Model training method and device, processing method and device, equipment and medium

Abstract

The application provides a model training method and device, a processing method and device, equipment and a medium. The training method of the positioning and speed measurement model is applied to a GNSS receiver and includes the following steps: acquiring a training signal group, acquiring a pseudo code in the GNSS receiver, setting a first coefficient according to the pseudo code, setting a first parameter, performing convolution processing on the input signal according to the first coefficient and the first parameter to obtain feature information of the input signal, setting a second coefficient and a second parameter, and processing the training signal group and the feature information according to the second coefficient and the second parameter to obtain a training positioning and speed measurement result of the GNSS receiver. The application simplifies the data processing process and improves the efficiency while ensuring the accuracy of the output positioning and speed measurement result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of satellite signal processing, and in particular to a model training method and device, a processing method and device, equipment and a medium. BACKGROUND

[0002] The overall design of a traditional GNSS receiver chip is generally as shown in Figure 1 In a traditional GNSS receiver chip, the process from intermediate frequency data IFdata input to positioning result output generally needs to go through the following processing: first, the intermediate frequency data is subjected to carrier and pseudo-code correlation processing, and then the result of the correlation processing is input to a tracking module for frequency discrimination and phase discrimination, and after filtering processing, it is fed back to a numerically controlled oscillator (NCO) to realize accurate tracking of the input signal. In this process, in order to cope with the influence of different signal strengths and different carrier dynamics, professional algorithm personnel need to polish and debug the module parameters for a long time to ensure the stability and accuracy of signal tracking, and finally to ensure the accuracy of the observation quantity input to the Position Velocity Timing (PVT) module. Finally, according to the observation quantity information passed by the tracking module, combined with the satellite ephemeris and time information, the navigation positioning solution is completed, and the positioning and velocity result is output. This module is also a complex and key link in GNSS that affects performance, especially in urban environments, professional algorithm personnel are needed to effectively evaluate the observation quantity, give appropriate strategies, and at the same time, combined with the measured data, various motion models are constrained and optimized, which is also a long and laborious optimization process.

[0003] To solve the above problems, basically the following two solutions are used:

[0004] 1. Use machine learning to replace part of the tracking module and PVT module, use the powerful learning ability and nonlinear expression ability of machine learning to reduce complex manual design, and use rich data to improve the accuracy of the local module, so as to improve the performance of the overall system. However, local optimization cannot ensure global optimization, and the connection of these models also needs to be designed manually by professional personnel, only to reduce part of the workload.

[0005] 2. The vector GNSS receiver structure designed by using the vector receiver algorithm is as shown in Figure 2As shown, the input signal is processed through correlation, through the discriminator or directly into the PVT module to form measurement information, reducing the design of the independent loop and the parameters that need to be optimized, and using the spatial correlation to make the system performance better. It is just the above changes that put higher requirements on the design of the PVT module, involving the nonlinear expression ability of the PVT model, the selection and evaluation of reliable measurement information, and the need for a longer algorithm design and parameter optimization process to achieve the stability of the traditional structure.

[0006] In summary, the above solution does not reduce the complexity of algorithm design and the difficulty of algorithm optimization. Therefore, the present application proposes a model training method and device, a processing method and device, equipment and a medium to reduce the complexity of algorithm design and the difficulty of algorithm optimization while ensuring the accuracy of positioning and speed measurement results. SUMMARY

[0007] The present application provides a model training method and device, a processing method and device, equipment and a medium to solve the technical problems of the complexity of algorithm design and the difficulty of algorithm optimization for solving the positioning and speed measurement results of the carrier in the prior art.

