A global navigation satellite system positioning method, device and terminal

Abstract

This invention provides a Global Navigation Satellite System (GNSS) positioning method, apparatus, and terminal. The method includes: preprocessing acquired positioning observation data to obtain a parameter matrix; robustly processing the observation variance matrix in the parameter matrix to obtain a robust observation variance matrix; adaptively processing the target prediction variance matrix based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix to obtain an adaptively processed prediction variance matrix; and filtering the robust observation variance matrix and the adaptively processed prediction variance matrix to obtain GNSS position and velocity parameters. The prediction residual matrix is ​​determined based on the parameter matrix. This invention can eliminate the influence of spurious noise data during the adaptive positioning process, thereby improving positioning accuracy.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a global navigation satellite system positioning method, apparatus and terminal. Background Technology

[0002] Location information is closely related to modern human life, and with the advent of the Internet of Things era, the demand for high-precision location information is increasing. Smartphones are a crucial platform for providing location services, and research on high-precision positioning methods for smartphones has become a key focus of scientific research. Currently, GNSS data acquisition interfaces for smartphone devices have been opened, allowing developers to directly obtain raw pseudorange, carrier, and Doppler observations from the smartphone platform and use these values ​​for high-precision positioning calculations.

[0003] In the positioning calculation process, the data noise is modeled using a zero-mean normal distribution, and subsequent calculations are all extended based on this. However, due to differences in actual data, the modeling results may have certain deviations, leading to positioning errors. Therefore, a robust adaptive calculation process is needed during positioning to mitigate the impact of model mismatch and improve positioning accuracy. However, in existing technologies, the adaptive process is easily affected by pseudorange observation data noise, resulting in poor positioning accuracy. Summary of the Invention

[0004] This invention provides a global navigation satellite system positioning method, apparatus, and terminal to address the problem in the prior art where positioning accuracy is poor due to the influence of pseudorange observation data noise during the adaptive positioning process.

[0005] To address the aforementioned technical problems, the embodiments of the present invention provide the following technical solutions:

[0006] This invention provides a global navigation satellite system positioning method, comprising:

[0007] The acquired positioning observation data is preprocessed to obtain the parameter matrix;

[0008] The observation variance matrix in the parameter matrix is ​​subjected to robust processing to obtain the robust observation variance matrix;

[0009] Based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix, the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed prediction variance matrix.

[0010] The robust observation variance matrix and the adaptive prediction variance matrix are filtered to obtain the GNSS position and velocity parameters of the Global Navigation Satellite System.

[0011] The prediction residual matrix is ​​determined based on the parameter matrix.

[0012] Optionally, based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix, the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed prediction variance matrix, including:

[0013] Determine the carrier residual submatrix and the Doppler residual submatrix in the prediction residual matrix;

[0014] Based on the carrier residual submatrix, determine the first adaptive factor of the first target parameter;

[0015] Based on the Doppler residual submatrix, determine the second adaptive factor of the second target parameter;

[0016] Based on the first adaptive factor, the first sub-prediction variance of the first target parameter in the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed first sub-prediction variance.

[0017] Based on the second adaptive factor, the second sub-prediction variance of the second target parameter in the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed second sub-prediction variance.

[0018] The adaptively processed prediction variance matrix is ​​obtained based on the first sub-prediction variance and the second sub-prediction variance after adaptive processing.

[0019] The first target parameter includes a position parameter and an ambiguity parameter; the second target parameter includes a velocity parameter.

[0020] Optionally, the positioning observation data includes ephemeris data;

[0021] The parameter matrix includes: the design matrix, the observation variance matrix, and the observation residual matrix;

[0022] The observation variance matrix in the parameter matrix is ​​robustly processed to obtain a robust observation variance matrix, including:

[0023] The outlier data in the prediction residual matrix is ​​removed by the quartile method to obtain the prediction residual matrix after removing outlier data.

[0024] Based on the prediction residual matrix after excluding outlier data, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix.

[0025] The prediction residual matrix is ​​determined based on the design matrix, the observation residual matrix, and the parameter calculation results of the previous epoch.

[0026] The parameter calculation result of the previous epoch is determined based on the ephemeris data.

[0027] Optionally, outlier data in the prediction residual matrix are excluded using the quartile method to obtain a prediction residual matrix after excluding outlier data, including:

[0028] Using the data corresponding to the first upper quartile and the data corresponding to the first lower quartile as the boundaries of the sorted prediction residual matrix, the data of the prediction residual matrix outside the boundaries are excluded, and the prediction residual matrix after excluding the data outside the boundaries is obtained.

[0029] The data corresponding to the second upper quartile and the data corresponding to the second lower quartile are used as the boundary of the prediction residual matrix after excluding the data outside the boundary, and the data of the prediction residual matrix outside the boundary are excluded.

[0030] If the number of exclusions exceeds a preset number, or if the difference between the data corresponding to the upper quartile and the data corresponding to the lower quartile is less than a preset threshold, the exclusion ends, and the prediction residual matrix after excluding outlier data is obtained.

[0031] The upper quartiles include the first upper quartile and the second upper quartile; the lower quartiles include the first lower quartile and the second lower quartile.

[0032] Optionally, the method is characterized in that, based on the prediction residual matrix after excluding outlier data, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix, including:

[0033] The prediction residual matrix after excluding outlier data is standardized to obtain the standardized residual matrix;

[0034] Based on the standardized residual matrix, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix.

[0035] Optionally, based on the standardized residual matrix, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix, including:

[0036] The magnification factor is determined based on the standardized residual matrix, the preset robustness scheme, and the preset empirical coefficients.

[0037] The observation variance matrix is ​​expanded according to the expansion factor to obtain the robust observation variance matrix.

[0038] Optionally, the scaling factor is determined based on the standardized residual matrix, the preset robustness scheme, and the preset empirical coefficients, including:

[0039] Based on the preset robustness scheme and preset empirical coefficients, the data in the standardized residual matrix are divided into effective residual data, usable residual data, and harmful residual data.

[0040] Based on the preset empirical coefficients and the standardized residual matrix, the first magnification factor corresponding to the effective residual data, the second magnification factor corresponding to the usable residual data, and the third magnification factor corresponding to the harmful residual data are determined respectively.

[0041] Optionally, the data of the observed variance matrix are expanded according to the expansion factor, including:

[0042] Based on the first magnification factor, the data of the observation variance matrix corresponding to the effective residual data is magnified;

[0043] The data of the observation variance matrix corresponding to the available residual data is expanded according to the second expansion factor;

[0044] The observation variance matrix corresponding to the harmful residual data is expanded according to the third expansion factor.

