A precise attitude determination method and system based on stepwise resolution of GNSS multi-antenna multi-frequency ambiguity

Abstract

This invention discloses a precise attitude determination method and system based on the step-by-step resolution of GNSS multi-antenna multi-frequency ambiguities. First, the coordinates of the reference antenna in the ECEF coordinate system are calculated using reference antenna observations. Then, the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna is calculated. Next, based on GNSS multi-antenna multi-frequency observation data, a double-difference pseudorange and carrier phase observation equation is constructed between the reference antenna and each auxiliary antenna on the carrier. Subsequently, based on the double-difference pseudorange and carrier phase observation equation, a wide-lane carrier phase observation equation is constructed, and the double-difference wide-lane ambiguity of each baseline is obtained. This invention achieves high-precision attitude determination by step-by-step resolution of GNSS multi-antenna multi-frequency ambiguities and a combination of direct and least squares methods. Furthermore, by utilizing the advantage of easily fixed ambiguities in multi-frequency combined observations, the resolution efficiency of GNSS multi-frequency attitude determination is improved. This not only makes reasonable use of GNSS multi-frequency observations but also improves the accuracy of attitude determination.

Description

Technical Field

[0001] This invention relates to the field of satellite navigation attitude determination technology, specifically to a precise attitude determination method and system based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution. Background Technology

[0002] Global Navigation Satellite Systems (GNSS) have become an indispensable infrastructure in modern economic development and daily life. With the gradual improvement of GNSS constellations and the increasing broadcasting of multi-frequency signals, high-precision attitude determination technology has rapidly evolved from traditional single-frequency and single-baseline modes to a highly reliable solution framework based on multiple antennas and multiple frequencies. Successful ambiguity fixation is crucial for achieving high-precision attitude determination. The BeiDou-3 and Galileo satellite navigation systems broadcast observation data in five civilian frequency bands: B1C / B1I / B3I / B2a / B2b and E1 / E6 / E5a / E5 / E5b. The dissemination of multi-frequency observation signals provides a signal foundation for constructing long-wavelength, low-noise combined observation values.

[0003] Currently, most existing GNSS attitude determination models typically solve for each baseline first and then calculate the carrier attitude angle based on the baseline. However, they fix the original ambiguity with a relatively short wavelength. As the frequency increases, the ambiguity dimension increases significantly, leading to a significant decrease in the computational efficiency of the attitude determination model. At the same time, simple stacking of observations generally cannot fully utilize the advantages of multi-frequency observation combination signals in ambiguity resolution, and using baselines to solve attitude angles cannot reasonably allocate the redundant information of the baselines, resulting in a dilution of attitude angle accuracy. Therefore, after solving for the attitude angles, the least squares attitude solution equation is listed and the attitude angles are solved again. However, this process has low computational efficiency for high-dimensional matrix operations and cannot utilize the advantages of multi-frequency observation combination signals in ambiguity resolution. Therefore, it is necessary to design a precise attitude determination method and system based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies. To better and more effectively address the limitations of current GNSS attitude determination models, which typically first solve for each baseline and then calculate the carrier attitude angles based on the baselines, but fix the original ambiguity with a relatively short wavelength, this leads to a significant increase in the ambiguity dimension as the frequency increases, resulting in a significant decrease in the computational efficiency of the attitude determination model. Furthermore, simple stacking of observations generally fails to fully utilize the advantages of multi-frequency observation combination signals in ambiguity resolution, and using baselines to solve for attitude angles cannot reasonably allocate redundant baseline information, causing a dilution of attitude angle accuracy. Therefore, based on the solved attitude angles, a least-squares attitude solution method is proposed. The process involves recalculating the attitude angle, but this process is computationally inefficient for high-dimensional matrix operations and cannot utilize the advantages of multi-frequency observation combination signals in ambiguity resolution. To address this, a precise attitude determination method and system based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution is proposed. This method achieves high-precision attitude determination by step-by-step resolution of GNSS multi-antenna multi-frequency ambiguity and a combination of direct and least squares methods. Furthermore, by utilizing the advantage of easily fixed ambiguity in multi-frequency combined observations, the resolution efficiency of GNSS multi-frequency attitude determination is improved. This not only makes reasonable use of GNSS multi-frequency observations but also improves the accuracy of attitude resolution.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A precise attitude determination method based on step-by-step ambiguity resolution using GNSS multi-antenna multi-frequency methods includes the following steps:

[0007] Step A: Calculate the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observations, and then calculate the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna.