[0008] In a first aspect, the present application provides a training method of a positioning and velocity measurement model, applied to a GNSS receiver, comprising: obtaining a training signal set, the training signal set comprising an input signal of the GNSS receiver, historical positioning information of the GNSS receiver, path propagation information of the input signal, and current trajectory information and velocity information of a visible satellite, the path propagation information being delay information of the input signal from being transmitted by the visible satellite to being received by the GNSS receiver, the current trajectory information and velocity information of the visible satellite referring to trajectory information and velocity information at a moment when the input signal is transmitted by the visible satellite; obtaining a pseudo code in the GNSS receiver, and setting first coefficients according to the pseudo code, the first coefficients comprising a number and a size of convolution kernels of a convolution layer, the size comprising a number of points of the convolution kernels and an arrangement of the number of points; setting first parameters, the first parameters comprising a first weight parameter and a first bias parameter of each point of the convolution kernels; performing convolution processing on the input signal according to the first coefficients and the first parameters to obtain feature information of the input signal; setting second coefficients and second parameters, the second coefficients comprising a number of layers of the fully connected layer and a number of points of each layer of the fully connected layer, the second parameters comprising a second weight parameter and a second bias parameter of each point of the fully connected layer; processing the training signal set and the feature information according to the second coefficients and the second parameters to obtain a training positioning and velocity measurement result of the GNSS receiver; judging whether the training positioning and velocity measurement result meets a first preset condition, if yes, saving the first coefficients, the second coefficients, the first parameters and the second parameters to generate the positioning and velocity measurement model, and if not, modifying the first parameters and the second parameters to obtain the positioning and velocity measurement model meeting the first preset condition.

[0009] The present application simplifies the data processing process while ensuring the accuracy of the output positioning and velocity measurement result, and solves the technical problems of complexity of algorithm design and difficulty of algorithm optimization for solving the positioning and velocity measurement result of a carrier in the prior art.

[0010] Optionally, the setting of the first coefficients according to the pseudo code comprises: setting a number of convolution kernels of a first layer in the convolution layer according to a number of satellites of the visible satellite; setting a size of the convolution kernels of the first layer according to a length of the pseudo code; and setting a number and a size of convolution kernels of the remaining layers in the convolution layer according to an output dimension of a previous layer. The present application has the beneficial effect of obtaining appropriate first coefficients of the positioning and velocity measurement model.

[0011] Optionally, the processing the training signal group, the feature information according to the second coefficient and the second parameter to obtain the training positioning and velocity measurement result of the carrier comprises: calculating a sum of a first product and the second bias parameter to obtain the first motion trajectory and the first motion velocity of the GNSS receiver, the first product being obtained by multiplying the historical positioning information, the feature information, and the path propagation information by the second weight parameter corresponding to each point in each layer of the full connection layer.

[0012] Optionally, before the judging whether the training positioning and velocity measurement result meets the first preset condition, the method further comprises: setting a first coefficient and a second coefficient, the first coefficient having a value of 0-1, the second coefficient having a value of 0-1, and the sum of the first coefficient and the second coefficient being 1. The beneficial effect is that the subsequent judging process of the training positioning and velocity measurement result is optimized by setting the first coefficient and the second coefficient.

[0013] Optionally, before the judging whether the training positioning and velocity measurement result meets the first preset condition, the method further comprises: obtaining an actual positioning and velocity measurement result of the GNSS receiver, the actual positioning and velocity measurement result being obtained by actual observation, and the actual positioning and velocity measurement result comprising a second motion trajectory and a second motion velocity.

[0014] Optionally, the judging whether the training positioning and velocity measurement result meets the first preset condition comprises: obtaining a first bias and a second bias, the first bias being a trajectory bias of the first motion trajectory compared with the second motion trajectory, and the second bias being a difference between the first motion velocity and the second motion velocity; obtaining a sum of a first product and a second product, the first product being a product of the first coefficient and the first bias, and the second product being a product of the second coefficient and the second bias; judging whether the sum of the first product and the second product is less than a preset threshold value, if yes, the training positioning and velocity measurement result meets the first preset condition; otherwise, the training positioning and velocity measurement result does not meet the first preset condition.

[0015] Optionally, before the obtaining the training signal group, the method further comprises: establishing a data set, the data set comprising a training set and a test set, the training set and the test set each comprising a plurality of training signal groups; and the obtaining the training signal group comprises: obtaining the training signal group in the training set.

[0016] Optionally, after the positioning and speed measurement model is generated, further comprising: obtaining a training signal group in the test set; obtaining an output result of the positioning and speed measurement model according to the training signal group in the test set, judging whether the output result satisfies a second preset condition, if yes, completing the training of the positioning test model; otherwise, modifying the first parameter and the second parameter to obtain a positioning and speed measurement model satisfying the second preset condition.

