Robust adaptive positioning method and device and electronic equipment

Abstract

The invention discloses a robust adaptive positioning method, a robust adaptive positioning device and electronic equipment. The method comprises the steps that a double-difference observation model is constructed according to a pseudo-range observation value and a carrier phase observation value in observation data received by a base station receiver and a monitoring station receiver, a pseudo-range double-difference observation value and a carrier double-difference observation value are obtained, and the observation data comprise the pseudo-range observation value, the carrier phase observation value and a signal-to-noise ratio observation value; and resolving the double-difference observation model by using a preset robust adaptive Kalman filtering algorithm to obtain a measured value residual vector, and taking the measured value residual vector as the position information of the monitoring station end receiver.

Description

Technical Field

[0001] The present invention relates to the field of satellite navigation and positioning technology, and in particular to an anti-error adaptive positioning method, device and electronic equipment. Background Art

[0002] In recent years, with the continuous development and popularization of the global satellite navigation system (GNSS), low-cost GNSS receivers (a device for receiving global navigation satellite system signals) have been widely used due to their low cost and low power consumption. However, low-cost GNSS receivers are susceptible to observation environment errors, especially in complex urban scenes. GNSS multipath errors (referring to the fact that during the propagation of global navigation satellite system signals, due to the reflection and scattering of various objects in the surrounding environment, the receiver receives multiple signal copies transmitted through different paths in addition to the signal directly from the satellite) are serious, gross errors and cycle slips occur frequently, resulting in a decrease in observation quality and affecting the performance of the Kalman filter. In severe cases, it may even cause the filter to diverge and fail to achieve positioning. The existing technology can solve the problems of gross errors and cycle slips well. For example, the MW-GF combination is used to detect cycle slips, which can detect various types of cycle slips more completely and eliminate them. The Raida criterion can eliminate gross errors. However, there is currently no clear method to eliminate multipath errors.

[0003] The existing Kalman filter algorithm is very sensitive to errors. Therefore, the impact of multipath errors on the accuracy of positioning results is more significant, and the existing Kalman filter algorithm lacks robustness to errors. This makes the algorithm model seriously affected by "outdated" information when the measurement value residual vector (the difference between the actual observation value and the predicted observation value) in the algorithm model is affected by multipath errors. The measurement update value of the state vector (the measurement update value is the new estimate obtained by correcting the prior estimate value in combination with the actual measurement value during the state estimation process) will be significantly worse in accuracy and reliability, which may cause the filter to diverge and fail to achieve positioning. In summary, since the existing technology cannot eliminate multipath errors, the existing technology has the problem of inaccurate positioning results or even the inability to achieve positioning. Summary of the invention

[0004] The embodiments of the present invention provide a robust adaptive positioning method, device and electronic device to solve the problem in the prior art that the positioning result is inaccurate or even the positioning cannot be achieved due to the inability to eliminate multipath errors.

[0005] The specific technical solutions are as follows:

[0006] In a first aspect, an embodiment of the present invention provides a robust adaptive positioning method, including:

[0007] According to the pseudorange observation values ​​and carrier phase observation values ​​in the observation data received by the base station and the monitoring station receivers, a double difference observation model is constructed, and pseudorange double difference observation values ​​and carrier double difference observation values ​​are obtained, wherein the observation data includes: pseudorange observation values, carrier phase observation values ​​and signal-to-noise ratio observation values;

[0008] The double difference observation model is solved by using a preset robust adaptive Kalman filter algorithm to obtain a measurement value residual vector, and the measurement value residual vector is used as the position information of the receiver at the monitoring station end; wherein the acquisition process of the robust adaptive Kalman filter algorithm is:

[0009] For each common-view satellite of the reference station and the monitoring station, respectively obtain the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station receivers, and perform inter-station difference on the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station to obtain the inter-station differential signal-to-noise ratio corresponding to each common-view satellite;

[0010] Obtaining historical signal-to-noise ratio observation values ​​of each common-view satellite received by the reference station and the monitoring station within a preset time period, and determining an inter-station differential signal-to-noise ratio threshold interval according to the historical signal-to-noise ratio observation values ​​of each common-view satellite;

[0011] The double-difference observation model is filtered and solved using a standard Kalman filter algorithm to obtain a measurement value residual vector, and a standardized residual is determined according to the measurement value residual vector and the median of its modulus length; the measurement value residual vector includes: a pseudorange measurement value residual vector and a carrier phase measurement value residual vector;

[0012] According to the weight and standardized residual of each observation vector, the preset division interval, the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval, an anti-error weight function is constructed; the observation vector is a pseudorange double difference observation vector and a carrier double difference observation vector;

[0013] Constructing an adaptive factor for a standard Kalman filter algorithm based on the discrepancy values ​​of each component of the prediction vector and the robustness weight function;

[0014] The constructed adaptive factor is introduced into the standard Kalman filter algorithm to obtain a robust adaptive Kalman filter algorithm.

[0015] Optionally, the step of constructing a double-difference observation model according to pseudorange observation values ​​and carrier phase observation values ​​in the observation data received by the reference station and the monitoring station receivers includes:

[0016] For each common-view satellite, the non-difference observation model corresponding to the reference station is constructed according to the pseudo-range observation values ​​and carrier phase observation values ​​received by the reference station receiver:

[0017]

[0018] The non-difference observation model corresponding to the monitoring station is constructed according to the pseudo-range observation values ​​and carrier phase observation values ​​received by the monitoring station receiver:

[0019]

[0020] Among them, the superscript s represents the common view satellite, the subscript r represents the receiver at the monitoring station, the subscript k represents the signal frequency, and c represents the speed of light in vacuum, c = 299792458 m / s. is the carrier wavelength of the common-view satellite s signal frequency k, It is the common-view satellite phase hardware delay. is the pseudo-range hardware delay between the satellite antenna and the satellite signal transmitter at the kth signal frequency, dt s represents the satellite clock error, represents the non-difference pseudorange observation value at the reference station, represents the non-difference carrier phase observation value at the reference station, Represents the geometric distance between the base station and the satellite, represents the tropospheric delay of the slant path at the reference station, represents the ionospheric delay of the slant path of the signal frequency k at the reference station. is the integer ambiguity at the base station; B b,k is the phase hardware delay of the base station receiver at the kth signal frequency point, b b,k is the pseudo-range hardware delay between the signal correlator at the kth signal frequency and the base station receiver antenna, and They represent the pseudorange measurement error and carrier phase measurement error at the reference station, dt b Indicates the base station receiver clock error, represents the non-difference pseudo-range observation value at the monitoring station, represents the non-difference carrier phase observation value at the monitoring station, represents the geometric distance between the monitoring station and the satellite, represents the slant path tropospheric delay at the monitoring station end; represents the ionospheric delay of the slant path of the signal frequency k at the monitoring station end, is the integer ambiguity at the monitoring station, B r,k is the phase hardware delay of the receiver at the kth signal frequency monitoring station, b r,k is the pseudo-range hardware delay between the signal correlator at the kth signal frequency and the receiver antenna at the monitoring station, and They represent the pseudorange measurement error and carrier phase measurement error at the monitoring station, respectively, and dt r Indicates the receiver clock error at the monitoring station;

[0021] Obtain the elevation angle of each common-view satellite of the reference station and the monitoring station, take the common-view satellite with the highest elevation angle as the reference satellite m, and take all the common-view satellites of the reference station and the monitoring station except the reference satellite m as non-reference satellites n;

[0022] For the reference satellite m, the non-difference pseudo-range observation values ​​corresponding to the monitoring station are used and carrier phase observations Subtract the non-difference pseudorange observation corresponding to the reference station and carrier phase observations The inter-station single-difference observation model corresponding to the reference satellite m is obtained:

[0023]

[0024] For each non-reference satellite n, the non-difference pseudo-range observation value corresponding to the monitoring station is used and carrier phase observations Subtract the non-difference pseudorange observation corresponding to the reference station and carrier phase observations The inter-station single-difference observation model corresponding to the non-reference satellite n is obtained:

[0025]

[0026] in, All represent single-difference pseudorange observations, Both represent single-difference carrier phase observations, denote the single-difference geometric distances between satellites m and n, denote the inter-station single-difference slant path tropospheric delays of satellites m and n, respectively, denote the inter-station single-difference slant path ionospheric delays of satellites m and n, respectively, denote the inter-station single-difference integer ambiguity of satellite m and n, respectively, and b b represents the pseudorange hardware delay of the base station receiver, b r represents the pseudorange hardware delay of the monitoring station receiver, denote the single-difference pseudorange observation noise between satellites m and n, denote the single-difference phase observation noise between satellites m and n, is the carrier wavelength of the common-view satellite m signal frequency point k, is the carrier wavelength of the common-view satellite n signal frequency k, B b represents the base station receiver phase hardware delay, B r Indicates the phase hardware delay of the monitoring station receiver;

[0027] For each non-reference satellite n, the non-difference pseudo-range observation value of the monitoring station corresponding to the reference satellite m is used and carrier phase observations Subtract the undifferenced pseudorange observation value of the reference station corresponding to the non-reference satellite n and carrier phase observations The double difference observation model is obtained:

[0028]

[0029] in, represents the double-difference pseudorange observation value, represents the double-difference carrier phase observation, represents the double-difference slant-path tropospheric delay; represents the slant path ionospheric delay; λ k is the carrier wavelength of the common-view satellite signal frequency k; is the double difference integer ambiguity; and They represent pseudorange measurement error and carrier phase measurement error respectively.