[0008] Step B: Construct the double-difference pseudorange and carrier phase observation equations between the reference antenna and each auxiliary antenna on the carrier based on GNSS multi-antenna multi-frequency observation data;

[0009] Step C: Construct the wide-lane carrier phase observation equation based on the double-difference pseudorange and carrier phase observation equation and obtain the double-difference wide-lane ambiguity of each baseline. Then fix the double-difference wide-lane ambiguity of each baseline and obtain the baseline vector of each baseline in the ECEF coordinate system.

[0010] Step D: Based on the rotation matrix from the ECEF coordinate system to the navigation coordinate system, the baseline vectors of each baseline in the ECEF coordinate system are transformed to the navigation coordinate system to obtain the coordinates of each baseline in the navigation coordinate system. Then, the approximate attitude angle of the vehicle is calculated using the coordinates of each baseline in the navigation coordinate system.

[0011] Step E: Construct a general attitude rotation matrix from the general vehicle coordinate system to the navigation coordinate system using the general attitude angles of the vehicle.

[0012] Step F: Construct the real attitude rotation matrix from the real vehicle coordinate system to the navigation coordinate system, and decompose the real rotation matrix into a general attitude rotation matrix and an attitude update rotation matrix. Then, decompose the attitude update rotation matrix and obtain the attitude decomposition result.

[0013] Step G: Construct an attitude solution model based on the attitude decomposition results, and then use the least squares method to calculate the attitude solution model to obtain the floating-point solution of the carrier phase ambiguity of the first frequency band and the variance-covariance matrix of the carrier phase ambiguity of the first frequency band.

[0014] Step H: Based on the floating-point solution of the carrier phase ambiguity of the first frequency band and the variance-covariance matrix of the carrier phase ambiguity of the first frequency band, the LAMBDA algorithm is used to fix the carrier phase ambiguity of the first frequency band and obtain the carrier attitude angle correction. Then, the carrier attitude angle correction is combined with the carrier approximate attitude angle to obtain the carrier true attitude angle.

[0015] The aforementioned precise attitude determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution, step A, calculates the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observations, and then calculates the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna. The specific steps are as follows.

[0016] Step A1: Calculate the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observations. The antenna of the GNSS receiver on the carrier includes one reference antenna and three auxiliary antennas. Specifically, the coordinates of the reference antenna in the ECEF coordinate system are calculated based on the reference antenna observations using real-time dynamic differential RTK positioning, single-point positioning, precise single-point positioning and INS combined positioning.

[0017] Step A2: Calculate the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna, as shown in formula (1).

[0018] (1)

[0019] in, This is the rotation matrix from the ECEF coordinate system to the navigation coordinate system. The latitude of the reference antenna. The longitude of the reference antenna.

[0020] The aforementioned precise attitude determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution, step B, involves constructing the double-difference pseudorange and carrier phase observation equations between the reference antenna and each auxiliary antenna on the carrier based on GNSS multi-antenna multi-frequency observation data, as shown in formula (2).

[0021] (2)

[0022] in, As a reference antenna, As an auxiliary antenna, For the station star double difference operator, The pseudorange observation values ​​for the nth frequency band are those of the reference antenna a and the auxiliary antenna i. The nth frequency band carrier phase observation values ​​of reference antenna a and auxiliary antenna i are given. The single-difference line-of-sight matrix of the reference antenna. This is a diagonal matrix whose diagonal values ​​are the wavelengths of the nth frequency band. It is a zero matrix. Baseline Baseline vector coordinates in the ECEF coordinate system The double-difference ambiguity of the nth frequency band between the reference antenna a and the auxiliary antenna i. This is to account for observation noise error.