[0017] In a second aspect, the present application provides a data processing method, comprising: obtaining an input signal, and inputting the input signal into a positioning and speed measurement model obtained by the training method of the positioning and speed measurement model according to any one of the first aspect, obtaining an output result of the positioning and speed measurement model according to the input signal.

[0018] In a third aspect, the present application provides a training device for positioning and velocity measurement model, which is used to execute the positioning and velocity measurement model training method in any one of the first aspect, comprising: a first acquisition unit, a second acquisition unit, a parameter setting unit, a convolution processing unit, a positioning and velocity measurement result generation unit, and a judgment unit; the first acquisition unit is used to acquire a training signal group, the training signal group comprises input signals of a GNSS receiver, historical positioning information of the GNSS receiver, path propagation information of the input signals, and current trajectory information and velocity information of visible satellites, the path propagation information is delay information of the input signals from being emitted by the visible satellites to being received by the GNSS receiver, and the current trajectory information and velocity information of the visible satellites refer to the trajectory information and velocity information at the moment when the input signals are emitted by the visible satellites; the second acquisition unit is used to acquire a pseudo code in the GNSS receiver; the parameter setting unit is used to set a first coefficient according to the pseudo code, the first coefficient comprises the number and size of convolution kernels of convolution layers, and the size comprises the number of points of the convolution kernels and the arrangement of the number of points; the parameter setting unit is also used to set a first parameter, the first parameter comprises a first weight parameter and a first bias parameter of each point of the convolution kernels; the convolution processing unit is used to perform convolution processing on the input signals according to the first coefficient and the first parameter, so as to obtain feature information of the input signals; the parameter setting unit is also used to set a second coefficient and a second parameter, the second coefficient comprises the number of layers of the fully connected layers and the number of points of each layer of the fully connected layers, and the second parameter comprises a second weight parameter and a second bias parameter of each point of the fully connected layers; the positioning and velocity measurement result output unit is used to process the training signal group and the feature information according to the second coefficient and the second parameter, so as to obtain a training positioning and velocity measurement result of the GNSS receiver; and the judgment unit is used to judge whether the training positioning and velocity measurement result meets a first preset condition, if yes, the first coefficient, the second coefficient, the first parameter, and the second parameter are saved to generate a positioning and velocity measurement model, otherwise, the first parameter and the second parameter are modified to obtain a positioning and velocity measurement model meeting the first preset condition.

[0019] In a fourth aspect, the present application provides a data processing device, comprising an acquisition module, an input module, and an output module; the acquisition module is used to acquire input signals, the input module is used to input the input signals into a positioning and velocity measurement model obtained by the positioning and velocity measurement model training device in the third aspect, and the output module is used to acquire an output result of the positioning and velocity measurement model according to the input signals.

[0020] In a fifth aspect, the present application provides a storage medium having stored thereon a computer program, which, when executed by a processor, performs the steps of the training method of the positioning and speed measurement model according to any one of the first aspect or the data processing method according to the second aspect.

[0021] In a sixth aspect, the present application provides a terminal comprising a memory and a processor, the memory having stored thereon a computer program capable of being executed on the processor, the processor performing the steps of the training method of the positioning and speed measurement model according to any one of the first aspect or the data processing method according to the second aspect when executing the computer program.

[0022] The advantages of the above-mentioned second to sixth aspects can be referred to the description of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 A schematic diagram of a conventional GNSS receiver chip structure in the prior art;

[0024] Figure 2 A schematic diagram of a vector GNSS receiver structure in the prior art;

[0025] Figure 3 A flowchart of an embodiment of the training method of the positioning and speed measurement model provided by the present application;

[0026] Figure 4 A flowchart of an embodiment of the data processing method provided by the present application;

[0027] Figure 5 A schematic diagram of an embodiment of the training device of the positioning and speed measurement model provided by the present application;

[0028] Figure 6 A schematic diagram of an embodiment of the data processing device provided by the present application;