[0030] Optionally, the step of obtaining, for each common-view satellite of the reference station and the monitoring station, respectively the signal-to-noise ratio observation values ​​received by the receivers of the reference station and the monitoring station, and performing inter-station differences on the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station to obtain the inter-station differential signal-to-noise ratio corresponding to each common-view satellite includes:

[0031] Under the assumption that the receivers and antennas of the base station and the monitoring station are of the same type and the observation environment is an ideal unobstructed environment, for each common-view satellite of the base station and the monitoring station, obtain the signal-to-noise ratio observation value received by the base station receiver and the signal-to-noise ratio observation received by the monitoring station receiver

[0032] The signal-to-noise ratio observation received by the base station receiver Subtract the signal-to-noise ratio observation received by the monitoring station receiver Obtain the inter-station differential signal-to-noise ratio corresponding to the common-view satellite

[0033] Optionally, the step of obtaining a historical signal-to-noise ratio observation value of each common-view satellite received by the reference station and the monitoring station within a preset time period, and determining an inter-station differential signal-to-noise ratio threshold interval according to the historical signal-to-noise ratio observation value of each common-view satellite includes:

[0034] For each satellite in common view between the base station and the monitoring station, obtain the signal-to-noise ratio observation value received by the base station receiver within the preset historical time period. and the signal-to-noise ratio observation received by the monitoring station receiver

[0035] The signal-to-noise ratio observation value received by the base station receiver during the historical period is used Subtract the signal-to-noise ratio observation received by the monitoring station receiver Obtain the inter-station differential signal-to-noise ratio corresponding to the common-view satellite during the historical time period

[0036] For each common-view satellite, the standard deviation σ of the inter-station differential signal-to-noise ratio corresponding to the common-view satellite in the historical time period is calculated respectively, and the Laida criterion is used to calculate the inter-station differential signal-to-noise ratio corresponding to each common-view satellite in the interval range of [-3σ, 3σ] Determine the reliability interval

[0037] According to the reliable interval range corresponding to each common-view satellite;

[0038] Using the formula Determine the threshold interval of the differential signal-to-noise ratio between stations [(ΔSNR) min ,(ΔSNR) max ], where superscript s indicates common-view satellites; floor() indicates rounding down, and roundup() indicates rounding up;

[0039] The step of filtering and solving the double-difference observation model using a standard Kalman filter algorithm to obtain a measurement value residual vector, and determining a standardized residual according to the measurement value residual vector and the median of its modulus length includes:

[0040] The double-difference observation model is filtered and solved using the standard Kalman filter algorithm to obtain the measurement value residual vector v i ,in, Subscript k represents the signal frequency k, v i is the residual vector of the i-th measurement value, R i is the system noise matrix of the ith measurement value residual vector, represents the variance-covariance matrix of the measurement residual vector, represents the variance-covariance matrix of the residual vector of the measured value prior to the test, The transpose of the system noise matrix representing the residual vector of the measurement value at the signal frequency point k;

[0041] Using the formula Calculate the standardized residual |v i |, where sigma0 is the median modulus of the measurement value residual vector.

[0042] Optionally, the step of constructing an anti-error weight function according to the weight and standardized residual of each observation vector, a preset partition interval, the inter-station differential signal-to-noise ratio, and the inter-station differential signal-to-noise ratio threshold interval includes:

[0043] According to the weight of each observation vector, the standardized residual of each observation vector, the preset partition intervals [1.0, 1.5] and [2.0, 3.0], and the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval [(ΔSNR) min ,(ΔSNR) max ], construct the robust weight function:

[0044] in, represents the weight of the observation vector i after the robust estimation adjustment, p i represents the weight of observation vector i, |v i | represents the standardized residual of observation vector i, Td represents the differential signal-to-noise ratio threshold between stations, and the value range of Td is [(ΔSNR) min ,(ΔSNR) max ], k0 and k1 are both constants, k0∈[1.0,1.5], k1∈[2.0,3.0], |v i The area with |≤k0 is the copyright protection area, and k0<|v i The area with |≤k1 is the downgraded area and The area is a denial zone.

[0045] Optionally, the step of constructing an adaptive factor for a standard Kalman filter algorithm according to the mismatch values ​​of each component of the prediction vector and the robustness weight function comprises:

[0046] According to the weights adjusted after the robust estimation of each observation vector, the state vector robust estimation of each observation vector at the k signal frequency point is calculated.

[0047] The state prediction vector of the observation vector at the kth signal frequency point is calculated based on the state vector robust estimation of the k-1th signal frequency point of each observation vector and the double difference observation model.

[0048] Using the formula Calculate the discrepancy value of each state prediction vector, where represents the discrepancy value of the j-th state prediction vector;

[0049] According to the discrepancy values ​​of each component of the prediction vector and the anti-error weight function, the formula is used Constructing an adaptive factor for the standard Kalman filter algorithm, sigma1 represents the median length of the prediction vector mismatch value; the subscript k represents the k signal frequency point; represents the adaptive factor of the j-th state prediction vector; c0 and c1 are the thresholds of the adaptive factor, c0 = 2.5, c1 = 7;

[0050] The constructed adaptive factor is introduced into the standard Kalman filter algorithm to obtain the formula of the robust adaptive Kalman filter algorithm: Among them, the subscript k represents the k signal frequency point, and the superscript T represents the transpose of the matrix. is the observation vector equivalent weight matrix, is the prediction vector weight matrix, A k is the coefficient matrix in the double-difference observation model, is the prediction vector adaptive factor, is the estimated value of the state vector, L k is the observation vector.

[0051] Optionally, after the step of solving the double-difference observation model using a preset robust adaptive Kalman filter algorithm to obtain a measurement value residual vector, the method further includes:

[0052] For each measurement value residual vector, respectively, according to the measurement value residual vector v k and the measurement value residual vector v k The corresponding variance-covariance matrix Construction Statistics λ k In order to obey the chi-square distribution with m degrees of freedom, determine whether the measured value residual vector belongs to the interval (Q1-C1·IQR, Q3+C1·IQR). If it does, keep the measured value residual vector; if it does not, remove the measured value residual vector. Q1 and Q3 are the residual values ​​of the measured value residual vector at the 25% and 75% positions after the residual values ​​of each measured value residual vector are arranged from small to large. IQR=Q3-Q1, C1=1.5, and the superscript T represents the transpose of the matrix.

[0053] The retained measurement residual vector is used as the position information of the receiver at the monitoring station.

[0054] In a second aspect, an embodiment of the present invention provides a robust adaptive positioning device, including:

[0055] A construction module is used to construct a double-difference observation model according to pseudo-range observation values ​​and carrier phase observation values ​​in the observation data received by the base station and the monitoring station receiver, and obtain pseudo-range double-difference observation values ​​and carrier double-difference observation values, wherein the observation data includes: pseudo-range observation values, carrier phase observation values ​​and signal-to-noise ratio observation values;

[0056] A solution module, used to solve the double-difference observation model by using a preset robust adaptive Kalman filter algorithm, obtain a measurement value residual vector, and use the measurement value residual vector as the position information of the receiver at the monitoring station end;

[0057] The device also includes: an algorithm acquisition module, which is used to:

[0058] For each common-view satellite of the reference station and the monitoring station, respectively obtain the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station receivers, and perform inter-station difference on the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station to obtain the inter-station differential signal-to-noise ratio corresponding to each common-view satellite;

[0059] Obtaining historical signal-to-noise ratio observation values ​​of each common-view satellite received by the reference station and the monitoring station within a preset time period, and determining an inter-station differential signal-to-noise ratio threshold interval according to the historical signal-to-noise ratio observation values ​​of each common-view satellite;

[0060] The double-difference observation model is filtered and solved using a standard Kalman filter algorithm to obtain a measurement value residual vector, and a standardized residual is determined according to the measurement value residual vector and the median of its modulus length; the measurement value residual vector includes: a pseudorange measurement value residual vector and a carrier phase measurement value residual vector;

[0061] According to the weight and standardized residual of each observation vector, the preset division interval, the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval, an anti-error weight function is constructed; the observation vector is a pseudorange double difference observation vector and a carrier double difference observation vector;

[0062] Constructing an adaptive factor for a standard Kalman filter algorithm based on the discrepancy values ​​of each component of the prediction vector and the robustness weight function;

[0063] The constructed adaptive factor is introduced into the standard Kalman filter algorithm to obtain a robust adaptive Kalman filter algorithm.

[0064] Optionally, the construction module 301 is specifically used for:

[0065] For each common-view satellite, the non-difference observation model corresponding to the reference station is constructed according to the pseudo-range observation values ​​and carrier phase observation values ​​received by the reference station receiver:

[0066]

[0067] The non-difference observation model corresponding to the monitoring station is constructed according to the pseudo-range observation values ​​and carrier phase observation values ​​received by the monitoring station receiver:

[0068]

[0069] Among them, the superscript s represents the common view satellite, the subscript r represents the receiver at the monitoring station, the subscript k represents the signal frequency, and c represents the speed of light in vacuum, c = 299792458 m / s. is the carrier wavelength of the common-view satellite s signal frequency k, It is the common-view satellite phase hardware delay. is the pseudo-range hardware delay between the satellite antenna and the satellite signal transmitter at the kth signal frequency, dt s represents the satellite clock error, represents the non-difference pseudorange observation value at the reference station, represents the non-difference carrier phase observation value at the reference station, Represents the geometric distance between the base station and the satellite, represents the tropospheric delay of the slant path at the reference station, represents the ionospheric delay of the slant path of the signal frequency k at the reference station. is the integer ambiguity at the base station; B b,k is the phase hardware delay of the base station receiver at the kth signal frequency point, b b,k is the pseudo-range hardware delay between the signal correlator at the kth signal frequency and the base station receiver antenna, and They represent the pseudorange measurement error and carrier phase measurement error at the reference station, dt b Indicates the base station receiver clock error, represents the non-difference pseudo-range observation value at the monitoring station, represents the non-difference carrier phase observation value at the monitoring station, represents the geometric distance between the monitoring station and the satellite, represents the slant path tropospheric delay at the monitoring station end; represents the ionospheric delay of the slant path of the signal frequency k at the monitoring station end, is the integer ambiguity at the monitoring station, B r,k is the phase hardware delay of the receiver at the kth signal frequency monitoring station, b r,k is the pseudo-range hardware delay between the signal correlator at the kth signal frequency and the receiver antenna at the monitoring station, and They represent the pseudorange measurement error and carrier phase measurement error at the monitoring station, respectively, and dt r Indicates the receiver clock error at the monitoring station;

[0070] Obtain the elevation angle of each common-view satellite of the reference station and the monitoring station, take the common-view satellite with the highest elevation angle as the reference satellite m, and take all the common-view satellites of the reference station and the monitoring station except the reference satellite m as non-reference satellites n;

[0071] For the reference satellite m, the non-difference pseudo-range observation values ​​corresponding to the monitoring station are used and carrier phase observations Subtract the non-difference pseudorange observation corresponding to the reference station and carrier phase observations The inter-station single-difference observation model corresponding to the reference satellite m is obtained:

[0072]

[0073] For each non-reference satellite n, the non-difference pseudo-range observation value corresponding to the monitoring station is used and carrier phase observations Subtract the non-difference pseudorange observation corresponding to the reference station and carrier phase observations The inter-station single-difference observation model corresponding to the non-reference satellite n is obtained:

[0074]

[0075] in, All represent single-difference pseudorange observations, Both represent single-difference carrier phase observations, denote the single-difference geometric distances between satellites m and n, denote the inter-station single-difference slant path tropospheric delays of satellites m and n, respectively, denote the inter-station single-difference slant path ionospheric delays of satellites m and n, respectively, denote the inter-station single-difference integer ambiguity of satellite m and n, respectively, and b b represents the pseudorange hardware delay of the base station receiver, b r represents the pseudorange hardware delay of the monitoring station receiver, denote the single-difference pseudorange observation noise between satellites m and n, denote the single-difference phase observation noise between satellites m and n, is the carrier wavelength of the common-view satellite m signal frequency point k, is the carrier wavelength of the common-view satellite n signal frequency k, B b represents the base station receiver phase hardware delay, B r Indicates the phase hardware delay of the monitoring station receiver;

[0076] For each non-reference satellite n, the non-difference pseudo-range observation value of the monitoring station corresponding to the reference satellite m is used and carrier phase observations Subtract the undifferenced pseudorange observation value of the reference station corresponding to the non-reference satellite n and carrier phase observations The double difference observation model is obtained:

[0077]

[0078] in, represents the double-difference pseudorange observation value, represents the double-difference carrier phase observation, represents the double-difference slant-path tropospheric delay; represents the slant path ionospheric delay; λ k is the carrier wavelength of the common-view satellite signal frequency k; is the double difference integer ambiguity; and They represent pseudorange measurement error and carrier phase measurement error respectively.

[0079] Optional, algorithm acquisition module, specifically used for:

[0080] Under the assumption that the receivers and antennas of the base station and the monitoring station are of the same type and the observation environment is an ideal unobstructed environment, for each common-view satellite of the base station and the monitoring station, obtain the signal-to-noise ratio observation value received by the base station receiver and the signal-to-noise ratio observation received by the monitoring station receiver

[0081] The signal-to-noise ratio observation received by the base station receiver Subtract the signal-to-noise ratio observation received by the monitoring station receiver Obtain the inter-station differential signal-to-noise ratio corresponding to the common-view satellite

[0082] Optional, algorithm acquisition module, specifically used for:

[0083] For each satellite in common view between the base station and the monitoring station, obtain the signal-to-noise ratio observation value received by the base station receiver within the preset historical time period. and the signal-to-noise ratio observation received by the monitoring station receiver

[0084] The signal-to-noise ratio observation value received by the base station receiver during the historical period is used Subtract the signal-to-noise ratio observation received by the monitoring station receiver Obtain the inter-station differential signal-to-noise ratio corresponding to the common-view satellite during the historical time period

[0085] For each common-view satellite, the standard deviation σ of the inter-station differential signal-to-noise ratio corresponding to the common-view satellite in the historical time period is calculated respectively, and the Laida criterion is used to calculate the inter-station differential signal-to-noise ratio corresponding to each common-view satellite in the interval range of [-3σ, 3σ] Determine the reliability interval

[0086] According to the reliable interval range corresponding to each common-view satellite;

[0087] Using the formula Determine the threshold interval of the differential signal-to-noise ratio between stations [(ΔSNR) min ,(ΔSNR) max ], where superscript s indicates common-view satellites; floor() indicates rounding down, and roundup() indicates rounding up;

[0088] Optional, algorithm acquisition module, specifically used for:

[0089] The double-difference observation model is filtered and solved using the standard Kalman filter algorithm to obtain the measurement value residual vector v i ,in, Subscript k represents the signal frequency k, v i is the residual vector of the i-th measurement value, R i is the system noise matrix of the ith measurement value residual vector, represents the variance-covariance matrix of the measurement residual vector, represents the variance-covariance matrix of the residual vector of the measured value prior to the test, The transpose of the system noise matrix representing the residual vector of the measurement value at the signal frequency point k;

[0090] Using the formula Calculate the standardized residual |v i |, where sigma0 is the median modulus of the measurement value residual vector.

[0091] Optional, algorithm acquisition module, specifically used for:

[0092] According to the weight of each observation vector, the standardized residual of each observation vector, the preset partition intervals [1.0, 1.5] and [2.0, 3.0], and the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval [(ΔSNR) min ,(ΔSNR) max ], construct the robust weight function:

[0093] in, represents the weight of the observation vector i after the robust estimation adjustment, p i represents the weight of observation vector i, |v i | represents the standardized residual of observation vector i, Td represents the differential signal-to-noise ratio threshold between stations, and the value range of Td is [(ΔSNR) min,(ΔSNR) max ], k0 and k1 are both constants, k0∈[1.0,1.5], k1∈[2.0,3.0], |v i The area with |≤k0 is the copyright protection area, and k0<|v i The area with |≤k1 is the downgraded area and The area is a denial zone.

[0094] Optional, algorithm acquisition module, specifically used for:

[0095] According to the weights adjusted after the robust estimation of each observation vector, the state vector robust estimation of each observation vector at the k signal frequency point is calculated.

[0096] The state prediction vector of the observation vector at the kth signal frequency point is calculated based on the state vector robust estimation of the k-1th signal frequency point of each observation vector and the double difference observation model.

[0097] Using the formula Calculate the discrepancy value of each state prediction vector, where represents the discrepancy value of the j-th state prediction vector;

[0098] According to the discrepancy values ​​of each component of the prediction vector and the anti-error weight function, the formula is used Constructing an adaptive factor for the standard Kalman filter algorithm, sigma1 represents the median length of the prediction vector mismatch value; the subscript k represents the k signal frequency point; represents the adaptive factor of the j-th state prediction vector; c0 and c1 are the thresholds of the adaptive factor, c0 = 2.5, c1 = 7;

[0099] The constructed adaptive factor is introduced into the standard Kalman filter algorithm to obtain the formula of the robust adaptive Kalman filter algorithm: Among them, the subscript k represents the k signal frequency point, and the superscript T represents the transpose of the matrix. is the observation vector equivalent weight matrix, is the prediction vector weight matrix, A k is the coefficient matrix in the double-difference observation model, is the prediction vector adaptive factor, is the estimated value of the state vector, L k is the observation vector.

[0100] Optionally, the device further includes: a verification module, configured to:

[0101] For each measurement value residual vector, respectively, according to the measurement value residual vector v kand the measurement value residual vector v k The corresponding variance-covariance matrix Construction Statistics λ k In order to obey the chi-square distribution with m degrees of freedom, determine whether the measured value residual vector belongs to the interval (Q1-C1·IQR, Q3+C1·IQR). If it does, keep the measured value residual vector; if it does not, remove the measured value residual vector. Q1 and Q3 are the residual values ​​of the measured value residual vector at the 25% and 75% positions after the residual values ​​of each measured value residual vector are arranged from small to large. IQR=Q3-Q1, C1=1.5, and the superscript T represents the transpose of the matrix.

[0102] The retained measurement residual vector is used as the position information of the receiver at the monitoring station.

[0103] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0104] The memory is used to store computer programs;

[0105] The processor is used to implement the method steps of the robust adaptive positioning described in the first aspect when executing the program stored in the memory.

[0106] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps of the robust adaptive positioning described in the first aspect are implemented.

[0107] In summary, the method provided by the embodiment of the present invention utilizes the correlation between multipath and signal-to-noise ratio to construct an adaptive factor for the standard Kalman filter algorithm, and obtains a robust adaptive Kalman filter algorithm for identifying multipath errors, and constructs a multipath error inter-station differential signal-to-noise ratio rejection threshold Td based on historical observation data, and includes the measurement values ​​exceeding the threshold into the IGGIII rejection zone, so that when there are abnormal values ​​such as multipath errors in the observation data, the robust adaptive Kalman filter algorithm can automatically adjust the weights, thereby effectively eliminating the influence of abnormal values ​​such as multipath errors in the observation data on the estimation results, thereby improving the accuracy of the estimation results.

[0108] In addition, the method provided by the embodiment of the present invention also takes into account the situation that the low-cost terminal positioning in a complex environment may be disturbed by various error factors. In the prior art, when the low-cost terminal positioning in a complex environment is disturbed by various error factors, the existing Kalman filter algorithm is usually based on a traditional random model, and continuously predicts and updates the estimated value of the positioning system state to achieve the optimal estimation of the positioning system and obtain accurate positioning results. However, the traditional random model cannot accurately reflect the accuracy of the measurement value, so the measurement value variance is used as a standardized residual construction factor, and an accurate standardized residual cannot be constructed, resulting in filter divergence. The method provided by the embodiment of the present invention uses the median of the modulus length of the residual vector instead of the measurement value variance as an element for constructing the standardized residual. Similarly, the median of the modulus length of the prediction vector mismatch value is used instead of the measurement value variance as an element for constructing the adaptive factor, which can construct accurate standardized residuals and adaptive factors, further improving the accuracy of the estimation results of the robust adaptive Kalman filter algorithm.

[0109] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0110] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0111] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0112] Figure 1 A schematic diagram of a flow chart of a robust adaptive positioning method provided by an embodiment of the present invention;

[0113] Figure 2 A schematic diagram of a process of obtaining a robust adaptive Kalman filter algorithm in an embodiment of the present invention;

[0114] Figure 3 A schematic diagram of the structure of a robust and adaptive positioning device provided by an embodiment of the present invention;

[0115] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0116] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0117] First, the professional terms involved in the embodiments of the present invention are explained:

[0118] 1. A co-viewed satellite refers to the same satellite observed by time-synchronized devices at two or more locations at the same time.

[0119] 2. A GNSS receiver is a device used to receive Global Navigation Satellite System (GNSS) signals.

[0120] 3. Observables. In Kalman filtering, observed quantities refer to information about the state of a system that can be obtained directly from sensors or measuring devices. For example, in a satellite positioning system, observed quantities can be pseudoranges, carrier phases, and other measurements of satellite signals received by a satellite receiver. Observable quantities usually contain certain noise and errors. The purpose of Kalman filtering is to estimate the true state of the system by combining the system's prior knowledge (state model) and observed quantities, and to continuously update the estimated value to improve accuracy.