[0023] The aforementioned precise attitude determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution, step C, involves constructing a wide-lane carrier phase observation equation based on the double-difference pseudorange and carrier phase observation equations and obtaining the double-difference wide-lane ambiguity of each baseline. Then, the double-difference wide-lane ambiguity of each baseline is fixed, and the baseline vector of each baseline in the ECEF coordinate system is obtained. The specific steps are as follows.

[0024] Step C1: Construct the wide-lane carrier phase observation equation based on the double-difference pseudorange and carrier phase observation equation, and obtain the double-difference wide-lane ambiguity of each baseline, as shown in formula (3).

[0025] (3)

[0026] in, Baseline Double-difference wide-lane carrier phase observations of the combination of frequency band 1 and frequency band n. This is a diagonal matrix whose diagonal values ​​are the combined wavelengths of the first and nth frequency bands. Baseline Wide-lane carrier phase ambiguity in combination of the first frequency band and the nth frequency band;

[0027] Step C2 involves fixing the double-difference wide-lane ambiguity of each baseline and obtaining the baseline vector of each baseline in the ECEF coordinate system. Specifically, this is achieved by using the wide-lane carrier phase observation equation combined with the elevation angle weighting model and the least squares method to obtain the floating-point solution of the double-difference wide-lane ambiguity of each baseline and the variance-covariance matrix of the double-difference wide-lane ambiguity of each baseline. Then, the LAMBDA algorithm or Bootstrapping algorithm is used to fix the double-difference wide-lane ambiguity of each baseline and obtain the baseline vector of each baseline in the ECEF coordinate system. .

[0028] The aforementioned precise attitude determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution, in step D, involves transforming the baseline vectors of each baseline in the ECEF coordinate system to the navigation coordinate system using the rotation matrix from the ECEF coordinate system to the navigation coordinate system, thereby obtaining the coordinates of each baseline in the navigation coordinate system. Then, the approximate attitude angles of the vehicle are calculated using the coordinates of each baseline in the navigation coordinate system. The specific steps are as follows.

[0029] Step D1: Based on the rotation matrix from the ECEF coordinate system to the navigation coordinate system, transform the baseline vectors of each baseline in the ECEF coordinate system to the navigation coordinate system to obtain the coordinates of each baseline in the navigation coordinate system, as shown in formula (4).

[0030] (4)

[0031] in, This represents the baseline vector of each baseline in the navigation coordinate system;

[0032] Step D2: Calculate the approximate attitude angles of the vehicle using the coordinates of each baseline in the navigation coordinate system, wherein the approximate attitude angles of the vehicle include the approximate heading angles of the vehicle. Carrier general pitch angle and the approximate roll angle of the carrier The specific steps are as follows:

[0033] Step D21, based on a baseline parallel to the longitudinal axis of the carrier. Calculate the approximate heading angle of the vehicle. and approximate pitch angle of the carrier Specifically, as shown in formulas (5) and (6) respectively,

[0034] (5)

[0035] (6)

[0036] in, Baseline The baseline vector in the navigation coordinate system;

[0037] Step D22, using baseline Combined with approximate heading angle and approximate pitch angle Calculate the approximate roll angle of the carrier. Specifically, as shown in formula (7),

[0038] (7)

[0039] in, Baseline The baseline vector in the navigation coordinate system.

[0040] In the aforementioned precise attitude determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution, step E involves constructing a general attitude rotation matrix from the general vehicle coordinate system to the navigation coordinate system using the general vehicle attitude angle, as shown in formula (8).

[0041] (8)

[0042] in, This is a rough attitude rotation matrix from the approximate vehicle coordinate system to the navigation coordinate system. For approximate attitude coordinate system, For cosine, It is the sine.

[0043] The aforementioned precise attitude determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution, in step F, involves constructing a true attitude rotation matrix from the real carrier coordinate system to the navigation coordinate system, decomposing the true rotation matrix into a preliminary attitude rotation matrix and an attitude update rotation matrix, and then decomposing the attitude update rotation matrix to obtain the attitude decomposition result. The specific steps are as follows.