[0029] Figure 7 A schematic diagram of an embodiment of the terminal provided by the present application. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. In the description of the embodiments of the present application, the terms used in the following embodiments are only for the purpose of describing the specific embodiments and are not intended to be limiting to the present application. As used in the specification and the appended claims of the present application, the singular forms “a,” “the,” “said,” and “this” are intended to include the plural forms as well, e.g., “one or more,” unless the context clearly indicates otherwise. It will be further understood that the terms “comprises,” “comprising,” “includes,” and “including,” when used in the following embodiments of the present application, specify the presence of the stated features, integers, steps, actions, or components, but do not preclude the presence or addition of one or more other features, integers, steps, actions, or components. The term “and / or” used in the description of the present application refers to the association relationship of the associated objects, which means that there can be three kinds of relationships; for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural. The character “ / ” generally represents an “or” relationship between the associated objects.

[0031] In the present specification, the reference to “one embodiment” or “some embodiments” etc. means that a particular feature, structure or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Therefore, the statements “in one embodiment”, “in some embodiments”, “in other some embodiments”, “in yet some embodiments” etc. appearing in different places in the present specification are not necessarily all referring to the same embodiment, but mean “one or more but not all embodiments”, unless otherwise specifically emphasized. The terms “include”, “comprise”, “have” and their variants mean “including but not limited to”, unless otherwise specifically emphasized. The term “connection” includes direct connection and indirect connection, unless otherwise specified. “First”, “second” are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features.

[0032] In the embodiments of the present application, the words “exemplarily” or “for example” are used to mean as an example, illustration or description. Any embodiment or design scheme described as “exemplarily” or “for example” in the embodiments of the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the words “exemplarily” or “for example” are used to present the relevant concept in a specific manner.

[0033] The present application provides a model training method and device, a processing method and device, equipment and a medium, to solve the technical problems of complexity of algorithm design and difficulty of algorithm optimization for solving the positioning and speed measurement results of a carrier in the prior art.

[0034] The present application provides a positioning and speed measurement model training method applied to a GNSS receiver, and the flowchart is as followsFigure 3 As shown, comprising:

[0035] S101: Obtain a training signal group, the training signal group comprising an input signal of a GNSS receiver, historical positioning information of the GNSS receiver, path propagation information of the input signal, and current trajectory information and speed information of a visible satellite, the path propagation information being delay information of the input signal from being emitted by the visible satellite to being received by the GNSS receiver, the current trajectory information and speed information of the visible satellite referring to trajectory information and speed information at a moment when the visible satellite emits the input signal; the delay information comprising ionospheric delay information and tropospheric delay information of the input signal from being emitted by the visible satellite to being received by the GNSS receiver;

[0036] S102: Obtain a pseudo code in the GNSS receiver, and set a first coefficient according to the pseudo code, the first coefficient comprising a number and a size of a convolution kernel of a convolution layer, the size comprising a point number of the convolution kernel and an arrangement of the point number;

[0037] S103: Set a first parameter, the first parameter comprising a first weight parameter and a first bias parameter of each point of the convolution kernel;

[0038] S104: Perform convolution processing on the input signal according to the first coefficient and the first parameter, to obtain feature information of the input signal;

[0039] S105: Set a second coefficient and a second parameter, the second coefficient comprising a number of layers of the fully connected layer and a point number of each layer of the fully connected layer, the second parameter comprising a second weight parameter and a second bias parameter of each point of the fully connected layer;

[0040] S106: Perform processing on the training signal group and the feature information according to the second coefficient and the second parameter, to obtain a training positioning and velocity measurement result of the GNSS receiver;

[0041] S107: Determine whether the training positioning and velocity measurement result meets a first preset condition, if yes, save the first coefficient, the second coefficient, the first parameter, and the second parameter, to generate a positioning and velocity measurement model; otherwise, modify the first parameter and the second parameter, to obtain a positioning and velocity measurement model meeting the first preset condition.

[0042] The positioning and velocity measurement model obtained through the embodiment simplifies a data processing process while ensuring accuracy of an output positioning and velocity measurement result, and solves technical problems of complexity of algorithm design and difficulty of algorithm optimization for solving a positioning and velocity measurement result of a carrier in the prior art.