[0121] 4. Signal frequency refers to the specific frequency value used to distinguish different signals in satellite communication systems.

[0122] 5. The discrepancy between the components of the prediction vector usually refers to the difference between the components of the prediction vector and the corresponding components of the actual observed value or true value. For example, if the prediction vector represents a prediction of a system state, such as the three-dimensional coordinates of the position (x, y, z) and the three-dimensional components of the velocity (vx, vy, vz), then the discrepancy is the difference between the actual observed position and velocity components and the corresponding components in the prediction vector. These discrepancies can be used to evaluate the accuracy of the prediction, and in the update step of the Kalman filter, the prediction value will be adjusted according to the size of the discrepancy to obtain an estimate that is closer to the true state.

[0123] 6. Non-difference observation model: the observation signals received by a single station from different satellites are processed independently without inter-station or inter-satellite difference. The observation value of each station includes the true distance from the satellite to the station and various error terms.

[0124] 7. Inter-station single-difference observation model: For the same satellite, observations are made at two stations at the same time, and the observation values ​​of the two stations are subtracted to obtain the inter-station single-difference observation value. Its purpose is to eliminate or weaken some common error sources, such as satellite clock errors.

[0125] 8. Standardized residual is the quantity obtained after standardizing the residual. It can be used to evaluate the fitting effect of the model and detect outliers.

[0126] 9. Pseudorange is the distance measurement value obtained by multiplying the time difference between the satellite signal reception time and the satellite signal transmission time by the speed of light. However, due to the influence of factors such as satellite clock error, receiver clock error, and atmospheric delay, this distance is not the true geometric distance, so it is called pseudorange.

[0127] 10. Carrier phase refers to the phase of the carrier signal transmitted by the satellite at the receiver.

[0128] In order to solve the problem in the prior art that the positioning result is inaccurate or even the positioning cannot be achieved due to the inability to eliminate multipath errors. The embodiment of the present invention provides a robust adaptive positioning method, such as Figure 1 As shown, the method includes:

[0129] S101: constructing a double-difference observation model according to pseudo-range observation values ​​and carrier phase observation values ​​in the observation data received by the base station and the monitoring station receivers, and obtaining pseudo-range double-difference observation values ​​and carrier double-difference observation values, wherein the observation data includes: pseudo-range observation values, carrier phase observation values ​​and signal-to-noise ratio observation values;

[0130] In practical applications, the base station is a GNSS-RTK base station, which is a GNSS receiving device fixed at a known precise coordinate position. The main function of the GNSS-RTK base station is to continuously receive satellite signals and transmit the observation data to the monitoring station (also called a rover or mobile station) in real time. By comparing and differentially processing the data with the monitoring station, it provides the monitoring station with high-precision positioning correction information, thereby improving the positioning accuracy of the monitoring station.

[0131] Specifically, the step of constructing a double difference observation model according to pseudorange observation values ​​and carrier phase observation values ​​in the observation data received by the base station and the monitoring station receivers includes:

[0132] For each common-view satellite, the non-difference observation model corresponding to the reference station is constructed according to the pseudo-range observation values ​​and carrier phase observation values ​​received by the reference station receiver:

[0133]

[0134] The non-difference observation model corresponding to the monitoring station is constructed according to the pseudo-range observation values ​​and carrier phase observation values ​​received by the monitoring station receiver:

[0135]

[0136] Among them, the superscript s represents the common view satellite, the subscript r represents the receiver at the monitoring station, the subscript k represents the signal frequency, and c represents the speed of light in vacuum, c = 299792458 m / s. is the carrier wavelength of the common-view satellite s signal frequency k, It is the common-view satellite phase hardware delay. is the pseudo-range hardware delay between the satellite antenna and the satellite signal transmitter at the kth signal frequency, dt s represents the satellite clock error, represents the non-difference pseudorange observation value at the reference station, represents the non-difference carrier phase observation value at the reference station, Represents the geometric distance between the base station and the satellite, represents the tropospheric delay of the slant path at the reference station, represents the ionospheric delay of the slant path of the signal frequency k at the reference station. is the integer ambiguity at the base station; B b,k is the phase hardware delay of the base station receiver at the kth signal frequency point, b b,k is the pseudo-range hardware delay between the signal correlator at the kth signal frequency and the base station receiver antenna, and They represent the pseudorange measurement error and carrier phase measurement error at the reference station, dt b Indicates the base station receiver clock error, represents the non-difference pseudo-range observation value at the monitoring station, represents the non-difference carrier phase observation value at the monitoring station, represents the geometric distance between the monitoring station and the satellite, represents the slant path tropospheric delay at the monitoring station end; represents the ionospheric delay of the slant path of the signal frequency k at the monitoring station end, is the integer ambiguity at the monitoring station, B r,k is the phase hardware delay of the receiver at the kth signal frequency monitoring station, b r,k is the pseudo-range hardware delay between the signal correlator at the kth signal frequency and the receiver antenna at the monitoring station, and They represent the pseudorange measurement error and carrier phase measurement error at the monitoring station, respectively, and dt r Indicates the receiver clock error at the monitoring station;

[0137] Obtain the elevation angle of each common-view satellite of the reference station and the monitoring station, take the common-view satellite with the highest elevation angle as the reference satellite m, and take all the common-view satellites of the reference station and the monitoring station except the reference satellite m as non-reference satellites n;

[0138] For the reference satellite m, the non-difference pseudo-range observation values ​​corresponding to the monitoring station are used and carrier phase observations Subtract the non-difference pseudorange observation corresponding to the reference station and carrier phase observations The inter-station single-difference observation model corresponding to the reference satellite m is obtained:

[0139]

[0140] For each non-reference satellite n, the non-difference pseudo-range observation value corresponding to the monitoring station is used and carrier phase observations Subtract the non-difference pseudorange observation corresponding to the reference station and carrier phase observations The inter-station single-difference observation model corresponding to the non-reference satellite n is obtained:

[0141]

[0142] in, All represent single-difference pseudorange observations, Both represent single-difference carrier phase observations, denote the single-difference geometric distances between satellites m and n, denote the inter-station single-difference slant path tropospheric delays of satellites m and n, respectively, denote the inter-station single-difference slant path ionospheric delays of satellites m and n, respectively, denote the inter-station single-difference integer ambiguity of satellite m and n, respectively, and b b represents the pseudorange hardware delay of the base station receiver, b r represents the pseudorange hardware delay of the monitoring station receiver, denote the single-difference pseudorange observation noise between satellites m and n, denote the single-difference phase observation noise between satellites m and n, is the carrier wavelength of the common-view satellite m signal frequency point k, is the carrier wavelength of the common-view satellite n signal frequency k, B b represents the base station receiver phase hardware delay, B r Indicates the phase hardware delay of the monitoring station receiver;

[0143] For each non-reference satellite n, the non-difference pseudo-range observation value of the monitoring station corresponding to the reference satellite m is used and carrier phase observations Subtract the undifferenced pseudorange observation value of the reference station corresponding to the non-reference satellite n and carrier phase observations The double difference observation model is obtained:

[0144]

[0145] in, represents the double-difference pseudorange observation value, represents the double-difference carrier phase observation, represents the double-difference slant-path tropospheric delay; represents the slant path ionospheric delay; λ k is the carrier wavelength of the common-view satellite signal frequency k; is the double difference integer ambiguity; and They represent pseudorange measurement error and carrier phase measurement error respectively.

[0146] S102: using a preset robust adaptive Kalman filter algorithm to solve the double-difference observation model, obtaining a measurement value residual vector, and using the measurement value residual vector as the position information of the receiver at the monitoring station.

[0147] In other embodiments, after solving the double-difference observation model using the robust adaptive Kalman filter algorithm, the solution result can also be checked, for example, the solution result can be checked by using the overall residual chi-square threshold or the post-test residual threshold one by one, or the solution result can be checked by using the overall residual chi-square threshold and the post-test residual threshold double check method to determine the validity of the final result. Specifically, after solving the double-difference observation model using the preset robust adaptive Kalman filter algorithm and obtaining the measurement value residual vector, the step of checking the solution result by using the overall residual chi-square threshold and the post-test residual threshold double check method can include:

[0148] For each measurement value residual vector, respectively, according to the measurement value residual vector v k and the measurement value residual vector v k The corresponding variance-covariance matrix Construction Statistics λ k In order to obey the chi-square distribution with m degrees of freedom, determine whether the measured value residual vector belongs to the interval (Q1-C1·IQR, Q3+C1·IQR). If it does, keep the measured value residual vector; if it does not, remove the measured value residual vector. Q1 and Q3 are the residual values ​​of the measured value residual vector at the 25% and 75% positions after the residual values ​​of each measured value residual vector are arranged from small to large. IQR=Q3-Q1, C1=1.5, and the superscript T represents the transpose of the matrix.

[0149] The retained measurement residual vector is used as the location information of the receiver at the monitoring station.

[0150] In actual applications, if the measurement is abnormal, compared with the standard value λ kThe value will become larger. At this time, the measurement update should be abandoned to isolate the abnormal measurement. Under normal measurement conditions, it is determined whether the measurement value residual vector belongs to the interval (Q1-C1·IQR, Q3+C1·IQR).

[0151] The variance-covariance matrix reflects the degree of uncertainty in the state estimate. The diagonal elements of the covariance matrix represent the variance of each state variable, and the off-diagonal elements represent the correlation between the state variables. A smaller variance means a smaller uncertainty in the state estimate and a more reliable estimate.

