time difference between the k-th positioning moment (abbreviated as the k-th moment) and the (k - 1)-th positioning moment (abbreviated as the (k - 1)-th moment).
[0039] Exemplarily, the user's positioning data may include the user's position information and speed information. The user's position information may include three coordinate values of the user in the real coordinate system, and the user's speed information may include three components of the user's speed in the real coordinate system. That is, X k =[x k , y k , z k , v xk , v yk , v zk T , where x k , y k , z k are respectively the position coordinates of the user on the x, y, and z axes at the k-th moment, and v xk , v yk , v zk are respectively the components of the user's speed on the x, y, and z axes at the k-th moment.
[0040] S102. Establish the observation equation of the system.
[0041] Exemplarily, the observation equation of the positioning system may satisfy the following formula:
[0042] G k = M k X k + v k (1.3)
[0043] Among them, G k is the rough estimated value of the user's positioning data at the k-th moment, M k is the measurement matrix, and vk is the observation noise vector.
[0044] Among them:
[0045]
[0046] Optionally, G k may be obtained according to the least squares positioning solution.
[0047] S103. Establish a single-point cost function for the graph optimization model.
[0048] Exemplarily, the single-point cost function of the graph optimization model may satisfy the following formula:
[0049]
[0050] where \(e_{k, mea}=(G k - M k X k ) represents the observation residual at the \(k\)-th moment. The observation residual can reflect the difference between the measured value and the measured predicted value calculated according to the state vector. \(e_{k, sta}=(X k - N k,k-1 X k-1 ) represents the state prediction residual at the \(k\)-th moment, indicating the deviation between the state prediction value and the actual state value. is the inverse matrix of the observation noise covariance matrix, and \(P k,k-1 is the covariance matrix of the state estimation error.
[0051] By minimizing the above single-point cost function, the optimal estimated value of the state vector at the current moment (i.e., the \(k\)-th moment) can be obtained.
[0052] S104. Construct a global cost function for the graph optimization model based on historical observation data.
[0053] In one example, the graph optimization model can optimize and solve the state vectors \(X' k = [X k-p+1 , X k-p+2 ,... X k T at a total of \(p\) moments in the past and the current moment as unknowns; by using historical observation data (i.e., the historical rough estimates of user positioning data) to solve the global optimal solutions at \(p\) moments, rather than being limited to the observation data at the current moment. Therefore, a global cost function for the graph optimization model can be constructed based on historical observation data, and the global optimal estimated value of \(X k can be obtained by solving the minimum value of the global cost function.
[0054] Exemplarily, the global cost function of the graph optimization model may satisfy the following formula:
[0055]
[0056] Among them, J′(X'k) is the global cost function of the graph optimization model, p - 1 is equal to the number of time steps traced forward during the global optimization of the graph optimization model, and G i is the rough estimate of the user's positioning data at the i-th moment, and M i is the measurement matrix, and X i is the refined estimate of the user's positioning data at the i-th moment, and R i is the original observation noise covariance matrix at the i-th moment, and N i,i-1 is the refined estimate X of the user's positioning data at the (i - 1)-th moment i-1 to X i is the state transition matrix, and P i,i-1 is the covariance matrix of the state estimation error at the i-th moment.
[0057] S105. Based on the robust adaptive factor, transform the global cost function to obtain the cost function of the graph optimization model.
[0058] In one example, the global cost function in step S104 regards the observation noise v k as Gaussian white noise, and its noise statistical information is usually regarded as a known constant obtained from empirical values, that is, the noise covariance matrix R is the same at each moment. However, in the actual satellite positioning environment, factors such as multipath effects and atmospheric interference cause large errors in the observed values, and the observation noise does not strictly conform to the assumed Gaussian distribution. To reduce the problem of limited positioning accuracy caused by inaccurate observation noise modeling, R can be subjected to robust processing, and a robust adaptive factor is introduced to adjust the size of the observation noise covariance matrix.
[0059] Exemplarily, the robust adaptive factor can satisfy the following formula:
[0060]
[0061] Among them, λ k is the robust adaptive factor at the k-th moment, e k,mea is the observation residual at the k-th moment, and c0 is the observation residual threshold.
[0062] Exemplarily, the observation noise covariance matrix adjusted by the robust adaptive factor can satisfy the following formula:
[0063]
[0064] Therefore, the final cost function of the graph optimization model can satisfy the following formula:
[0065]
[0066] Among them, J(X' k) represents the final cost function of the graph optimization model.