[0043] In some embodiments, the setting the first coefficient according to the pseudo code comprises: setting the number of convolution kernels of a first layer in the convolution layer according to the number of satellites of the visible satellites; setting the size of the convolution kernel of the first layer according to the length of the pseudo code; and setting the number and size of the convolution kernels of the remaining layers in the convolution layer according to the output dimension of the previous layer. Generally, the more the number of convolution layers is, the stronger the ability of the positioning and velocity measurement model to extract nonlinear features is, especially in the scenario of multipath interference, the feature extraction capability can be improved by increasing the number of convolution layers. However, the number of convolution layers cannot be infinite, and the specific design needs to be determined according to the actual situation.

[0044] In some embodiments, the processing the training signal group, the feature information according to the second coefficient and the second parameter to obtain the training positioning and velocity measurement result of the carrier comprises: calculating the sum of a first product and the second bias parameter to obtain the first motion trajectory and the first motion velocity of the GNSS receiver, the first product being obtained by multiplying the historical positioning information, the feature information, and the path propagation information by the second weight parameter corresponding to each point of each layer in the full connection layer. The full connection layer mainly describes the spatial correlation between the feature information of each satellite and the time correlation before and after, and if the randomness of the carrier motion is considered, the number of full connection layers can be appropriately increased to enhance the learning ability of the network, but the number of full connection layers cannot be infinite, and the specific design needs to be determined according to the actual situation. For example, in the process of solving the first motion trajectory, the historical positioning information A corresponds to the first weight coefficient a and the first bias parameter b, the feature information B corresponds to the second weight coefficient c and the second bias parameter d, the path information C in the path propagation information corresponds to the third weight coefficient e and the third bias parameter f, and the time information D in the path propagation information corresponds to the fourth weight coefficient g and the fourth bias parameter h. Then, the first motion trajectory and the first motion velocity are obtained by solving the formula Aa+b+Bc+d+Ce+f+Dg+h. The solving process of the first motion velocity is the same, except that the weight coefficients and bias parameters corresponding to the historical positioning information, the feature information, the path information, and the time information in the path propagation information may be different.

[0045] In some embodiments, before judging whether the training positioning and velocity measurement result meets the first preset condition, the method further comprises: setting a first coefficient and a second coefficient, the value of the first coefficient being 0-1, the value of the second coefficient being 0-1, and the sum of the first coefficient and the second coefficient being 1. The beneficial effect is that by setting the first coefficient and the second coefficient, the subsequent judgment process of the training positioning and velocity measurement result is optimized.

[0046] In some embodiments, before the judging whether the training positioning and speed measurement result meets the first preset condition, further comprising: obtaining an actual positioning and speed measurement result of the carrier, the actual positioning and speed measurement result is obtained through actual observation, and the actual positioning and speed measurement result comprises a second motion trajectory and a second motion speed.

[0047] In some embodiments, the judging whether the training positioning and speed measurement result meets the first preset condition comprises: obtaining a first deviation and a second deviation, the first deviation is a trajectory deviation of the first motion trajectory compared with the second motion trajectory, and the second deviation is a difference between the first motion speed and the second motion speed; obtaining a sum of a first product and a second product, the first product is a product of the first coefficient and the first deviation, and the second product is a product of the second coefficient and the second deviation; judging whether the sum of the first product and the second product is less than a preset threshold value, if yes, the training positioning and speed measurement result meets the first preset condition; otherwise, the training positioning and speed measurement result does not meet the first preset condition, and the parameters of the convolution kernel and the parameters of the full connection layer are retrained. Optionally, the first coefficient is 0.3, 0.4, 0.5, 0.6, 0.8 or 0.9, and the second coefficient is 0.3, 0.4, 0.5, 0.6, 0.8 or 0.9.

[0048] In some embodiments, before the obtaining the training signal group, comprising: establishing a data set, the data set comprising a training set and a test set, and the training set and the test set each comprising a plurality of training signal groups; and the obtaining the training signal group comprises: obtaining a training signal group in the training set. The data in the data set can be actually collected data or generated data through a simulator, and then the data needs to be normalized and randomly processed, and then divided into the training set and the test set according to a certain proportion.

[0049] In some embodiments, after the generating the positioning and speed measurement model, further comprising: obtaining a training signal group in the test set; obtaining an output result of the positioning and speed measurement model according to the training signal group in the test set, judging whether the output result meets a second preset condition, if yes, completing the training of the positioning test model; otherwise, modifying the first parameter and the second parameter to obtain a positioning and speed measurement model meeting the second preset condition.