[0045] This invention also provides a global navigation satellite system positioning device, comprising:

[0046] The first processing module is used to preprocess the acquired positioning observation data to obtain a parameter matrix;

[0047] The second processing module is used to perform robust processing on the observation variance matrix in the parameter matrix to obtain a robust observation variance matrix.

[0048] The third processing module is used to adaptively process the target prediction variance matrix based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix to obtain the adaptively processed prediction variance matrix.

[0049] The fourth processing module is used to filter the robust observation variance matrix and the adaptive prediction variance matrix to obtain the GNSS position and velocity parameters of the Global Navigation Satellite System.

[0050] The prediction residual matrix is ​​determined based on the parameter matrix.

[0051] Optionally, the third processing module includes:

[0052] The first determining unit is used to determine the carrier residual submatrix and the Doppler residual submatrix in the prediction residual matrix;

[0053] The second determining unit is used to determine the first adaptive factor of the first target parameter based on the carrier residual sub-matrix.

[0054] The third determining unit is used to determine the second adaptive factor of the second target parameter based on the Doppler residual sub-matrix.

[0055] The first processing unit is configured to adaptively process the first sub-prediction variance of the first target parameter in the target prediction variance matrix according to the first adaptive factor, so as to obtain the first sub-prediction variance after adaptive processing.

[0056] The second processing unit is configured to adaptively process the second sub-prediction variance of the second target parameter in the target prediction variance matrix according to the second adaptive factor, and obtain the adaptively processed second sub-prediction variance.

[0057] The first acquisition unit is used to obtain the adaptively processed prediction variance matrix based on the adaptively processed first sub-prediction variance and the adaptively processed second sub-prediction variance.

[0058] The first target parameter includes a position parameter and an ambiguity parameter; the second target parameter includes a velocity parameter.

[0059] Optionally, the positioning observation data includes ephemeris data;

[0060] The parameter matrix includes: the design matrix, the observation variance matrix, and the observation residual matrix;

[0061] The second processing module includes:

[0062] The third processing unit is used to exclude outlier data in the prediction residual matrix using the quartile method to obtain the prediction residual matrix after excluding outlier data.

[0063] The fourth processing unit is used to perform robust processing on the observation variance matrix based on the prediction residual matrix after excluding outlier data, to obtain the robust observation variance matrix.

[0064] The prediction residual matrix is ​​determined based on the design matrix, the observation residual matrix, and the parameter calculation results of the previous epoch.

[0065] The parameter calculation result of the previous epoch is determined based on the ephemeris data.

[0066] Optionally, the third processing unit is specifically used for:

[0067] Using the data corresponding to the first upper quartile and the data corresponding to the first lower quartile as the boundaries of the sorted prediction residual matrix, the data of the prediction residual matrix outside the boundaries are excluded, and the prediction residual matrix after excluding the data outside the boundaries is obtained.

[0068] The data corresponding to the second upper quartile and the data corresponding to the second lower quartile are used as the boundary of the prediction residual matrix after excluding the data outside the boundary, and the data of the prediction residual matrix outside the boundary are excluded.

[0069] If the number of exclusions exceeds a preset number, or if the difference between the data corresponding to the upper quartile and the data corresponding to the lower quartile is less than a preset threshold, the exclusion ends, and the prediction residual matrix after excluding outlier data is obtained.

[0070] The upper quartiles include the first upper quartile and the second upper quartile; the lower quartiles include the first lower quartile and the second lower quartile.

[0071] Optionally, the fourth processing unit is specifically used for:

[0072] The prediction residual matrix after excluding outlier data is standardized to obtain the standardized residual matrix;

[0073] Based on the standardized residual matrix, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix.

[0074] Optionally, the fourth processing unit is specifically used for:

[0075] The magnification factor is determined based on the standardized residual matrix, the preset robustness scheme, and the preset empirical coefficients.

[0076] The observation variance matrix is ​​expanded according to the expansion factor to obtain the robust observation variance matrix.

[0077] Optionally, the fourth processing unit is specifically used for:

[0078] Based on the preset robustness scheme and preset empirical coefficients, the data in the standardized residual matrix are divided into effective residual data, usable residual data, and harmful residual data.

[0079] Based on the preset empirical coefficients and the standardized residual matrix, the first magnification factor corresponding to the effective residual data, the second magnification factor corresponding to the usable residual data, and the third magnification factor corresponding to the harmful residual data are determined respectively.

[0080] Optionally, the fourth processing unit is specifically used for:

[0081] Based on the first magnification factor, the data of the observation variance matrix corresponding to the effective residual data is magnified;

[0082] The data of the observation variance matrix corresponding to the available residual data is expanded according to the second expansion factor;

[0083] The observation variance matrix corresponding to the harmful residual data is expanded according to the third expansion factor.

[0084] This invention also provides a terminal, comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the global navigation satellite system positioning method as described above.

[0085] This invention also provides a readable storage medium storing a program that, when executed by a processor, implements the steps of the global navigation satellite system positioning method as described above.

[0086] The beneficial effects of this invention are:

[0087] The present invention obtains positioning observation data, acquires a parameter matrix based on the positioning observation data, performs robust processing on the observation variance matrix in the parameter matrix, and adaptively processes the target prediction variance matrix in the filtering algorithm based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix to obtain position parameters and velocity parameters. This can eliminate the influence of spurious noise data and improve positioning accuracy during the adaptive positioning process. Attached Figure Description

[0088] Figure 1 One of the flowcharts represents a global navigation satellite system positioning method provided in an embodiment of the present invention;

[0089] Figure 2 A flowchart illustrating the adaptive processing provided in an embodiment of the present invention;

[0090] Figure 3 A flowchart illustrating the process of excluding outlier data from the prediction residual matrix provided in this embodiment of the invention;

[0091] Figure 4A flowchart illustrating the robustness processing provided in an embodiment of the present invention;

[0092] Figure 5 The second flowchart illustrates the global navigation satellite system positioning method provided in this embodiment of the invention.

[0093] Figure 6 This is a schematic diagram of the structure of the global navigation satellite system positioning device provided in an embodiment of the present invention;

[0094] Figure 7 This is a schematic diagram showing the structure of the terminal provided in an embodiment of the present invention. Detailed Implementation

[0095] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0096] This invention addresses the problem that the positioning accuracy is poor due to the influence of pseudorange observation data noise during the adaptive positioning process, and provides a global navigation satellite system positioning method, device, and terminal.

[0097] like Figure 1 As shown, this embodiment of the invention provides a global navigation satellite system positioning method, including:

[0098] Step 101: Preprocess the acquired positioning observation data to obtain the parameter matrix.