[0067] The robust adaptive factor can compare the magnitude relationship between the observation residual and the observation residual threshold. When the observation residual is greater than the threshold, there are gross errors in the observed quantity corresponding to that moment. At this time, the observation noise cannot accurately reflect the true observation error. Therefore, by using the covariance matrix R of the observation noise at p moments i transformed into so as to realize the automatic adjustment of the covariance matrix of the observation noise. When there are large errors in the observation data, the value of the covariance matrix of the observation noise is increased through the robust factor, thereby reducing the weight of the current observed quantity in the calculation of the optimal state estimate and enhancing the robustness and positioning accuracy of the system in complex environments.
[0068] The method for improving satellite positioning accuracy based on the optimization of the robust adaptive factor provided by the embodiments of the present invention can be applied to electronic devices such as mobile terminals, personal laptop computers, supercomputers, etc. The embodiments of the present invention do not impose any restrictions on the specific types of electronic devices.
[0069] Figure 2 The flowchart of an implementation of a method for improving satellite positioning accuracy based on the optimization of the robust adaptive factor provided by the embodiments of the present invention is shown. By way of example and not limitation, this method can be applied to the above-mentioned electronic devices. This method may include steps S201 - S202, which will be described below.
[0070] S201, obtain the rough estimate value of the current moment of the user positioning data.
[0071] Exemplarily, the rough estimate value of the current moment of the user positioning data can be obtained by solving using the least squares method.
[0072] S202, based on the graph optimization model, globally optimize the user positioning data according to the rough estimate value of the current moment and the historical rough estimate values of the user positioning data, and solve to obtain the refined estimate value of the current moment of the user positioning data.
[0073] In one example, the cost function of the graph optimization model, that is, the above formula (1.9), can be solved, and the X that makes this cost function reach the minimum value k is used as the refined estimate value of the current moment of the positioning data.
[0074] Exemplarily, based on the Levenberg - Marquardt method, the cost function of the graph optimization model can be iteratively solved according to the rough estimate value of the current moment and the historical rough estimate values of the user positioning data to globally optimize the user positioning data and obtain the refined estimate values of the user position information and speed information at the current moment.
[0075] Since the cost function of the graph optimization model for calculating the user's position information and speed information in the present invention is constructed based on the robust adaptive factor, and the robust adaptive factor can adjust the size of the observation noise covariance matrix in the cost function to be proportional to the size of the observation residual. When there are large errors in the observation data, the value of the observation noise covariance matrix is increased through the robust factor, thereby reducing the weight of the current observation in the calculation of the optimal state estimate and enhancing the robustness and positioning accuracy of the system in complex environments.
[0076] Figure 3 The flowchart of an implementation of a method for calculating user positioning data based on the Levenberg-Marquardt method provided by an embodiment of the present invention is shown. By way of example and not limitation, this method is a specific possible implementation of the above step S202. This method may include steps S301 - S305, which are described below.
[0077] S301, perform a first-order expansion on the refined estimated value of the error cost function in the cost function after the nth iteration, and substitute the expanded error cost function into the cost function to obtain the cost function linearized in the nth iteration.
[0078] In one example, the cost function of the graph optimization model can be rewritten in a general form first, then the error cost function is expanded at the refined estimated value after the nth iteration, and then the expanded error cost function is substituted into the cost function of the graph optimization model to obtain the cost function linearized in the nth iteration.
[0079] Exemplarily, the cost function in the general form of the graph optimization model may satisfy the following formula:
[0080]
[0081] where rj(X'k) is the error cost function, including the observation residual ej,mea = (G j -M j X' j ) and the state prediction residual ej,sta = (X' j -N j,j-1 X j-1 ), Σ- 1 represents matrix and matrix.
[0082] Exemplarily, the expanded error cost function may satisfy the following formula:
[0083]
[0084] where ΔX is the state increment, and D j is the Jacobian matrix of the error cost function with respect to the state vector.
[0085] Exemplarily, the linearized cost function can satisfy the following formula:
[0086]
[0087] where is the linearized cost function, is the gradient vector,
[0088] S302. Take the derivative of the cost function linearized in the n-th round with respect to the state increment, and set the derivative of the cost function to 0 to obtain the state increment after the (n + 1)-th round of iteration.
[0089] Exemplarily, the state increment after the (n + 1)-th round of iteration can satisfy the following formula:
[0090] (G + μI)ΔX n+1 = -b
[0091] where D j is the Jacobian matrix of the error cost function with respect to the state vector, μ is the damping factor, I is the identity matrix, ΔX n+1 is the state increment after the (n + 1)-th round of iteration, and b is the gradient vector.
[0092] S303. Take the sum of the refined estimate value after the n-th round of iteration and the state increment after the (n + 1)-th round of iteration as the refined estimate value after the (n + 1)-th round of iteration.
[0093] Exemplarily, the refined estimate value after the (n + 1)-th round of iteration can satisfy the following formula:
[0094]
[0095] where is the refined estimate value after the (n + 1)-th round of iteration, is the refined estimate value after the n-th round of iteration.