[0050] Based on the training method of the positioning and speed measurement model in any of the above embodiments, the application provides a data processing method, and a flow thereof is shown in Figure 4 as follows.

[0051] S201: obtaining an input signal;

[0052] S202: input the input signal into the positioning and speed measurement model obtained by the training method of the positioning and speed measurement model according to any one of the above embodiments;

[0053] S203: obtain the output result of the positioning and speed measurement model according to the input signal.

[0054] Based on the training method of the positioning and speed measurement model according to any one of the above embodiments, the application further provides a training device of the positioning and speed measurement model, such as Figure 5As shown, the method for training the positioning and speed measurement model according to any one of the above is used to execute the method, which comprises a first acquisition unit 301, a second acquisition unit 302, a convolution processing unit 303, a parameter setting unit 304, a positioning and speed measurement result generation unit 305, and a judgment unit 306. The first acquisition unit 301 is used to acquire a training signal group, which comprises an input signal of a GNSS receiver, historical positioning information of the GNSS receiver, path propagation information of the input signal, and current trajectory information and speed information of a visible satellite. The path propagation information is delay information of the input signal from being transmitted by the visible satellite to being received by the GNSS receiver. The current trajectory information and speed information of the visible satellite refer to the trajectory information and speed information at the moment when the input signal is transmitted by the visible satellite. The second acquisition unit 302 is used to acquire a pseudo code in the GNSS receiver. The parameter setting unit 304 is used to set a first coefficient according to the pseudo code, wherein the first coefficient comprises the number and size of convolution kernels of a convolution layer, and the size comprises the number of points of the convolution kernel and the arrangement of the number of points. The parameter setting unit 304 is further used to set a first parameter, wherein the first parameter comprises a first weight parameter and a first bias parameter of each point of the convolution kernel. The convolution processing unit 303 is used to perform convolution processing on the input signal according to the first coefficient and the first parameter, so as to obtain feature information of the input signal. The parameter setting unit 304 is further used to set a second coefficient and a second parameter, wherein the second coefficient comprises the number of layers of the full connection layer and the number of points of each layer of the full connection layer, and the second parameter comprises a second weight parameter and a second bias parameter of each point of the full connection layer. The positioning and speed measurement result generation unit 305 is used to process the training signal group and the feature information according to the second coefficient and the second parameter, so as to obtain a training positioning and speed measurement result of the GNSS receiver. The judgment unit 306 is used to judge whether the training positioning and speed measurement result meets a first preset condition. If yes, the first coefficient, the second coefficient, the first parameter, and the second parameter are saved, so as to generate a positioning and speed measurement model. If not, the first parameter and the second parameter are modified, so as to obtain a positioning and speed measurement model meeting the first preset condition. All related contents of each step involved in the above method embodiment can be cited to the function description of the corresponding unit module, which will not be described here.

[0055] Based on the training device of the positioning and speed measurement model provided in the embodiment, the present application provides a data processing device, which comprises Figure 6As shown, the terminal can include one or more processors 501, a memory 502, a display 503, one or more application programs (not shown), and one or more computer programs 504, wherein the one or more computer programs 504 are stored in the memory 502 and configured to be executed by the one or more processors 501, and the one or more computer programs 504 include instructions that can be used to execute the steps in the respective embodiments described above.

[0056] Through tests, it is found that the accuracy of the output result (indicating the positioning and speed measurement result) of the positioning and speed measurement model according to the input signal provided by the present application is consistent with the accuracy of the output result of the conventional GNSS receiver, and the processing speed of the present application is faster than that of the conventional GNSS receiver, that is, the processing efficiency is improved.

[0057] In some other embodiments of the present application, the terminal can refer to the training device or the data processing device of the positioning and speed measurement model in the above-described embodiments, such as Figure 7 As shown, the terminal can include one or more processors 501, a memory 502, a display 503, one or more application programs (not shown), and one or more computer programs 504, wherein the one or more computer programs 504 are stored in the memory 502 and configured to be executed by the one or more processors 501, and the one or more computer programs 504 include instructions that can be used to execute the steps in the respective embodiments described above.