[0099] It should be noted that the global navigation satellite system positioning method provided in this embodiment of the invention can be applied to terminals, such as mobile terminals with camera or display functions, such as smartphones, wearable devices, smart TVs, etc., or the method can also be applied to servers, etc., and this embodiment of the invention is not limited thereto.

[0100] In this embodiment of the invention, the method of positioning using the Global Navigation Satellite System is described as an example using a smartphone.

[0101] Before preprocessing the acquired positioning observation data, the raw positioning observation data of the smartphone is obtained through the smartphone's GNSS data acquisition interface. The positioning observation data specifically includes time observation data, carrier observation data, observed satellite information, data accuracy, etc., and pseudo-observation data is calculated based on the acquired time observation data. The calculation method is as follows:

[0102] ρ=(t Rx -t Tx )·c

[0103] Where ρ represents the calculated pseudorange observation data value, and t Rx t represents the signal reception time. Txdenoted by , where represents the satellite signal transmission time, and c represents the speed of light in a vacuum.

[0104] The pseudorange observation data, carrier observation data, and Doppler observation data values ​​relative to satellite 's' acquired by a smartphone can be expressed in the following form:

[0105]

[0106] Among them, P s L s D s These represent the pseudorange observation, carrier wave observation, and Doppler observation values ​​of the smartphone relative to satellite 's', respectively. dT represents the clock difference of a mobile phone. s Indicates satellite clock bias. d orb ,d ion ,d trop These are orbital error, ionospheric delay error, and tropospheric delay error, respectively. ε represents the integer ambiguity of the carrier phase, and λ represents the wavelength. P ε L ε D These represent pseudorange, carrier, and Doppler observation noise, respectively. The superscript indicates the rate of change of the data.

[0107] After acquiring positioning observation data, the smartphone preprocesses the data. This preprocessing includes satellite coordinate calculation, clock error calculation, atmospheric error calculation, relativistic bias calculation, and Earth rotation bias calculation. During relative positioning, it is necessary to determine the common-view satellites between the reference station and the positioning observation data. After preprocessing the positioning observation data, a parameter matrix is ​​generated. Optionally, the parameter matrix includes: a design matrix B, an observation residual matrix l, and an observation variance matrix R.

[0108] Step 102: Perform robust processing on the observation variance matrix in the parameter matrix to obtain the robust observation variance matrix.

[0109] In this embodiment of the invention, in order to prevent the optimistic use of observation data within an epoch, it is necessary to make the observation variance matrix R more consistent with the actual situation of the observation values. That is, it is necessary to amplify the poor observation variance data to weaken its impact on the positioning effect, which means to perform robust processing on the observation variance matrix R.

[0110] In this embodiment of the invention, the observation variance matrix R is robustly processed by combining the prediction residual vector data in the prediction residual matrix.

[0111] Step 103: Based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix, perform adaptive processing on the target prediction variance matrix to obtain the adaptively processed prediction variance matrix.

[0112] The prediction residual matrix is ​​determined based on the parameter matrix.

[0113] In this embodiment of the invention, the unknown parameter prediction variance matrix (target prediction variance matrix) in the filtering process is adaptively processed. The adaptive process provided in this embodiment of the invention first separates the prediction residual matrix, extracts the pseudorange observation part, the carrier observation part (carrier residual sub-matrix), and the Doppler observation part (Doppler residual sub-matrix), and performs adaptive processing through the carrier residual sub-matrix and the Doppler residual sub-matrix, which can effectively avoid the influence of the pseudorange observation part on the positioning accuracy and improve the reliability of positioning.

[0114] Step 104: Filter the robust observation variance matrix and the adaptive prediction variance matrix to obtain the GNSS position and velocity parameters.

[0115] In this embodiment of the invention, the position and velocity parameters are calculated using a Kalman filter, or the Kalman filtering algorithm, based on the robust observation variance matrix and the adaptively processed prediction variance matrix. Specifically, the robust observation variance matrix and the adaptively processed prediction variance matrix are input into the Kalman filter for filtering to obtain accurate GNSS position and velocity parameters.

[0116] It should be noted that the method of inputting the robust observation variance matrix and the adaptive prediction variance matrix into the Kalman filter for filtering to obtain accurate GNSS position and velocity parameters is an existing technology and will not be elaborated here.

[0117] Optionally, based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix, the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed prediction variance matrix, including:

[0118] Determine the carrier residual submatrix and the Doppler residual submatrix in the prediction residual matrix;

[0119] Based on the carrier residual submatrix, determine the first adaptive factor of the first target parameter;

[0120] Based on the Doppler residual submatrix, determine the second adaptive factor of the second target parameter;

[0121] Based on the first adaptive factor, the first sub-prediction variance of the first target parameter in the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed first sub-prediction variance.

[0122] Based on the second adaptive factor, the second sub-prediction variance of the second target parameter in the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed second sub-prediction variance.

[0123] The adaptively processed prediction variance matrix is ​​obtained based on the first sub-prediction variance and the second sub-prediction variance after adaptive processing.

[0124] The first target parameter includes a position parameter and an ambiguity parameter; the second target parameter includes a velocity parameter.

[0125] The following details the process of adaptively processing the unknown parameter prediction variance matrix (target prediction variance matrix).

[0126] Generally, the variance of the estimated unknown parameters gradually decreases during Kalman filtering, but this result tends to be more ideal. In some cases, the calculated variance of the unknown parameters may be overly optimistic, failing to accurately represent the actual situation. During Kalman filtering updates, this can lead to an excessive proportion of predicted information, hindering parameter updates; therefore, adaptive processing is necessary. Specifically, in the adaptive case, the predicted variance matrix needs to be adjusted... Become The optimal adaptive factor α given in existing adaptive navigation and positioning k The calculation method and process are shown in the following formula:

[0127]

[0128] in, The theoretical variance matrix represents the predicted residual vector. R represents the actual calculated variance matrix of the predicted residual vector. k Let α be the observation variance matrix. k This is the optimal adaptive factor.

[0129] α k Approximate possible values ​​are:

[0130]

[0131] in, The theoretical variance matrix represents the predicted residual vector. α represents the actual calculated variance matrix of the predicted residual vector. k This is the optimal adaptive factor.

[0132] in,

[0133]

[0134] In the formula, the design matrix B k To design the matrix, To predict the variance matrix, R k Let v be the observed variance matrix and v be the predicted residual matrix. The theoretical variance matrix represents the predicted residual vector. This represents the actual calculated variance matrix of the predicted residual vector.