[0096] S304. Determine whether the preset stop condition is satisfied.
[0097] In one example, if the stop condition is satisfied, step S305 can be performed.
[0098] Exemplarily, it can be determined whether the preset stop condition is satisfied by determining whether n is greater than or equal to the maximum number of iterations, or by determining whether the error obtained according to is greater than or equal to the preset error convergence threshold.
[0099] In another example, if the stop condition is not satisfied, let n = n + 1 and continue to iterate from step S301.
[0100] S305. Use the refined estimated value after the (n + 1)-th iteration as the refined estimated value at the current moment of the user positioning data.
[0101] Through the Levenberg-Marquardt method, the positioning result can be globally optimized using historical observation data and current observation data to obtain the global optimal solutions of the user's position information and speed information.
[0102] Figure 4 The figure shows a schematic structural diagram of a satellite positioning accuracy improvement device based on the optimization of a robust adaptive factor provided by an embodiment of the present invention. By way of example and not limitation, the device 400 may include an acquisition unit 410 and a processing unit 420.
[0103] Exemplarily, the acquisition unit 410 may be used to obtain a rough estimated value at the current moment of the user positioning data; the processing unit 420 may globally optimize the user positioning data based on a graph optimization model according to the rough estimated value at the current moment of the user positioning data and the historical rough estimated values, and solve to obtain the refined estimated value at the current moment of the user positioning data; wherein, the cost function of the graph optimization model is constructed based on a robust adaptive factor, and the robust adaptive factor is used to adjust the size of the observation noise covariance matrix in the cost function according to the magnitude relationship between the observation residual and a preset observation residual threshold, so that the size of the observation noise covariance matrix is positively correlated with the size of the observation residual.
[0104] Figure 5 The figure shows a comparison schematic diagram of the positioning root mean square error provided by an embodiment of the present invention.
[0105] See Figure 5 , which shows the root mean square error of the user positioning data determined according to the method provided by the present invention, the root mean square error of the user positioning data determined according to the traditional graph optimization model, and the root mean square error of the user positioning data determined according to the least squares method.
[0106] It can be seen from Figure 5 that the root mean square error of the user positioning data determined according to the method provided by the present invention is less than the traditional two methods in most cases, and the comprehensive performance of the present invention is better.
[0107] Since the cost function of the graph optimization model for calculating the user's position information and speed information in the present invention is constructed based on a robust adaptive factor, and the robust adaptive factor can adjust the size of the observation noise covariance matrix in the cost function to be proportional to the size of the observation residual, thereby realizing the dynamic adjustment of the observation noise covariance matrix. Compared with the traditional graph optimization method in which the observation noise is calculated as a fixed value, the present invention can improve the accuracy of the calculated user position and speed information and enhance the robustness of the positioning system in a complex environment.
[0108] Figure 6 The following shows a schematic structural diagram of an electronic device provided by an embodiment of the present invention. As Figure 6 shown, the electronic device 600 may include: at least one processor 610 ( Figure 6 only one processor is shown in the figure), a memory 620, and a computer program 630 stored in the memory 620 and executable on the at least one processor 610. When the processor 610 executes the computer program 630, the steps in any of the above method embodiments are implemented.
[0109] The electronic device 600 may be a processing device such as a robot that can implement the above method. The specific type of the electronic device in the embodiment of the present invention is not limited in any way.
[0110] Those skilled in the art can understand that Figure 6 merely an example of the electronic device 600, which does not constitute a limitation on the electronic device. It may include more or fewer components than shown in the figure, or combine some components, or different components. For example, the electronic device 600 may further include an input / output interface.
[0111] The so-called processor 610 may be a central processing unit (CPU), and the processor 610 may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASTCs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0112] The memory 620 may be an internal storage unit, such as a hard disk or memory, in some embodiments. The memory 620 may also be an external storage device, such as a plug-in hard disk, a Smart Memory Card (SMC), a Secure Digital (SD) card, a Flash Card, etc., in other embodiments. Further, the memory 620 may include both an internal storage unit and an external storage device. The memory 620 is used to store an operating system, application programs, a Boot Loader, data, and other programs, such as the program code of the computer program. The memory 620 may also be used to temporarily store data that has been output or will be output.
[0113] It should be understood that the sequence numbers of the steps in the above embodiments do not indicate the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
[0114] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiments can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of mutual distinction and do not limit the protection scope of the present invention. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be described herein again.
[0115] The embodiments of the present invention also provide a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments can be implemented.
[0116] The embodiments of the present invention provide a computer program product, and when the computer program product runs on an electronic device, the electronic device can execute to implement the steps in the above method embodiments.