[0058] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is taken as an example for illustration, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0059] The functional units in each of the embodiments of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0060] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application essentially or in other words the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the method of the embodiments of the present application. The foregoing storage medium includes: a flash memory, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk, and various media that can store program codes.

[0061] The above is only a specific implementation of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any change or replacement within the technical scope disclosed by the embodiments of the present application should be covered in the protection scope of the embodiments of the present application. Therefore, the protection scope of the embodiments of the present application should be subject to the protection scope of the claims.

Claims

1. A training method for a positioning and speed measurement model, characterized in that, The application is applied to a GNSS receiver, comprising: obtaining a training signal group, the training signal group comprising an input signal of the GNSS receiver, historical positioning information of the GNSS receiver, path propagation information of the input signal, and current trajectory information and speed information of a visible satellite, the path propagation information being delay information of the input signal from being emitted by the visible satellite to being received by the GNSS receiver, the current trajectory information and speed information of the visible satellite referring to trajectory information and speed information at a moment when the visible satellite emits the input signal; obtaining a pseudo code in the GNSS receiver, and setting a first coefficient according to the pseudo code, comprising: setting a number of convolution kernels of a first layer in a convolution layer according to a number of satellites of the visible satellite; setting a size of the convolution kernel of the first layer according to a length of the pseudo code; setting a number and a size of convolution kernels of the remaining layers in the convolution layer according to an output dimension of a previous layer; the first coefficient comprising the number and the size of the convolution kernels of the convolution layer, the size comprising a number of points of the convolution kernel and an arrangement of the number of points; setting a first parameter, the first parameter comprising a first weight parameter and a first bias parameter of each point of the convolution kernel; performing convolution processing on the input signal according to the first coefficient and the first parameter to obtain feature information of the input signal; setting a second coefficient and a second parameter, the second coefficient comprising a number of layers of a full connection layer and a number of points of each layer of the full connection layer, the second parameter comprising a second weight parameter and a second bias parameter of each point of the full connection layer; processing the training signal group and the feature information according to the second coefficient and the second parameter to obtain a training positioning and velocity measurement result of the GNSS receiver; judging whether the training positioning and velocity measurement result meets a first preset condition, if yes, saving the first coefficient, the second coefficient, the first parameter and the second parameter to generate a positioning and velocity measurement model; otherwise, modifying the first parameter and the second parameter to obtain a positioning and velocity measurement model meeting the first preset condition.

2. The method of claim 1, wherein, the processing of the training signal group and the feature information according to the second coefficient and the second parameter to obtain the training positioning and velocity measurement result of the carrier, comprising: calculating a sum of a first product and the second bias parameter to obtain a first motion trajectory and a first motion speed of the GNSS receiver, the first product being obtained by multiplying the historical positioning information, the feature information and the path propagation information by a second weight parameter corresponding to each point of each layer of the full connection layer.

3. The method of claim 2, wherein, Before the judging whether the training positioning and velocity measurement result meets the first preset condition, further comprising: setting the first coefficient and the second coefficient, the first coefficient taking a value of 0-1, the second coefficient taking a value of 0-1, and a sum of the first coefficient and the second coefficient being 1.

4. The method of claim 3, wherein, Before the judging whether the training positioning and velocity measurement result meets the first preset condition, further comprising: obtaining an actual positioning and velocity measurement result of the GNSS receiver, the actual positioning and velocity measurement result being obtained by actual observation, the actual positioning and velocity measurement result comprising a second motion trajectory and a second motion velocity.

5. The method of claim 4, wherein, The determining whether the training positioning and velocity measurement result meets a first preset condition comprises: obtaining a first deviation and a second deviation, the first deviation being a trajectory deviation of the first motion trajectory relative to the second motion trajectory, and the second deviation being a difference between the first motion velocity and the second motion velocity; obtaining a sum of a first product and a second product, the first product being a product of the first coefficient and the first deviation, and the second product being a product of the second coefficient and the second deviation; determining whether the sum of the first product and the second product is less than a preset threshold value, if yes, the training positioning and velocity measurement result meets the first preset condition; otherwise, the training positioning and velocity measurement result does not meet the first preset condition.