[0135] The calculation process for the prediction residual matrix v is as follows:

[0136] Suppose the solution to the unknown parameters in the previous epoch is X. k-1 The prediction variance matrix is ​​Q k-1 The recursive result of the parameter solution from the previous epoch to the current epoch is: The prediction variance matrix is The design matrix for the current epoch is B. k The observation residual matrix is ​​l k The prediction residual matrix v can then be expressed in the following form:

[0137]

[0138] It should be noted that during the positioning process of smartphones, due to the influence of pseudorange noise, the above-mentioned optimal adaptive factor α... k The calculation results mainly vary with the pseudorange observation data values, making it difficult to accurately represent the actual situation of the unknown parameter prediction variance matrix (target prediction variance matrix), thus resulting in poor practical application effects.

[0139] In this embodiment of the invention, the prediction residual matrix is ​​first separated, and the pseudorange observation portion, carrier observation portion (carrier residual sub-matrix), and Doppler observation portion (Doppler residual sub-matrix) are extracted. The carrier observation portion is used to calculate the first sub-adaptation factor of the first target parameter (including position and ambiguity parameters), and the Doppler observation portion is used to calculate the second adaptive factor of the second target parameter (velocity parameter). The first adaptive factor is used to expand the first sub-prediction variance of the first target parameter (including position and ambiguity parameters) in the target prediction variance matrix, and the second adaptive factor is used to expand the second sub-prediction variance of the second target parameter (velocity parameter) in the target prediction variance matrix. This method effectively avoids the influence of pseudorange observation data noise on the adaptive factor calculation and implements the target prediction variance expansion process step by step, effectively utilizing the properties of the data.

[0140] The following is combined with Figure 2 The adaptive processing flow will be explained in detail.

[0141] The covariance matrix of the unknown parameter prediction variance matrix (target prediction variance matrix) is decomposed, and the prediction residual matrix is ​​calculated. The theoretical variance of the prediction residual matrix is ​​calculated, and the carrier component (carrier residual sub-matrix) and the Doppler observation component (Doppler residual sub-matrix) are extracted from the prediction residual matrix respectively. A test value (first adaptive factor) is calculated based on the carrier residual sub-matrix and the theoretical carrier residual sub-matrix. A test value (second adaptive factor) is calculated based on the Doppler residual sub-matrix and the theoretical Doppler residual matrix. The prediction variances of the position parameter and ambiguity parameter are amplified based on the test value (first adaptive factor) (first sub-prediction variance), and the prediction variance of the velocity parameter is amplified based on the test value (second adaptive factor) (second sub-prediction variance). The amplified first and second sub-prediction variances are then recombined to obtain the adaptively processed prediction variance matrix.

[0142] Optionally, the positioning observation data includes ephemeris data;

[0143] The parameter matrix includes: the design matrix, the observation variance matrix, and the observation residual matrix;

[0144] The observation variance matrix in the parameter matrix is ​​robustly processed to obtain a robust observation variance matrix, including:

[0145] The outlier data in the prediction residual matrix is ​​removed by the quartile method to obtain the prediction residual matrix after removing outlier data.

[0146] Based on the prediction residual matrix after excluding outlier data, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix.

[0147] The prediction residual matrix is ​​determined based on the design matrix, the observation residual matrix, and the parameter calculation results of the previous epoch.

[0148] The parameter calculation result of the previous epoch is determined based on the ephemeris data.

[0149] It should be noted that the positioning observation data also includes acquired ephemeris data. Specifically, smartphones can also obtain ephemeris data from the IGS website; optionally, the ephemeris data can be obtained via an FTP site. Since the standardized residual matrix is ​​key to reasonably amplifying the observation variance matrix, outlier data can cause excessively large standard deviations during the residual matrix standardization process, resulting in smaller calculation results for some standardized residual matrices. In this embodiment of the invention, the quartile method is used to exclude outlier data from the prediction residual matrix, obtaining a prediction residual matrix after outlier exclusion. Based on the prediction residual matrix after outlier exclusion, robustness processing is applied to the observation variance matrix to obtain the robust observation variance matrix. Using the quartile method to exclude outlier data from the prediction residual matrix avoids recalculating errors in the sequence (data of the prediction residual matrix), allows for the detection of multiple outlier data points at once, and effectively avoids the influence of outlier data in the prediction residual matrix data, which is beneficial for robust positioning calculations.

[0150] The prediction residual matrix v can then be expressed in the following form:

[0151]

[0152] Among them, X k-1 Q is the result of solving for the unknown parameters in the previous epoch. k-1 To predict the variance matrix. This is the recursive result of the parameter solution from the previous epoch to the current epoch. To predict the variance matrix. B k For the design matrix of the current epoch, l k To observe the residual matrix.

[0153] Optionally, outlier data in the prediction residual matrix are excluded using the quartile method to obtain a prediction residual matrix after excluding outlier data, including:

[0154] The predicted residual matrix is ​​sorted to obtain the sorted predicted residual matrix;

[0155] Using the data corresponding to the first upper quartile and the data corresponding to the first lower quartile as the boundaries of the sorted prediction residual matrix, the data of the prediction residual matrix outside the boundaries are excluded, and the prediction residual matrix after excluding the data outside the boundaries is obtained.

[0156] The data corresponding to the second upper quartile and the data corresponding to the second lower quartile are used as the boundary of the prediction residual matrix after excluding the data outside the boundary, and the data of the prediction residual matrix outside the boundary are excluded.

[0157] If the number of exclusions exceeds a preset number, or if the difference between the data corresponding to the upper quartile and the data corresponding to the lower quartile is less than a preset threshold, the exclusion ends, and the prediction residual matrix after excluding outlier data is obtained.

[0158] The upper quartiles include the first upper quartile and the second upper quartile; the lower quartiles include the first lower quartile and the second lower quartile.

[0159] The following is combined with Figure 3 The process of excluding outliers from the prediction residual matrix using the quartile method is explained in detail below:

[0160] Input the data sequence (data from the prediction residual matrix), setting all data to unlabeled. Sort the unlabeled sequence, calculate the data corresponding to the upper quartile (first upper quartile) and the data corresponding to the lower quartile (first lower quartile), determine the data corresponding to the first upper quartile as the upper bound, and the data corresponding to the first lower quartile as the lower bound, label and exclude data outside the bounds, recalculate the data corresponding to the upper quartile (second upper quartile) and the lower quartile (second lower quartile), and redetermine the data corresponding to the second upper quartile as the lower bound. The upper bound is determined by re-establishing the data corresponding to the second lower quartile as the lower bound. Data outside the bound are marked again and excluded. If the difference between the upper bound data corresponding to the upper quartile (which can be the first upper quartile or the second upper quartile) and the lower bound data corresponding to the lower quartile (correspondingly, which can be the first lower quartile or the second lower quartile) is less than a preset threshold or the number of markings (exclusions) exceeds a preset number (e.g., 5 times), all marked data are determined as outliers to be excluded and excluded. After exclusion, the prediction residual matrix after excluding outliers is obtained.