[0117] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such understanding, to implement all or part of the processes in the above method embodiments of the present invention, a computer program can be used to instruct relevant hardware to complete. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the photographing device / terminal device, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disc, etc. In some jurisdictions, according to legislation and patent practice, the computer-readable medium cannot be an electrical carrier signal and a telecommunication signal.
[0118] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0119] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.
Claims
1. A method for improving satellite positioning accuracy based on the optimization of a robust adaptive factor, characterized in that, Including: Based on the graph optimization model, globally optimize the user positioning data according to the rough estimate value and historical rough estimate value of the user positioning data at the current moment, and calculate the refined estimate value of the user positioning data at the current moment; Among them, the cost function of the graph optimization model is constructed based on a robust adaptive factor, and the robust adaptive factor is used to adjust the size of the observation noise covariance matrix in the cost function according to the size relationship between the observation residual and a preset observation residual threshold, so that the size of the observation noise covariance matrix is positively correlated with the size of the observation residual.
2. The method according to claim 1, characterized in that, The robust adaptive factor satisfies the following formula: where λ k is the robust adaptive factor at the k-th moment, and e k,mea is the observation residual at the k-th moment, and c0 is the observation residual threshold.
3. The method according to claim 2, wherein The cost function satisfies the following formula: Among them, J(X' k ) represents the cost function, p - 1 is equal to the number of time steps traced forward when the graph optimization model globally optimizes the user location data, G i is the rough estimate of the user location data at the i-th moment, M i is the measurement matrix, X i is the refined estimate of the user location data at the i-th moment, is the observation noise covariance matrix at the i-th moment adjusted by the robust adaptive factor λi, R i is the original observation noise covariance matrix at the i-th moment, N i,i-1 is the refined estimate X of the user location data at the (i - 1)-th moment i-1 to X i of the state transition matrix, P i,i-1 is the covariance matrix of the state estimation error at the i-th moment.
4. The method according to claim 1, wherein The step of globally optimizing the user positioning data according to the rough estimate value and historical rough estimate value of the user positioning data at the current moment based on the graph optimization model and calculating the refined estimate value of the user positioning data at the current moment includes: Based on the Levenberg-Marquardt method, iteratively solve the cost function of the graph optimization model according to the rough estimate value and historical rough estimate value of the user positioning data at the current moment to globally optimize the user positioning data and obtain the refined estimate value of the user positioning data at the current moment.
5. The method according to claim 4, wherein The step of iteratively solving the cost function of the graph optimization model according to the rough estimate value and historical rough estimate value of the user positioning data at the current moment based on the Levenberg-Marquardt method to globally optimize the user positioning data and obtain the refined estimate value of the user positioning data at the current moment includes: Perform a first-order expansion of the error cost function in the cost function at the refined estimate value after the nth iteration, and substitute the expanded error cost function into the cost function to obtain the cost function linearized in the nth round, where the error cost function includes the observation residual and state prediction residual in the cost function; Take the derivative of the cost function linearized in the nth round with respect to the state increment, and set the derivative cost function to 0 to obtain the state increment after the (n + 1)th iteration; Take the sum of the refined estimate value after the nth iteration and the state increment after the (n + 1)th iteration as the refined estimate value after the (n + 1)th iteration; Determine whether a preset stop condition is satisfied. If the preset stop condition is satisfied, take the refined estimate value after the (n + 1)th iteration as the refined estimate value of the user positioning data at the current moment; if the preset stop condition is not satisfied, continue the iteration.
6. The method according to claim 5, characterized in that The state increment after the (n + 1)th iteration satisfies the following formula: (G + μI)ΔX n+1 = -b Among them, D j is the Jacobian matrix of the error cost function with respect to the state vector, μ is the damping factor, I is the identity matrix, and ΔX n+1 is the state increment after the (n + 1)-th iteration, and b is the gradient vector.
7. The method according to claim 1, wherein The rough estimate values of the user position information and speed information at the current moment are obtained by performing preliminary positioning calculation according to the least squares method.
8. A satellite positioning accuracy improvement device based on the optimization of a robust adaptive factor, characterized in that, Including a processing unit, the processing unit is configured to: Based on the graph optimization model, globally optimize the user positioning data according to the rough estimate value and historical rough estimate value of the user positioning data at the current moment, and calculate the refined estimate value of the user positioning data at the current moment; Among them, the cost function of the graph optimization model is constructed based on a robust adaptive factor, and the robust adaptive factor is used to adjust the size of the observation noise covariance matrix in the cost function according to the magnitude relationship between the observation residual and a preset observation residual threshold, so that the size of the observation noise covariance matrix is positively correlated with the size of the observation residual.
9. An electronic device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, When the processor executes the computer program, the method described in any one of claims 1-7 is implemented.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the electronic device, the method described in any one of claims 1-7 is implemented.