6. The method of claim 1, wherein, Before obtaining the training signal group, the method comprises: establishing a data set, the data set comprising a training set and a test set, the training set and the test set each comprising a plurality of training signal groups; The obtaining the training signal group comprises:

7. The method of claim 6, wherein, obtaining a training signal group in the training set. After generating the positioning and velocity measurement model, the method further comprises: obtaining a training signal group in the test set; 8. A data processing method, characterized by, obtaining an output result of the positioning and velocity measurement model according to the training signal group in the test set, determining whether the output result meets a second preset condition, if yes, completing the training of the positioning and velocity measurement model; otherwise, modifying the first parameter and the second parameter to obtain a positioning and velocity measurement model meeting the second preset condition. The method comprises:

9. A training apparatus for positioning a speed model, characterized by obtaining an input signal and inputting the input signal into a positioning and velocity measurement model obtained by the training method of the positioning and velocity measurement model according to any one of claims 1-7, and obtaining an output result of the positioning and velocity measurement model according to the input signal. The method for executing the training method of the positioning and velocity measurement model according to any one of claims 1-7 comprises a first obtaining unit, a second obtaining unit, a parameter setting unit, a convolution processing unit, a positioning and velocity measurement result generating unit, and a determining unit. The first obtaining unit is configured to obtain a training signal group, the training signal group comprising an input signal of a GNSS receiver, historical positioning information of the GNSS receiver, path propagation information of the input signal, and current trajectory information and velocity information of a visible satellite, the path propagation information being delay information of the input signal from being emitted by the visible satellite to being received by the GNSS receiver, and the current trajectory information and velocity information of the visible satellite referring to trajectory information and velocity information at a moment when the input signal is emitted by the visible satellite. The second acquisition unit is configured to acquire a pseudo code in the GNSS receiver; and the parameter setting unit is configured to set a first coefficient according to the pseudo code, including: setting a number of convolution kernels in a first layer in the convolution layer according to a number of satellites of the visible satellites; setting a size of the convolution kernel in the first layer according to a length of the pseudo code; setting a number and a size of convolution kernels in the remaining layers in the convolution layer according to an output dimension of a previous layer; the first coefficient including the number and the size of the convolution kernels in the convolution layer, the size including a number of points of the convolution kernel and an arrangement of the points; and the parameter setting unit is further configured to set a first parameter, the first parameter including a first weight parameter and a first bias parameter of each point of the convolution kernel. The convolution processing unit is configured to perform convolution processing on the input signal according to the first coefficient and the first parameter, to obtain feature information of the input signal. The parameter setting unit is further configured to set a second coefficient and a second parameter, the second coefficient including a number of layers of the fully connected layer and a number of points of each layer of the fully connected layer, and the second parameter including a second weight parameter and a second bias parameter of each point of the fully connected layer; and the positioning and velocity measurement result output unit is configured to perform processing on a training signal group and the feature information according to the second coefficient and the second parameter, to obtain a training positioning and velocity measurement result of the GNSS receiver. The judging unit is configured to judge whether the training positioning and velocity measurement result meets a first preset condition, if yes, save the first coefficient, the second coefficient, the first parameter and the second parameter, to generate a positioning and velocity measurement model; otherwise, modify the first parameter and the second parameter, to obtain a positioning and velocity measurement model meeting the first preset condition.

10. A data processing apparatus, characterized by, The method comprises an acquisition module, an input module and an output module. The acquisition module is configured to acquire an input signal; the input module is configured to input the input signal into a positioning and velocity measurement model obtained by the training device of the positioning and velocity measurement model in claim 9; and the output module is configured to acquire an output result of the positioning and velocity measurement model according to the input signal.

11. A storage medium having stored thereon a computer program, characterized in that The computer program, when executed by a processor, performs the steps of the training method of the positioning and velocity measurement model in any one of claims 1 to 7 or the data processing method in claim 8. 12.A terminal comprising a memory and a processor, the memory having stored thereon a computer program capable of running on the processor, characterized in that, The computer program, when executed by a processor, performs the steps of the training method of the positioning and velocity measurement model in any one of claims 1 to 7 or the data processing method in claim 8.

Images