[0161] Optionally, the method is characterized in that, based on the prediction residual matrix after excluding outlier data, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix, including:

[0162] The prediction residual matrix after excluding outlier data is standardized to obtain the standardized residual matrix;

[0163] Based on the standardized residual matrix, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix.

[0164] It should be noted that the solution result for the unknown parameters in the previous epoch is X. k-1 The prediction variance matrix is ​​Q k-1The recursive result of the parameter solution from the previous epoch to the current epoch is: The prediction variance matrix is The design matrix for the current epoch is B. k The observation residual matrix is ​​l k The aforementioned prediction residual matrix v can be expressed in the following form:

[0165]

[0166] In theory, the prediction residual matrix v should satisfy the following conditions:

[0167]

[0168] Among them, B k To design the matrix, To predict the variance matrix.

[0169] Theoretically, the predicted residual matrix vector data should follow a normal distribution with a mean of 0. However, due to the bias of unknown parameters, the predicted residual matrix vector data may exhibit small deviations. Standardizing the predicted residual matrix vector data yields the standardized residual matrix. It can determine the degree of deviation of the predicted residual vector data from the center, and amplify the data in the corresponding observation variance matrix according to the degree of deviation.

[0170] Optionally, based on the standardized residual matrix, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix, including:

[0171] The magnification factor is determined based on the standardized residual matrix, the preset robustness scheme, and the preset empirical coefficients.

[0172] The observation variance matrix is ​​expanded according to the expansion factor to obtain the robust observation variance matrix.

[0173] In this embodiment of the invention, the data in the observation variance matrix that is amplified according to the degree of deviation of the predicted residual vector data from the center is done by an amplification factor. The specific amplification process is as follows:

[0174]

[0175] Where, r ii R represents the observation variance matrix k The i-th diagonal element, γ ii Indicates the expansion factor. This represents the result after the variance elements are magnified.

[0176] The expansion factor is determined based on the standardized residual matrix, the preset robustness scheme, and the preset empirical coefficient.

[0177] Optionally, the scaling factor is determined based on the standardized residual matrix, the preset robustness scheme, and the preset empirical coefficients, including:

[0178] Based on the preset robustness scheme and preset empirical coefficients, the data in the standardized residual matrix are divided into effective residual data, usable residual data, and harmful residual data.

[0179] Based on the preset empirical coefficients and the standardized residual matrix, the first magnification factor corresponding to the effective residual data, the second magnification factor corresponding to the usable residual data, and the third magnification factor corresponding to the harmful residual data are determined respectively.

[0180] In this embodiment of the invention, according to the IGGⅢ robustness scheme, preset empirical coefficients k0 and k1 are set. k0 and k1 divide the standardized residual matrix array into three segments: effective residual data, usable residual data, and harmful residual data. A first expansion coefficient corresponding to the effective residual data, a second expansion coefficient corresponding to the usable residual data, and a third expansion coefficient corresponding to the harmful residual data are set respectively.

[0181] Optionally, the data of the observed variance matrix are expanded according to the expansion factor, including:

[0182] Based on the first magnification factor, the data of the observation variance matrix corresponding to the effective residual data is magnified;

[0183] The data of the observation variance matrix corresponding to the available residual data is expanded according to the second expansion factor;

[0184] The observation variance matrix corresponding to the harmful residual data is expanded according to the third expansion factor.

[0185] In this embodiment of the invention, based on the above-described IGGⅢ robustness scheme, we have:

[0186]

[0187] Where k0 and k1 are preset empirical coefficients, For the standardized residual matrix The i-th element.

[0188] In other words, the standardized residual matrix data is divided into three segments by pre-setting empirical coefficients k0 and k1: effective residual data, usable residual data, and harmful residual data. The data of the observation variance matrix corresponding to the effective residual data remains unchanged, the data of the observation variance matrix corresponding to the usable residual data is expanded to a certain extent, and the data of the observation variance matrix corresponding to the harmful residual data is expanded to the maximum.

[0189] The following is combined with Figure 4 The process of resistance treatment will be explained in detail.

[0190] The pseudorange prediction residual matrix, Doppler residual matrix, and carrier residual matrix in the prediction residual matrix are calculated. Outlier data in the pseudorange prediction residual matrix and Doppler residual matrix are excluded using the quartile method. Newly appearing satellite residual matrices are removed from the carrier residual matrix, and a threshold is set to exclude outlier data in the carrier prediction residual matrix. The error of the normalized residual matrix corresponding to the pseudorange prediction residual matrix after excluding outlier data is calculated. The error of the normalized residual matrix corresponding to the Doppler residual matrix after excluding outlier data is calculated. The error of the normalized residual matrix corresponding to the carrier residual matrix after excluding outlier data is calculated. The IGGIII scheme is used to amplify the data in the corresponding observation variance matrix.

[0191] The following is combined with Figure 5 This section details the process of using the Global Navigation Satellite System (GNSS) for smartphone positioning.

[0192] Within a single epoch, firstly, the smartphone's position and velocity parameters are updated. When the current epoch is the processing epoch, the position and velocity parameters obtained from the smartphone are used for initialization. Then, the obtained position and velocity parameters are preprocessed, including satellite coordinate calculation, clock error interpolation, and modeling error calculation, generating parameter matrices, including: design matrix, observation residual matrix, and observation variance matrix. The observation variance matrix is ​​robustly processed. When it is determined from the ephemeris data that the current epoch is not the initial epoch, the target prediction variance matrix is ​​adaptively processed, and then based on the ephemeris... When the data determines that the current epoch is the initial epoch, the position parameters are directly estimated using Kalman filtering. When the data determines that the current epoch is not the initial epoch, the position and velocity parameters are estimated using Kalman filtering based on the robust observation variance matrix and the adaptively processed target prediction variance matrix. The process then determines whether the current epoch is the final epoch. If the current epoch is the final epoch, the process ends. If the current epoch is not the final epoch, the position and velocity parameters for the next epoch are obtained. The above steps are repeated to achieve continuous updating of the position and velocity parameters.