[0152] In a specific embodiment, the Figure 2 The steps shown are used to obtain the robust adaptive Kalman filter algorithm. The specific process steps are as follows:

[0153] S201: for each common-view satellite of the reference station and the monitoring station, respectively obtain the signal-to-noise ratio observation values ​​received by the base station and the monitoring station receivers, and perform inter-station difference on the signal-to-noise ratio observation values ​​received by the base station and the monitoring station to obtain the inter-station differential signal-to-noise ratio corresponding to each common-view satellite;

[0154] For each common-view satellite of the reference station and the monitoring station, respectively obtain the signal-to-noise ratio observation values ​​received by the base station and the monitoring station receivers, and perform inter-station difference on the signal-to-noise ratio observation values ​​received by the base station and the monitoring station to obtain the inter-station differential signal-to-noise ratio corresponding to each common-view satellite, including:

[0155] Under the assumption that the receivers and antennas of the base station and the monitoring station are of the same type and the observation environment is an ideal unobstructed environment, for each common-view satellite of the base station and the monitoring station, obtain the signal-to-noise ratio observation value received by the base station receiver and the signal-to-noise ratio observation received by the monitoring station receiver

[0156] The signal-to-noise ratio observation received by the base station receiver Subtract the signal-to-noise ratio observation received by the monitoring station receiver Get the inter-station differential signal-to-noise ratio corresponding to the common-view satellite

[0157] In GNSS-RTK (Global Navigation Satellite System-Real Time Kinematic) technology, it is assumed that the GNSS receivers and antenna types of the base station b and the monitoring station r are the same, the observation environment is an ideal unobstructed environment, and the base station and the monitoring station have a common view satellite s. Then, the error terms of the signals received by the base station and the monitoring station from the common view satellite s are basically the same. The signal-to-noise ratio observations received by the two stations are differentiated between the stations, which can be expressed as:

[0158]

[0159] In theory, It should obey Gaussian normal distribution. In landslide monitoring, the base station is usually deployed in an open environment without obstruction. The monitoring station environment is relatively complex. Due to the constraints of the selected point location, there are usually vegetation and terrain obstructions around it, and the multipath effect is serious. When the satellite signal of the monitoring station is blocked by vegetation and terrain, serious signal diffraction and reflection phenomena occur. The sequence will show large fluctuations.

[0160] S202: Obtain historical signal-to-noise ratio observation values ​​of each common-view satellite received by the reference station and the monitoring station within a preset time period, and determine an inter-station differential signal-to-noise ratio threshold interval according to the historical signal-to-noise ratio observation values ​​of each common-view satellite;

[0161] Specifically, the steps of obtaining the historical signal-to-noise ratio observation value of each common-view satellite received by the reference station and the monitoring station within a preset time period, and determining the inter-station differential signal-to-noise ratio threshold range according to the historical signal-to-noise ratio observation value of each common-view satellite include:

[0162] For each satellite in common view between the base station and the monitoring station, obtain the signal-to-noise ratio observation value received by the base station receiver within the preset historical time period. and the signal-to-noise ratio observation received by the monitoring station receiver

[0163] Use the signal-to-noise ratio observations received by the base station receiver during the historical period Subtract the signal-to-noise ratio observation received by the monitoring station receiver Get the inter-station differential signal-to-noise ratio corresponding to the common-view satellite in the historical time period

[0164] For each common-view satellite, the standard deviation σ of the inter-station differential signal-to-noise ratio corresponding to the common-view satellite in the historical time period is calculated respectively, and the Laida criterion is used to calculate the inter-station differential signal-to-noise ratio corresponding to each common-view satellite in the interval range of [-3σ, 3σ] Determine the reliability interval

[0165] According to the reliable interval range corresponding to each common-view satellite;

[0166] Using the formula Determine the threshold interval of the differential signal-to-noise ratio between stations [(ΔSNR) min ,(ΔSNR) max ], where the superscript s indicates the common-view satellite; floor() indicates rounding down, and roundup() indicates rounding up.

[0167] During landslide monitoring, the environment around the receiver remains basically unchanged in the short term, and the satellite motion trajectory is periodically repetitive, and its multipath effect is also periodically repetitive. Therefore, the recent historical signal-to-noise ratio observations can be used to make inter-station differences between the common view satellite s of the base station and the monitoring station to obtain When the environment is ideal, The sequence approximately obeys the Gaussian normal distribution, and its standard deviation σ is calculated. The Raida criterion is used to select the common-view satellite s in the interval [-3σ, 3σ]. Sequence reliability interval range To determine the interval range of the overall inter-station differential signal-to-noise ratio sequence of all common-view satellites [(ΔSNR) min ,(ΔSNR) max ], the reliable interval range of all common-view satellites is screened and counted, and the smallest one among all common-view satellites is Round down to get (ΔSNR) min : The smallest satellite among all satellites in common view Round up to get (ΔSNR) max .

[0168] In the embodiment of the present invention, the criterion for identifying and eliminating multipath errors based on the reference station signal-to-noise ratio observation data can be expressed as:

[0169]

[0170] S203: Use a standard Kalman filter algorithm to filter and solve the double difference observation model to obtain a measurement value residual vector, and determine a standardized residual according to the measurement value residual vector and the median of its modulus length; the measurement value residual vector includes: a pseudorange measurement value residual vector and a carrier phase measurement value residual vector;

[0171] Specifically, the double difference observation model is filtered and solved using a standard Kalman filter algorithm to obtain a measurement value residual vector, and the steps of determining a standardized residual according to the measurement value residual vector and the median of its modulus length include:

[0172] Use the standard Kalman filter algorithm to filter and solve the double difference observation model to obtain the measurement value residual vector v i ,in, Subscript k represents the signal frequency k, v i is the residual vector of the i-th measurement value, R i is the system noise matrix of the ith measurement value residual vector, represents the variance-covariance matrix of the measurement residual vector, represents the variance-covariance matrix of the residual vector of the measured value prior to the test, The transpose of the system noise matrix representing the residual vector of the measurement value at the signal frequency point k;

[0173] Using the formula Calculate the standardized residual |v i |, where sigma0 is the median modulus of the measurement value residual vector.

[0174] S204: constructing an anti-error weight function according to the weight and standardized residual of each observation vector, the preset division interval, the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval; the observation vector is a pseudorange double difference observation vector and a carrier double difference observation vector;

[0175] In practical applications, the robust weight function may be a three-segment robust weight function of IGGIII.

[0176] Specifically, the step of constructing an anti-error weight function according to the weight and standardized residual of each observation vector, the preset division interval, the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval includes:

[0177] According to the weight of each observation vector, the standardized residual of each observation vector, the preset partition intervals [1.0, 1.5] and [2.0, 3.0], and the differential signal-to-noise ratio between stations and the threshold interval of the differential signal-to-noise ratio between stations [(ΔSNR) min ,(ΔSNR) max ], construct the robust weight function:

[0178] in, represents the weight of the observation vector i after the robust estimation adjustment, p i represents the weight of observation vector i, |v i | represents the standardized residual of observation vector i, Td represents the differential signal-to-noise ratio threshold between stations, and the value range of Td is [(ΔSNR) min ,(ΔSNR) max ], k0 and k1 are both constants, k0∈[1.0,1.5], k1∈[2.0,3.0], |v i The area with |≤k0 is the copyright protection area, and k0<|v i The area with |≤k1 is the downgraded area and The area is a denial zone.

[0179] The standardized residual is greater than the threshold k1 or the differential signal-to-noise ratio between stations. The measurement values ​​greater than Td are included in the rejection zone and assigned infinite variance, so that the measurement values ​​are directly eliminated in the embodiment of the present invention; the measurement values ​​whose standardized residuals fall into the interval [k0, k1] are weighted according to the weight reduction model of the IGGIII three-segment robustness weight function; the measurement value weights remain unchanged in other cases.

[0180] Robust estimation refers to reducing or eliminating the impact of observational errors on parameter estimation by selecting an appropriate parameter estimation method.

[0181] S205: constructing an adaptive factor for the standard Kalman filter algorithm according to the discrepancy values ​​of each component of the prediction vector and the robustness weight function;

[0182] Specifically, the steps of constructing an adaptive factor for the standard Kalman filter algorithm according to the mismatch values ​​of each component of the prediction vector and the robustness weight function include:

[0183] According to the weights adjusted after the robust estimation of each observation vector, the state vector robust estimation of each observation vector at the k signal frequency point is calculated.

[0184] According to the state vector robust estimation and double difference observation model of each observation vector at the k-1th signal frequency point, the state prediction vector of the observation vector at the kth signal frequency point is calculated and obtained.

[0185] Using the formula Calculate the discrepancy value of each state prediction vector, where represents the discrepancy value of the j-th state prediction vector;

[0186] According to the discrepancy values ​​of each component of the prediction vector and the anti-error weight function, the formula is used Constructing an adaptive factor for the standard Kalman filter algorithm, sigma1 represents the median length of the prediction vector mismatch value; the subscript k represents the k signal frequency point; represents the adaptive factor of the j-th state prediction vector; c0 and c1 are the thresholds of the adaptive factor, c0=2.5, c1=7.

[0187] During the Kalman filtering process, if there is a large error in the dynamic model, if the system noise is too small at this time, it will not be able to accurately describe the prediction parameter error, which will affect the subsequent measurement part. Therefore, in order to reasonably set the system noise, the embodiment of the present invention constructs a dynamic adjustment state of the adaptive factor, and through the dynamic adjustment of the adaptive factor, extracts the prediction vector and the prediction vector weight matrix, eliminates the difference between the carrier motion trajectory of the dynamic receiver and the prediction information, and improves the reliability of positioning.

[0188] S206: Introducing the constructed adaptive factor into the standard Kalman filter algorithm to obtain a robust adaptive Kalman filter algorithm.

[0189] Specifically, the constructed adaptive factor is introduced into the standard Kalman filter algorithm to obtain the formula of the robust adaptive Kalman filter algorithm: Among them, the subscript k represents the k signal frequency point, and the superscript T represents the transpose of the matrix. is the observation vector equivalent weight matrix, is the prediction vector weight matrix, A k is the coefficient matrix in the double-difference observation model, is the prediction vector adaptive factor, is the estimated value of the state vector, L k is the observation vector.

[0190] In summary, the method provided by the embodiment of the present invention utilizes the correlation between multipath and signal-to-noise ratio to construct an adaptive factor for the standard Kalman filter algorithm, and obtains a robust adaptive Kalman filter algorithm for identifying multipath errors, and constructs a multipath error inter-station differential signal-to-noise ratio rejection threshold Td based on historical observation data, and includes the measurement values ​​exceeding the threshold into the IGGIII rejection zone, so that when there are abnormal values ​​such as multipath errors in the observation data, the robust adaptive Kalman filter algorithm can automatically adjust the weights, thereby effectively eliminating the influence of abnormal values ​​such as multipath errors in the observation data on the estimation results, thereby improving the accuracy of the estimation results.