[0193] To verify the effectiveness of the method described in this invention, four smartphones were used for vehicle-mounted data experiments. The smartphone model numbers and other information are shown in Table 1 below. During the test, a Huace topographic receiver was used to collect baseline data, and Huace's commercial calculation software CHO was used for data processing. The calculation results were used as baseline data and compared with the smartphone calculation results. The test was conducted on April 21, 2021, with a smartphone data sampling rate of 1 second and a collection time of 1100 epochs.

[0194] Table 1 Smartphone Test Information

[0195]

[0196] To specifically verify the advantages of this invention over existing robust and adaptive methods, different schemes were used to locate the collected data, and the positioning accuracy was compared. The calculation schemes used included: (a) positioning calculation without robustness and adaptation; (b) positioning calculation using existing robust and adaptive methods; and (c) positioning calculation using the method described in this invention. Finally, the positioning results of the three schemes were compared to verify the positioning performance of the method described in this invention. The calculation results of the three schemes are shown in Table 2 below.

[0197] Table 2 Calculation results for the three schemes

[0198]

[0199] As shown in Table 2 above, both Scheme (b) and Scheme (c) show improved positioning performance compared to Scheme (a). Comparing the planar positioning results and the elevation positioning results of Scheme (b) and Scheme (c) respectively, it can be seen that Scheme (c) consistently improves the accuracy of both planar and elevation positioning compared to Scheme (b).

[0200] The Global Navigation Satellite System (GNSS) positioning method provided in this invention, specifically a step-by-step robust adaptive calculation method for GNSS positioning on smartphones, considers the characteristics of smartphone data and effectively improves smartphone positioning accuracy compared to existing methods. It fully considers the properties of smartphone GNSS data, discarding pseudorange observations during the adaptive process and calculating adaptive factors using both carrier and Doppler observations. Then, it implements step-by-step adaptation of position and velocity variances, effectively improving computational reliability. The method employs quartiles to remove outliers from the prediction residuals before performing residual standardization and observation variance processing. This process weakens the influence of outliers and improves robustness and reliability. Furthermore, the proposed quartiles method for detecting outliers has lower complexity than existing methods, reducing the computational burden on the smartphone.

[0201] like Figure 6 As shown, this embodiment of the invention also provides a global navigation satellite system positioning device, comprising:

[0202] The first processing module 601 is used to preprocess the acquired positioning observation data to obtain a parameter matrix;

[0203] The second processing module 602 is used to perform robust processing on the observation variance matrix in the parameter matrix to obtain a robust observation variance matrix.

[0204] The third processing module 603 is used to adaptively process the target prediction variance matrix based on the carrier residual submatrix and the Doppler residual submatrix in the prediction residual matrix to obtain the adaptively processed prediction variance matrix.

[0205] The fourth processing module 604 is used to filter the robust observation variance matrix and the adaptive prediction variance matrix to obtain the GNSS position parameters and velocity parameters of the Global Navigation Satellite System.

[0206] The prediction residual matrix is ​​determined based on the parameter matrix.

[0207] The apparatus provided in this embodiment of the invention acquires positioning observation data, obtains a parameter matrix based on the positioning observation data, performs robust processing on the observation variance matrix in the parameter matrix, and adaptively processes the target prediction variance matrix in the filtering algorithm based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix to obtain position parameters and velocity parameters. This can eliminate the influence of spurious noise data and improve positioning accuracy during the adaptive positioning process.

[0208] Optionally, the third processing module 603 includes:

[0209] The first determining unit is used to determine the carrier residual submatrix and the Doppler residual submatrix in the prediction residual matrix;

[0210] The second determining unit is used to determine the first adaptive factor of the first target parameter based on the carrier residual sub-matrix.

[0211] Based on the Doppler residual submatrix, determine the second adaptive factor of the second target parameter;

[0212] Based on the first adaptive factor, the first sub-prediction variance of the first target parameter in the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed first sub-prediction variance.

[0213] Based on the second adaptive factor, the second sub-prediction variance of the second target parameter in the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed second sub-prediction variance.

[0214] The adaptively processed prediction variance matrix is ​​obtained based on the first sub-prediction variance and the second sub-prediction variance after adaptive processing.

[0215] The first target parameter includes a position parameter and an ambiguity parameter; the second target parameter includes a velocity parameter.

[0216] Optionally, the positioning observation data includes ephemeris data;

[0217] The parameter matrix includes: the design matrix, the observation variance matrix, and the observation residual matrix;

[0218] The second processing module 602 includes:

[0219] The third processing unit is used to exclude outlier data in the prediction residual matrix using the quartile method to obtain the prediction residual matrix after excluding outlier data.

[0220] The fourth processing unit is used to perform robust processing on the observation variance matrix based on the prediction residual matrix after excluding outlier data, to obtain the robust observation variance matrix.

[0221] The prediction residual matrix is ​​determined based on the design matrix, the observation residual matrix, and the parameter calculation results of the previous epoch.

[0222] The parameter calculation result of the previous epoch is determined based on the ephemeris data.

[0223] Optionally, the third processing unit is specifically used for:

[0224] The predicted residual matrix is ​​sorted to obtain the sorted predicted residual matrix;

[0225] Using the data corresponding to the first upper quartile and the data corresponding to the first lower quartile as the boundaries of the sorted prediction residual matrix, the data of the prediction residual matrix outside the boundaries are excluded, and the prediction residual matrix after excluding the data outside the boundaries is obtained.

[0226] The data corresponding to the second upper quartile and the data corresponding to the second lower quartile are used as the boundary of the prediction residual matrix after excluding the data outside the boundary, and the data of the prediction residual matrix outside the boundary are excluded.

[0227] If the number of exclusions exceeds a preset number, or if the difference between the data corresponding to the upper quartile and the data corresponding to the lower quartile is less than a preset threshold, the exclusion ends, and the prediction residual matrix after excluding outlier data is obtained.

[0228] The upper quartiles include the first upper quartile and the second upper quartile; the lower quartiles include the first lower quartile and the second lower quartile.

[0229] Optionally, the fourth processing unit is specifically used for:

[0230] The prediction residual matrix after excluding outlier data is standardized to obtain the standardized residual matrix;

[0231] Based on the standardized residual matrix, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix.

[0232] Optionally, the fourth processing unit is specifically used for:

[0233] The magnification factor is determined based on the standardized residual matrix, the preset robustness scheme, and the preset empirical coefficients.

[0234] The observation variance matrix is ​​expanded according to the expansion factor to obtain the robust observation variance matrix.

[0235] Optionally, the fourth processing unit is specifically used for:

[0236] Based on the preset robustness scheme and preset empirical coefficients, the data in the standardized residual matrix are divided into effective residual data, usable residual data, and harmful residual data.

[0237] Based on the preset empirical coefficients and the standardized residual matrix, the first magnification factor corresponding to the effective residual data, the second magnification factor corresponding to the usable residual data, and the third magnification factor corresponding to the harmful residual data are determined respectively.

[0238] Optionally, the fourth processing unit is specifically used for:

[0239] Based on the first magnification factor, the data of the observation variance matrix corresponding to the effective residual data is magnified;

[0240] The data of the observation variance matrix corresponding to the available residual data is expanded according to the second expansion factor;

[0241] The observation variance matrix corresponding to the harmful residual data is expanded according to the third expansion factor.

[0242] It should be noted that the global navigation satellite system positioning device provided in this embodiment of the invention is a device capable of performing the above-described global navigation satellite system positioning method. Therefore, both the left and right embodiments of the above-described global navigation satellite system positioning method are applicable to this device and can achieve the same or similar technical effects.

[0243] like Figure 7 As shown, this embodiment of the invention also provides a terminal, including: a processor 701, a memory 702, and a transceiver 703; the memory 702 is used to store program instructions; the transceiver 703 is used to send and receive data under the control of the processor 701; the processor 701 is used to read the program instructions in the memory 702 and perform the following operations:

[0244] The acquired positioning observation data is preprocessed to obtain the parameter matrix;

[0245] The observation variance matrix in the parameter matrix is ​​subjected to robust processing to obtain the robust observation variance matrix;

[0246] Based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix, the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed prediction variance matrix.

[0247] The robust observation variance matrix and the adaptive prediction variance matrix are filtered to obtain the GNSS position and velocity parameters of the Global Navigation Satellite System.

[0248] The prediction residual matrix is ​​determined based on the parameter matrix.

[0249] Optionally, the processor 701 is specifically used for:

[0250] Determine the carrier residual submatrix and the Doppler residual submatrix in the prediction residual matrix;

[0251] Based on the carrier residual submatrix, determine the first adaptive factor of the first target parameter;

[0252] Based on the Doppler residual submatrix, determine the second adaptive factor of the second target parameter;

[0253] Based on the first adaptive factor, the first sub-prediction variance of the first target parameter in the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed first sub-prediction variance.

[0254] Based on the second adaptive factor, the second sub-prediction variance of the second target parameter in the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed second sub-prediction variance.

[0255] The adaptively processed prediction variance matrix is ​​obtained based on the first sub-prediction variance and the second sub-prediction variance after adaptive processing.

[0256] The first target parameter includes a position parameter and an ambiguity parameter; the second target parameter includes a velocity parameter.

[0257] Optionally, the positioning observation data includes ephemeris data;

[0258] The parameter matrix includes: the design matrix, the observation variance matrix, and the observation residual matrix;

[0259] The processor 701 is specifically used for:

[0260] The outlier data in the prediction residual matrix is ​​removed by the quartile method to obtain the prediction residual matrix after removing outlier data.

[0261] Based on the prediction residual matrix after excluding outlier data, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix.

[0262] The prediction residual matrix is ​​determined based on the design matrix, the observation residual matrix, and the parameter calculation results of the previous epoch.

[0263] The parameter calculation result of the previous epoch is determined based on the ephemeris data.

[0264] Optionally, the processor 701 is specifically used for:

[0265] The predicted residual matrix is ​​sorted to obtain the sorted predicted residual matrix;

[0266] Using the data corresponding to the first upper quartile and the data corresponding to the first lower quartile as the boundaries of the sorted prediction residual matrix, the data of the prediction residual matrix outside the boundaries are excluded, and the prediction residual matrix after excluding the data outside the boundaries is obtained.

[0267] The data corresponding to the second upper quartile and the data corresponding to the second lower quartile are used as the boundary of the prediction residual matrix after excluding the data outside the boundary, and the data of the prediction residual matrix outside the boundary are excluded.

[0268] If the number of exclusions exceeds a preset number, or if the difference between the data corresponding to the upper quartile and the data corresponding to the lower quartile is less than a preset threshold, the exclusion ends, and the prediction residual matrix after excluding outlier data is obtained.

[0269] The upper quartiles include the first upper quartile and the second upper quartile; the lower quartiles include the first lower quartile and the second lower quartile.

[0270] Optionally, the processor 701 is specifically used for:

[0271] The prediction residual matrix after excluding outlier data is standardized to obtain the standardized residual matrix;

[0272] Based on the standardized residual matrix, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix.

[0273] Optionally, the processor 701 is specifically used for:

[0274] The magnification factor is determined based on the standardized residual matrix, the preset robustness scheme, and the preset empirical coefficients.

[0275] The observation variance matrix is ​​expanded according to the expansion factor to obtain the robust observation variance matrix.

[0276] Optionally, the processor 701 is specifically used for:

[0277] Based on the preset robustness scheme and preset empirical coefficients, the data in the standardized residual matrix are divided into effective residual data, usable residual data, and harmful residual data.

[0278] Based on the preset empirical coefficients and the standardized residual matrix, the first magnification factor corresponding to the effective residual data, the second magnification factor corresponding to the usable residual data, and the third magnification factor corresponding to the harmful residual data are determined respectively.

[0279] Optionally, the processor 701 is specifically used for:

[0280] Based on the first magnification factor, the data of the observation variance matrix corresponding to the effective residual data is magnified;

[0281] The data of the observation variance matrix corresponding to the available residual data is expanded according to the second expansion factor;

[0282] The observation variance matrix corresponding to the harmful residual data is expanded according to the third expansion factor.

[0283] Among them, Figure 7In this context, the bus architecture can include any number of interconnected buses and bridges, specifically linking various circuits of one or more processors represented by processor 701 and memory represented by memory 702 together. The bus architecture can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides an interface. The transceiver 703 can be multiple elements, including transmitters and receivers, providing a unit for communicating with various other devices over a transmission medium, including wireless channels, wired channels, optical fibers, etc. For different user equipment, the user interface 704 can also be an interface capable of connecting external or internal devices, including but not limited to keypads, displays, speakers, microphones, joysticks, etc.

[0284] The processor 701 is responsible for managing the bus architecture and general processing, while the memory 702 can store the data used by the processor 701 when performing operations.

[0285] The processor 701 executes any of the methods provided in the embodiments of the present invention according to the obtained executable instructions by calling program instructions stored in the memory. The processor 701 and the memory 702 may also be physically separated.

[0286] This invention also provides a readable storage medium storing a program that, when executed by a processor, implements the steps of the global navigation satellite system positioning method as described above.

[0287] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions that cause a computer device (which may be a personal computer, server, or network device, etc.) to execute some steps of the transmission and reception methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0288] The above describes the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also within the scope of protection of the present invention.