[0191] In addition, the method provided by the embodiment of the present invention also takes into account the situation that the low-cost terminal positioning in a complex environment may be disturbed by various error factors. In the prior art, when the low-cost terminal positioning in a complex environment is disturbed by various error factors, the existing Kalman filter algorithm is usually based on a traditional random model, and continuously predicts and updates the estimated value of the positioning system state to achieve the optimal estimation of the positioning system and obtain accurate positioning results. However, the traditional random model cannot accurately reflect the accuracy of the measurement value, so the measurement value variance is used as a standardized residual construction factor, and an accurate standardized residual cannot be constructed, resulting in filter divergence. The method provided by the embodiment of the present invention uses the median of the modulus length of the residual vector instead of the measurement value variance as an element for constructing the standardized residual. Similarly, the median of the modulus length of the prediction vector mismatch value is used instead of the measurement value variance as an element for constructing the adaptive factor, which can construct accurate standardized residuals and adaptive factors, further improving the accuracy of the estimation results of the robust adaptive Kalman filter algorithm.

[0192] and Figure 1 Corresponding to the embodiment shown, the embodiment of the present invention also provides a GNSS single-frequency precision single-point positioning device. Figure 3 As shown, the device includes: a construction module 301, a solution module 302, wherein:

[0193] The construction module 301 is used to construct a double-difference observation model according to the pseudo-range observation values ​​and carrier phase observation values ​​in the observation data received by the reference station and the monitoring station receiver, and obtain the pseudo-range double-difference observation values ​​and the carrier double-difference observation values, wherein the observation data includes: pseudo-range observation values, carrier phase observation values ​​and signal-to-noise ratio observation values;

[0194] A solving module 302 is used to solve the double-difference observation model using a preset robust adaptive Kalman filter algorithm to obtain a measurement value residual vector, and use the measurement value residual vector as the position information of the receiver at the monitoring station end;

[0195] The device also includes: an algorithm acquisition module, which is used to:

[0196] For each common-view satellite of the reference station and the monitoring station, the signal-to-noise ratio observation values ​​received by the base station and the monitoring station receivers are obtained respectively, and the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station are inter-station differentiated to obtain the inter-station differential signal-to-noise ratio corresponding to each common-view satellite;

[0197] Obtain the historical signal-to-noise ratio observation value of each common-view satellite received by the reference station and the monitoring station within a preset time period, and determine the inter-station differential signal-to-noise ratio threshold interval according to the historical signal-to-noise ratio observation value of each common-view satellite;

[0198] The double-difference observation model is filtered and solved using a standard Kalman filter algorithm to obtain a measurement value residual vector, and a standardized residual is determined based on the measurement value residual vector and the median of its modulus length; the measurement value residual vector includes: a pseudo-range measurement value residual vector and a carrier phase measurement value residual vector;

[0199] According to the weight and standardized residual of each observation vector, the preset division interval, the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval, an anti-error weight function is constructed; the observation vector is a pseudorange double difference observation vector and a carrier double difference observation vector;

[0200] According to the discrepancy values ​​of each component of the prediction vector and the robustness weight function, an adaptive factor is constructed for the standard Kalman filter algorithm;

[0201] The constructed adaptive factor is introduced into the standard Kalman filter algorithm to obtain the robust adaptive Kalman filter algorithm.

[0202] Optionally, the construction module 301 is specifically used for:

[0203] For each common-view satellite, the non-difference observation model corresponding to the reference station is constructed according to the pseudo-range observation values ​​and carrier phase observation values ​​received by the reference station receiver:

[0204]

[0205] The non-difference observation model corresponding to the monitoring station is constructed according to the pseudo-range observation values ​​and carrier phase observation values ​​received by the monitoring station receiver:

[0206]

[0207] Among them, the superscript s represents the common view satellite, the subscript r represents the receiver at the monitoring station, the subscript k represents the signal frequency, and c represents the speed of light in vacuum, c = 299792458 m / s. is the carrier wavelength of the common-view satellite s signal frequency k, It is the common-view satellite phase hardware delay. is the pseudo-range hardware delay between the satellite antenna and the satellite signal transmitter at the kth signal frequency, dt s represents the satellite clock error, represents the non-difference pseudorange observation value at the reference station, represents the non-difference carrier phase observation value at the reference station, Represents the geometric distance between the base station and the satellite, represents the tropospheric delay of the slant path at the reference station, represents the ionospheric delay of the slant path of the signal frequency k at the reference station. is the integer ambiguity at the base station; B b,k is the phase hardware delay of the base station receiver at the kth signal frequency point, b b,k is the pseudo-range hardware delay between the signal correlator at the kth signal frequency and the base station receiver antenna, and They represent the pseudorange measurement error and carrier phase measurement error at the reference station, dt b Indicates the base station receiver clock error, represents the non-difference pseudo-range observation value at the monitoring station, represents the non-difference carrier phase observation value at the monitoring station, represents the geometric distance between the monitoring station and the satellite, represents the slant path tropospheric delay at the monitoring station end; represents the ionospheric delay of the slant path of the signal frequency k at the monitoring station end, is the integer ambiguity at the monitoring station, B r,k is the phase hardware delay of the receiver at the kth signal frequency monitoring station, b r,k is the pseudo-range hardware delay between the signal correlator at the kth signal frequency and the receiver antenna at the monitoring station, and They represent the pseudorange measurement error and carrier phase measurement error at the monitoring station, respectively, and dt r Indicates the receiver clock error at the monitoring station;

[0208] Obtain the elevation angle of each common-view satellite of the reference station and the monitoring station, take the common-view satellite with the highest elevation angle as the reference satellite m, and take all the common-view satellites of the reference station and the monitoring station except the reference satellite m as non-reference satellites n;

[0209] For the reference satellite m, the non-difference pseudo-range observation values ​​corresponding to the monitoring station are used and carrier phase observations Subtract the non-difference pseudorange observation corresponding to the reference station and carrier phase observations The inter-station single-difference observation model corresponding to the reference satellite m is obtained:

[0210]

[0211] For each non-reference satellite n, the non-difference pseudo-range observation value corresponding to the monitoring station is used and carrier phase observations Subtract the non-difference pseudorange observation corresponding to the reference station and carrier phase observations The inter-station single-difference observation model corresponding to the non-reference satellite n is obtained:

[0212]

[0213] in, All represent single-difference pseudorange observations, Both represent single-difference carrier phase observations, denote the single-difference geometric distances between satellites m and n, denote the inter-station single-difference slant path tropospheric delays of satellites m and n, respectively, denote the inter-station single-difference slant path ionospheric delays of satellites m and n, respectively, denote the inter-station single-difference integer ambiguity of satellite m and n, respectively, and b b represents the pseudorange hardware delay of the base station receiver, b r represents the pseudorange hardware delay of the monitoring station receiver, denote the single-difference pseudorange observation noise between satellites m and n, denote the single-difference phase observation noise between satellites m and n, is the carrier wavelength of the common-view satellite m signal frequency point k, is the carrier wavelength of the common-view satellite n signal frequency k, B b represents the base station receiver phase hardware delay, B r Indicates the phase hardware delay of the monitoring station receiver;

[0214] For each non-reference satellite n, the non-difference pseudo-range observation value of the monitoring station corresponding to the reference satellite m is used and carrier phase observations Subtract the undifferenced pseudorange observation value of the reference station corresponding to the non-reference satellite n and carrier phase observations The double difference observation model is obtained:

[0215]

[0216] in, represents the double-difference pseudorange observation value, represents the double-difference carrier phase observation, represents the double-difference slant-path tropospheric delay; represents the slant path ionospheric delay; λ k is the carrier wavelength of the common-view satellite signal frequency k; is the double difference integer ambiguity; and They represent pseudorange measurement error and carrier phase measurement error respectively.

[0217] Optional, algorithm acquisition module, specifically used for:

[0218] Under the assumption that the receivers and antennas of the base station and the monitoring station are of the same type and the observation environment is an ideal unobstructed environment, for each common-view satellite of the base station and the monitoring station, obtain the signal-to-noise ratio observation value received by the base station receiver and the signal-to-noise ratio observation received by the monitoring station receiver

[0219] The signal-to-noise ratio observation received by the base station receiver Subtract the signal-to-noise ratio observation received by the monitoring station receiver Get the inter-station differential signal-to-noise ratio corresponding to the common-view satellite

[0220] Optional, algorithm acquisition module, specifically used for:

[0221] For each satellite in common view between the base station and the monitoring station, obtain the signal-to-noise ratio observation value received by the base station receiver within the preset historical time period. and the signal-to-noise ratio observation received by the monitoring station receiver

[0222] Use the signal-to-noise ratio observations received by the base station receiver during the historical period Subtract the signal-to-noise ratio observation received by the monitoring station receiver Get the inter-station differential signal-to-noise ratio corresponding to the common-view satellite in the historical time period

[0223] For each common-view satellite, the standard deviation σ of the inter-station differential signal-to-noise ratio corresponding to the common-view satellite in the historical time period is calculated respectively, and the Laida criterion is used to calculate the inter-station differential signal-to-noise ratio corresponding to each common-view satellite in the interval range of [-3σ, 3σ] Determine the reliability interval

[0224] According to the reliable interval range corresponding to each common-view satellite;

[0225] Using the formula Determine the threshold interval of the differential signal-to-noise ratio between stations [(ΔSNR) min ,(ΔSNR) max ], where superscript s indicates common-view satellites; floor() indicates rounding down, and roundup() indicates rounding up;

[0226] Optional, algorithm acquisition module, specifically used for:

[0227] Use the standard Kalman filter algorithm to filter and solve the double difference observation model to obtain the measurement value residual vector v i ,in, Subscript k represents the signal frequency k, v i is the residual vector of the i-th measurement value, R i is the system noise matrix of the ith measurement value residual vector, represents the variance-covariance matrix of the measurement residual vector, represents the variance-covariance matrix of the residual vector of the measured value prior to the test, The transpose of the system noise matrix representing the residual vector of the measurement value at the signal frequency point k;

[0228] Using the formula Calculate the standardized residual |v i |, where sigma0 is the median modulus of the measurement value residual vector.