Claims

1. A global navigation satellite system positioning method, characterized in that, include: The acquired positioning observation data is preprocessed to obtain the parameter matrix; The observation variance matrix in the parameter matrix is ​​subjected to robust processing to obtain the robust observation variance matrix; Based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix, the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed prediction variance matrix; wherein, the prediction residual matrix is ​​determined based on the parameter matrix; The robust observation variance matrix and the adaptive prediction variance matrix are filtered to obtain the GNSS position and velocity parameters of the Global Navigation Satellite System. The positioning observation data includes ephemeris data; the parameter matrix includes: a design matrix, the observation variance matrix, and the observation residual matrix. The observation variance matrix in the parameter matrix is ​​robustly processed to obtain a robust observation variance matrix, including: The outlier data in the prediction residual matrix is ​​removed by the quartile method to obtain the prediction residual matrix after removing outlier data. Based on the prediction residual matrix after excluding outlier data, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix. The prediction residual matrix is ​​determined based on the design matrix, the observation residual matrix, and the parameter calculation results of the previous epoch. The parameter calculation result of the previous epoch is determined based on the ephemeris data.

2. The global navigation satellite system positioning method according to claim 1, characterized in that, Based on the carrier residual submatrix and Doppler residual submatrix in the prediction residual matrix, the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed prediction variance matrix, including: Determine the carrier residual submatrix and the Doppler residual submatrix in the prediction residual matrix; Based on the carrier residual submatrix, determine the first adaptive factor of the first target parameter; Based on the Doppler residual submatrix, determine the second adaptive factor of the second target parameter; Based on the first adaptive factor, the first sub-prediction variance of the first target parameter in the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed first sub-prediction variance. Based on the second adaptive factor, the second sub-prediction variance of the second target parameter in the target prediction variance matrix is ​​adaptively processed to obtain the adaptively processed second sub-prediction variance. The adaptively processed prediction variance matrix is ​​obtained based on the first sub-prediction variance and the second sub-prediction variance after adaptive processing. The first target parameter includes a position parameter and an ambiguity parameter; the second target parameter includes a velocity parameter.

3. The global navigation satellite system positioning method according to claim 1, characterized in that, Outlier data in the prediction residual matrix are removed using the quartile method to obtain the prediction residual matrix after outlier removal, including: The predicted residual matrix is ​​sorted to obtain the sorted predicted residual matrix; Using the data corresponding to the first upper quartile and the data corresponding to the first lower quartile as the boundaries of the sorted prediction residual matrix, the data of the prediction residual matrix outside the boundaries are excluded, and the prediction residual matrix after excluding the data outside the boundaries is obtained. The data corresponding to the second upper quartile and the data corresponding to the second lower quartile are used as the boundary of the prediction residual matrix after excluding the data outside the boundary, and the data of the prediction residual matrix outside the boundary are excluded. If the number of exclusions exceeds a preset number, or if the difference between the data corresponding to the upper quartile and the data corresponding to the lower quartile is less than a preset threshold, the exclusion ends, and the prediction residual matrix after excluding outlier data is obtained. The upper quartiles include the first upper quartile and the second upper quartile; the lower quartiles include the first lower quartile and the second lower quartile.

4. The global navigation satellite system positioning method according to claim 1, characterized in that, Based on the prediction residual matrix after excluding outlier data, the observation variance matrix is ​​robustly processed to obtain the robust observation variance matrix, including: The prediction residual matrix after excluding outlier data is standardized to obtain the standardized residual matrix; Based on the standardized residual matrix, the observation variance matrix is ​​subjected to robust processing to obtain the robust observation variance matrix.

5. The global navigation satellite system positioning method according to claim 4, characterized in that, Based on the standardized residual matrix, the observation variance matrix is ​​robustly processed to obtain the robust observation variance matrix, including: The magnification factor is determined based on the standardized residual matrix, the preset robustness scheme, and the preset empirical coefficients. The observation variance matrix is ​​expanded according to the expansion factor to obtain the robust observation variance matrix.

6. The global navigation satellite system positioning method according to claim 5, characterized in that, Based on the standardized residual matrix, the preset robustness scheme, and the preset empirical coefficients, the magnification factor is determined, including: Based on the preset robustness scheme and preset empirical coefficients, the data in the standardized residual matrix are divided into effective residual data, usable residual data, and harmful residual data. Based on the preset empirical coefficients and the standardized residual matrix, the first magnification factor corresponding to the effective residual data, the second magnification factor corresponding to the usable residual data, and the third magnification factor corresponding to the harmful residual data are determined respectively.

7. The global navigation satellite system positioning method according to claim 6, characterized in that, The data of the observed variance matrix are expanded according to the expansion factor, including: Based on the first magnification factor, the data of the observation variance matrix corresponding to the effective residual data is magnified; The data of the observation variance matrix corresponding to the available residual data is expanded according to the second expansion factor; The observation variance matrix corresponding to the harmful residual data is expanded according to the third expansion factor.

8. A global navigation satellite system positioning apparatus, characterized by, include: The first processing module is used to preprocess the acquired positioning observation data to obtain a parameter matrix; The second processing module is used to perform robust processing on the observation variance matrix in the parameter matrix to obtain a robust observation variance matrix. The third processing module is used to adaptively process the target prediction variance matrix based on the carrier residual submatrix and the Doppler residual submatrix in the prediction residual matrix to obtain the adaptively processed prediction variance matrix; wherein, the prediction residual matrix is ​​determined based on the parameter matrix; The fourth processing module is used to filter the robust observation variance matrix and the adaptive prediction variance matrix to obtain the GNSS position and velocity parameters of the Global Navigation Satellite System. The positioning observation data includes ephemeris data; the parameter matrix includes: a design matrix, the observation variance matrix, and the observation residual matrix. The second processing module includes: The third processing unit is used to exclude outlier data in the prediction residual matrix using the quartile method to obtain the prediction residual matrix after excluding outlier data. The fourth processing unit is used to perform robust processing on the observation variance matrix based on the prediction residual matrix after excluding outlier data, to obtain the robust observation variance matrix. The prediction residual matrix is ​​determined based on the design matrix, the observation residual matrix, and the parameter calculation results of the previous epoch. The parameter calculation result of the previous epoch is determined based on the ephemeris data.

9. A terminal, characterized in that, include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the global navigation satellite system positioning method as described in any one of claims 1 to 7.

10. A readable storage medium, characterized by, The readable storage medium stores a program that, when executed by a processor, implements the steps of the global navigation satellite system positioning method as described in any one of claims 1 to 7.

Images