[0229] Optional, algorithm acquisition module, specifically used for:

[0230] According to the weight of each observation vector, the standardized residual of each observation vector, the preset partition intervals [1.0, 1.5] and [2.0, 3.0], and the differential signal-to-noise ratio between stations and the threshold interval of the differential signal-to-noise ratio between stations [(ΔSNR) min ,(ΔSNR) max ], construct the robust weight function:

[0231] in, represents the weight of the observation vector i after the robust estimation adjustment, pi represents the weight of observation vector i, |v i | represents the standardized residual of observation vector i, Td represents the differential signal-to-noise ratio threshold between stations, and the value range of Td is [(ΔSNR) min ,(ΔSNR) max ], k0 and k1 are both constants, k0∈[1.0,1.5], k1∈[2.0,3.0], |v i The area with |≤k0 is the copyright protection area, and k0<|v i The area with |≤k1 is the downgraded area and The area is a denial zone.

[0232] Optional, algorithm acquisition module, specifically used for:

[0233] According to the weights adjusted after the robust estimation of each observation vector, the state vector robust estimation of each observation vector at the k signal frequency point is calculated.

[0234] According to the state vector robust estimation and double difference observation model of each observation vector at the k-1th signal frequency point, the state prediction vector of the observation vector at the kth signal frequency point is calculated and obtained.

[0235] Using the formula Calculate the discrepancy value of each state prediction vector, where represents the discrepancy value of the j-th state prediction vector;

[0236] According to the discrepancy values ​​of each component of the prediction vector and the anti-error weight function, the formula is used Constructing an adaptive factor for the standard Kalman filter algorithm, sigma1 represents the median length of the prediction vector mismatch value; the subscript k represents the k signal frequency point; represents the adaptive factor of the j-th state prediction vector; c0 and c1 are the thresholds of the adaptive factor, c0 = 2.5, c1 = 7;

[0237] Introduce the constructed adaptive factor into the standard Kalman filter algorithm to obtain the formula of the robust adaptive Kalman filter algorithm: Among them, the subscript k represents the k signal frequency point, and the superscript T represents the transpose of the matrix. is the observation vector equivalent weight matrix, is the prediction vector weight matrix, A k is the coefficient matrix in the double-difference observation model, is the prediction vector adaptive factor, is the estimated value of the state vector, L k is the observation vector.

[0238] Optionally, the device further includes: a verification module, configured to:

[0239] For each measurement value residual vector, respectively, according to the measurement value residual vector v k and the measurement value residual vector v k The corresponding variance-covariance matrix Construction Statistics λ k In order to obey the chi-square distribution with m degrees of freedom, determine whether the measured value residual vector belongs to the interval (Q1-C1·IQR, Q3+C1·IQR). If it does, keep the measured value residual vector; if it does not, remove the measured value residual vector. Q1 and Q3 are the residual values ​​of the measured value residual vector at the 25% and 75% positions after the residual values ​​of each measured value residual vector are arranged from small to large. IQR=Q3-Q1, C1=1.5, and the superscript T represents the transpose of the matrix.

[0240] The retained measurement residual vector is used as the location information of the receiver at the monitoring station.

[0241] In summary, the device provided by the embodiment of the present invention utilizes the correlation between multipath and signal-to-noise ratio to construct an adaptive factor for the standard Kalman filter algorithm, and obtains a robust adaptive Kalman filter algorithm for identifying multipath errors, and constructs a multipath error inter-station differential signal-to-noise ratio rejection threshold Td based on historical observation data, and includes the measurement values ​​exceeding the threshold into the IGGIII rejection zone, so that when there are abnormal values ​​such as multipath errors in the observation data, the robust adaptive Kalman filter algorithm can automatically adjust the weights, thereby effectively eliminating the influence of abnormal values ​​such as multipath errors in the observation data on the estimation results, thereby improving the accuracy of the estimation results.

[0242] and Figure 1 Corresponding to the embodiment shown, the embodiment of the present invention further provides an electronic device, see Figure 4 , including a processor 401, a communication interface 402, a memory 403 and a communication bus 404, wherein the processor 401, the communication interface 402, and the memory 403 communicate with each other through the communication bus 404;

[0243] Memory 403, used for storing computer programs;

[0244] The processor 401 is used to implement any of the robust adaptive positioning method steps described in the above embodiments when executing the program stored in the memory.

[0245] The electronic device provided by the embodiment of the present invention utilizes the correlation between multipath and signal-to-noise ratio to construct an adaptive factor for the standard Kalman filtering algorithm, and obtains a robust adaptive Kalman filtering algorithm for identifying multipath errors, and constructs a multipath error station-to-station differential signal-to-noise ratio rejection threshold Td based on historical observation data, and includes measurement values ​​exceeding the threshold into the IGGIII rejection zone, so that when there are abnormal values ​​such as multipath errors in the observation data, the robust adaptive Kalman filtering algorithm can automatically adjust the weights, thereby effectively eliminating the influence of abnormal values ​​such as multipath errors in the observation data on the estimation results, thereby improving the accuracy of the estimation results.

[0246] and Figure 1 Corresponding to the embodiment shown, an embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the method steps of any one of the anti-differential adaptive positioning methods described in the above embodiments are implemented.

[0247] The storage device provided by the embodiment of the present invention utilizes the correlation between multipath and signal-to-noise ratio to construct an adaptive factor for the standard Kalman filter algorithm, and obtains a robust adaptive Kalman filter algorithm for identifying multipath errors, and constructs a multipath error station-to-station differential signal-to-noise ratio rejection threshold Td based on historical observation data, and includes measurement values ​​exceeding the threshold into the IGGIII rejection zone, so that when there are abnormal values ​​such as multipath errors in the observation data, the robust adaptive Kalman filter algorithm can automatically adjust the weights, thereby effectively eliminating the influence of abnormal values ​​such as multipath errors in the observation data on the estimation results, thereby improving the accuracy of the estimation results.

[0248] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0249] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0250] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0251] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0252] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A robust adaptive positioning method, characterized in that: The method comprises: According to the pseudorange observation values ​​and carrier phase observation values ​​in the observation data received by the base station and the monitoring station receivers, a double difference observation model is constructed, and pseudorange double difference observation values ​​and carrier double difference observation values ​​are obtained, wherein the observation data includes: pseudorange observation values, carrier phase observation values ​​and signal-to-noise ratio observation values; The double difference observation model is solved by using a preset robust adaptive Kalman filter algorithm to obtain a measurement value residual vector, and the measurement value residual vector is used as the position information of the receiver at the monitoring station end; wherein the acquisition process of the robust adaptive Kalman filter algorithm is: For each common-view satellite of the reference station and the monitoring station, respectively obtain the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station receivers, and perform inter-station difference on the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station to obtain the inter-station differential signal-to-noise ratio corresponding to each common-view satellite; Obtaining historical signal-to-noise ratio observation values ​​of each common-view satellite received by the reference station and the monitoring station within a preset time period, and determining an inter-station differential signal-to-noise ratio threshold interval according to the historical signal-to-noise ratio observation values ​​of each common-view satellite; The double-difference observation model is filtered and solved using a standard Kalman filter algorithm to obtain a measurement value residual vector, and a standardized residual is determined according to the measurement value residual vector and the median of its modulus length; the measurement value residual vector includes: a pseudorange measurement value residual vector and a carrier phase measurement value residual vector; According to the weight and standardized residual of each observation vector, the preset division interval, the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval, an anti-error weight function is constructed; the observation vector is a pseudorange double difference observation vector and a carrier double difference observation vector; Constructing an adaptive factor for a standard Kalman filter algorithm based on the discrepancy values ​​of each component of the prediction vector and the robust weight function; The constructed adaptive factor is introduced into the standard Kalman filter algorithm to obtain a robust adaptive Kalman filter algorithm.

2. The method according to claim 1, characterized in that The step of constructing a double-difference observation model according to pseudo-range observation values ​​and carrier phase observation values ​​in the observation data received by the reference station and the monitoring station receivers includes: For each common-view satellite, the non-difference observation model corresponding to the reference station is constructed according to the pseudo-range observation values ​​and carrier phase observation values ​​received by the reference station receiver: The non-difference observation model corresponding to the monitoring station is constructed based on the pseudo-range observation values ​​and carrier phase observation values ​​received by the monitoring station receiver: Among them, the superscript s represents the common view satellite, the subscript r represents the receiver at the monitoring station, the subscript k represents the signal frequency, and c represents the speed of light in vacuum, c = 299792458 m / s. is the carrier wavelength of the common-view satellite s signal frequency k, It is the common-view satellite phase hardware delay. is the pseudo-range hardware delay between the satellite antenna and the satellite signal transmitter at the kth signal frequency, dt s represents the satellite clock error, represents the non-difference pseudorange observation value at the reference station, represents the non-difference carrier phase observation value at the reference station, Represents the geometric distance between the base station and the satellite, represents the tropospheric delay of the slant path at the reference station, represents the ionospheric delay of the slant path of the signal frequency k at the reference station. is the integer ambiguity at the base station; B b,k is the phase hardware delay of the base station receiver at the kth signal frequency point, b b,k is the pseudo-range hardware delay between the signal correlator at the kth signal frequency and the base station receiver antenna, and They represent the pseudorange measurement error and carrier phase measurement error at the reference station, dt b Indicates the base station receiver clock error, represents the non-difference pseudo-range observation value at the monitoring station, represents the non-difference carrier phase observation value at the monitoring station, represents the geometric distance between the monitoring station and the satellite, represents the slant path tropospheric delay at the monitoring station end; represents the ionospheric delay of the slant path of the signal frequency k at the monitoring station end, is the integer ambiguity at the monitoring station, B r,k is the phase hardware delay of the receiver at the kth signal frequency monitoring station, b r,k is the pseudo-range hardware delay between the signal correlator at the kth signal frequency and the receiver antenna at the monitoring station, and They represent the pseudorange measurement error and carrier phase measurement error at the monitoring station, respectively, and dt r Indicates the receiver clock error at the monitoring station; Obtain the elevation angle of each common-view satellite of the reference station and the monitoring station, take the common-view satellite with the highest elevation angle as the reference satellite m, and take all the common-view satellites of the reference station and the monitoring station except the reference satellite m as non-reference satellites n; For the reference satellite m, the non-difference pseudo-range observation values ​​corresponding to the monitoring station are used and carrier phase observations Subtract the non-difference pseudorange observation corresponding to the reference station and carrier phase observations The inter-station single-difference observation model corresponding to the reference satellite m is obtained: For each non-reference satellite n, the non-difference pseudo-range observation value corresponding to the monitoring station is used and carrier phase observations Subtract the non-difference pseudorange observation corresponding to the reference station and carrier phase observations The inter-station single-difference observation model corresponding to the non-reference satellite n is obtained: in, All represent single-difference pseudorange observations, Both represent single-difference carrier phase observations, denote the single-difference geometric distances between satellites m and n, denote the inter-station single-difference slant path tropospheric delays of satellites m and n, respectively, denote the inter-station single-difference slant path ionospheric delays of satellites m and n, respectively, denote the inter-station single-difference integer ambiguity of satellite m and n, respectively, and b b represents the pseudorange hardware delay of the base station receiver, b r represents the pseudorange hardware delay of the monitoring station receiver, denote the single-difference pseudorange observation noise between satellites m and n, denote the single-difference phase observation noise between satellites m and n, is the carrier wavelength of the common-view satellite m signal frequency point k, is the carrier wavelength of the common-view satellite n signal frequency k, B b represents the base station receiver phase hardware delay, B r Indicates the phase hardware delay of the monitoring station receiver; For each non-reference satellite n, the non-difference pseudo-range observation value of the monitoring station corresponding to the reference satellite m is used and carrier phase observations Subtract the undifferenced pseudorange observation value of the reference station corresponding to the non-reference satellite n and carrier phase observations The double difference observation model is obtained: in, represents the double-difference pseudorange observation value, represents the double-difference carrier phase observation value, represents the double-difference slant-path tropospheric delay; represents the slant path ionospheric delay; λ k is the carrier wavelength of the common-view satellite signal frequency k; is the double difference integer ambiguity; and They represent pseudorange measurement error and carrier phase measurement error respectively.

3. The method according to claim 1, characterized in that: The step of obtaining, for each common-view satellite of the reference station and the monitoring station, respectively the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station receivers, performing inter-station differences on the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station, and obtaining the inter-station differential signal-to-noise ratio corresponding to each common-view satellite includes: Under the assumption that the receivers and antennas of the base station and the monitoring station are of the same type and the observation environment is an ideal unobstructed environment, for each common-view satellite of the base station and the monitoring station, obtain the signal-to-noise ratio observation value received by the base station receiver and the signal-to-noise ratio observation received by the monitoring station receiver The signal-to-noise ratio observation received by the base station receiver Subtract the signal-to-noise ratio observation received by the monitoring station receiver Obtain the inter-station differential signal-to-noise ratio corresponding to the common-view satellite 4. The method according to claim 3, characterized in that The step of obtaining the historical signal-to-noise ratio observation value of each common-view satellite received by the reference station and the monitoring station within a preset time period, and determining the inter-station differential signal-to-noise ratio threshold range according to the historical signal-to-noise ratio observation value of each common-view satellite, comprises: For each satellite in common view between the base station and the monitoring station, obtain the signal-to-noise ratio observation value received by the base station receiver within the preset historical time period. and the signal-to-noise ratio observation received by the monitoring station receiver The signal-to-noise ratio observation value received by the base station receiver during the historical period is used Subtract the signal-to-noise ratio observation received by the monitoring station receiver Obtain the inter-station differential signal-to-noise ratio corresponding to the common-view satellite during the historical time period For each common-view satellite, the standard deviation σ of the inter-station differential signal-to-noise ratio corresponding to the common-view satellite in the historical time period is calculated respectively, and the Laida criterion is used to calculate the inter-station differential signal-to-noise ratio corresponding to each common-view satellite in the interval range of [-3σ, 3σ] Determine the reliability interval According to the reliable interval range corresponding to each common-view satellite; Using the formula Determine the threshold interval of the differential signal-to-noise ratio between stations [(ΔSNR) min ,(ΔSNR) max ], where superscript s indicates common-view satellites; floor() indicates rounding down, and roundup() indicates rounding up; The step of filtering and solving the double-difference observation model using a standard Kalman filter algorithm to obtain a measurement value residual vector, and determining a standardized residual according to the measurement value residual vector and the median of its modulus length includes: The double-difference observation model is filtered and solved using the standard Kalman filter algorithm to obtain the measurement value residual vector v i ,in, Subscript k represents the signal frequency k, v i is the residual vector of the i-th measurement value, R i is the system noise matrix of the ith measurement value residual vector, represents the variance-covariance matrix of the measurement residual vector, represents the variance-covariance matrix of the residual vector of the measured value prior to the test, represents the transpose of the variance amplification factor matrix of signal frequency point k; Using the formula Calculate the standardized residual |v i |, where sigma0 is the median modulus of the measurement value residual vector.

5. The method according to claim 4, characterized in that The step of constructing an anti-error weight function according to the weight and standardized residual of each observation vector, the preset division interval, the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval includes: According to the weight of each observation vector, the standardized residual of each observation vector, the preset partition intervals [1.0, 1.5] and [2.0, 3.0], and the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval [(ΔSNR) min ,(ΔSNR) max ], construct the robust weight function: in, represents the weight of the observation vector i after the robust estimation adjustment, p i represents the weight of observation vector i, |v i | represents the standardized residual of observation vector i, Td represents the differential signal-to-noise ratio threshold between stations, and the value range of Td is [(ΔSNR) min ,(ΔSNR) max ], k0 and k1 are both constants, k0∈[1.0,1.5], k1∈[2.0,3.0], |v i The area with |≤k0 is the copyright protection area, and k0<|v i The area with |≤k1 is the downgraded area and The area is a denial zone.

6. The method according to claim 5, characterized in that The step of constructing an adaptive factor for the standard Kalman filter algorithm according to the mismatch values ​​of each component of the prediction vector and the robustness weight function comprises: According to the weights adjusted after the robust estimation of each observation vector, the state vector robust estimation of each observation vector at the k signal frequency point is calculated. The state prediction vector of the observation vector at the kth signal frequency point is calculated based on the state vector robust estimation of the k-1th signal frequency point of each observation vector and the double difference observation model. Using the formula Calculate the discrepancy value of each state prediction vector, where represents the discrepancy value of the j-th state prediction vector; According to the discrepancy values ​​of each component of the prediction vector and the anti-error weight function, the formula is used Constructing an adaptive factor for the standard Kalman filter algorithm, sigma1 represents the median length of the prediction vector mismatch value; the subscript k represents the k signal frequency point; represents the adaptive factor of the j-th state prediction vector; c0 and c1 are the thresholds of the adaptive factor, c0 = 2.5, c1 = 7; The constructed adaptive factor is introduced into the standard Kalman filter algorithm to obtain the formula of the robust adaptive Kalman filter algorithm: Among them, the subscript k represents the k signal frequency point, and the superscript T represents the transpose of the matrix. is the observation vector equivalent weight matrix, is the prediction vector weight matrix, A k is the coefficient matrix in the double-difference observation model, is the prediction vector adaptive factor, is the estimated value of the state vector, L k is the observation vector.

7. The method according to claim 1, characterized in that After the step of solving the double difference observation model by using a preset robust adaptive Kalman filter algorithm to obtain a measurement value residual vector, the method further includes: For each measurement value residual vector, respectively, according to the measurement value residual vector v k and the measurement value residual vector v k The corresponding variance-covariance matrix Construction Statistics λ k To obey the chi-square distribution with m degrees of freedom, determine whether the measurement value residual vector belongs to the interval (Q1-C1·IQR,Q3+C1·IQR), if it belongs to, then keep the measurement value residual vector; if it does not belong to, then remove the measurement value residual vector, Q1, Q3 are the residual values ​​of the measurement value residual vector at the 25% and 75% positions after the residual values ​​of each measurement value residual vector are arranged from small to large, IQR=Q3-Q1, C1=1.5, and the superscript T represents the transpose of the matrix; The retained measurement residual vector is used as the position information of the receiver at the monitoring station.

8. A robust and adaptive positioning device, characterized in that: The device comprises: A construction module is used to construct a double-difference observation model according to pseudo-range observation values ​​and carrier phase observation values ​​in the observation data received by the base station and the monitoring station receiver, and obtain pseudo-range double-difference observation values ​​and carrier double-difference observation values, wherein the observation data includes: pseudo-range observation values, carrier phase observation values ​​and signal-to-noise ratio observation values; A solution module, used to solve the double-difference observation model by using a preset robust adaptive Kalman filter algorithm, obtain a measurement value residual vector, and use the measurement value residual vector as the position information of the receiver at the monitoring station end; The device also includes: an algorithm acquisition module, which is used to: For each common-view satellite of the reference station and the monitoring station, respectively obtain the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station receivers, and perform inter-station difference on the signal-to-noise ratio observation values ​​received by the reference station and the monitoring station to obtain the inter-station differential signal-to-noise ratio corresponding to each common-view satellite; Obtaining historical signal-to-noise ratio observation values ​​of each common-view satellite received by the reference station and the monitoring station within a preset time period, and determining an inter-station differential signal-to-noise ratio threshold interval according to the historical signal-to-noise ratio observation values ​​of each common-view satellite; The double-difference observation model is filtered and solved using a standard Kalman filter algorithm to obtain a measurement value residual vector, and a standardized residual is determined according to the measurement value residual vector and the median of its modulus length; the measurement value residual vector includes: a pseudorange measurement value residual vector and a carrier phase measurement value residual vector; According to the weight and standardized residual of each observation vector, the preset division interval, the inter-station differential signal-to-noise ratio and the inter-station differential signal-to-noise ratio threshold interval, an anti-error weight function is constructed; the observation vector is a pseudorange double difference observation vector and a carrier double difference observation vector; Constructing an adaptive factor for a standard Kalman filter algorithm based on the discrepancy values ​​of each component of the prediction vector and the robustness weight function; The constructed adaptive factor is introduced into the standard Kalman filter algorithm to obtain a robust adaptive Kalman filter algorithm.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; The memory is used to store computer programs; The processor is used to implement the method steps described in any one of claims 1-7 when executing the program stored in the memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps described in any one of claims 1 to 7 are implemented.

Images