[0044] Step F1: Construct the true attitude rotation matrix from the real vehicle coordinate system to the navigation coordinate system, and decompose the true rotation matrix into a preliminary attitude rotation matrix and an attitude update rotation matrix, as shown in formula (9).

[0045] (9)

[0046] in, This is the true attitude rotation matrix from the real vehicle coordinate system to the navigation coordinate system. This is the transformation matrix from the real carrier coordinate system to the approximate carrier coordinate system. For the true pitch angle of the carrier Approximate pitch angle of the carrier difference, For the true roll angle of the carrier Approximate roll angle of the carrier difference, For the carrier's true heading angle Approximate heading angle of the vehicle difference;

[0047] Step F2 involves decomposing the pose update rotation matrix and obtaining the pose decomposition result, as shown in formula (10).

[0048] ;

[0049] (10)

[0050] in, The difference between the actual posture of the carrier and its approximate posture. It is a 3×3 identity matrix, with superscript... It is a differential rotation matrix.

[0051] The aforementioned precise attitude determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution, step G, involves constructing an attitude resolution model based on the attitude decomposition results, and then using the least squares method to calculate the floating-point solution of the first frequency band carrier phase ambiguity and the variance-covariance matrix of the first frequency band carrier phase ambiguity. The specific steps are as follows.

[0052] Step G1: Construct an attitude calculation model based on the attitude decomposition results. The specific steps are as follows.

[0053] Step G11, establish baseline The attitude calculation model is shown in Equation (11).

[0054] , ;

[0055] ;

[0056] ;

[0057] (11)

[0058] in, These are the pseudorange observations for all frequency bands of the reference antenna a and the auxiliary antenna i. Baseline The double-difference wide-lane carrier phase observations for the first frequency band and all frequency band combinations, where E is the double-difference operator matrix. For combined wavelengths, It is a column vector, specifically n×1. , Use the ECEF coordinate system;

[0059] Step G12, for the baseline The attitude calculation model is parameterized and a baseline is obtained. The parameterized attitude calculation model is shown in Equation (12).

[0060] , , , ;

[0061] (12)

[0062] in, Baseline Double difference observations For the line-of-sight matrix, Baseline The ambiguity parameter design matrix, For wavelength matrix, It is a column vector;

[0063] Step G13, based on the baseline Establishing a baseline for the parameterized attitude solution model The attitude error equation is shown in equation (13).

[0064] ;

[0065] ;

[0066] (13)

[0067] in, Baseline The difference vector between the calculated value and the observed value. Baseline The design matrix for attitude angle update parameters, Update parameters for attitude angles;

[0068] Step G14, based on baseline The attitude error equations of baseline ab, baseline ac and baseline ad are constructed and together constitute the attitude solution model, as shown in formula (14).

[0069] (14)

[0070] in, , and These are the vectors of differences between the calculated and observed values ​​of baselines ab, ac, and ad, respectively. , and These are the attitude angle update parameter design matrices for baselines ab, ac, and ad, respectively. , and Design matrices for the ambiguity parameters of baselines ab, ac, and ad, respectively. , and These are the first-band ambiguity parameters for baselines ab, ac, and ad, respectively.

[0071] Step G2 involves using the least squares method to calculate the attitude solution model and obtain the floating-point solution of the carrier phase ambiguity in the first frequency band and the variance-covariance matrix of the carrier phase ambiguity in the first frequency band.

[0072] The aforementioned precise attitude determination method based on GNSS multi-antenna multi-frequency ambiguity stepwise resolution, in step H, uses the LAMBDA algorithm to fix the carrier phase ambiguity of the first frequency band based on the floating-point solution of the carrier phase ambiguity of the first frequency band and the variance-covariance matrix of the carrier phase ambiguity of the first frequency band, and obtains the carrier attitude angle correction. Then, the carrier attitude angle correction is combined with the approximate carrier attitude angle to obtain the true carrier attitude angle, as shown in formula (15).

[0073] (15).

[0074] A precise attitude determination system based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution includes a coordinate calculation module, an observation equation construction module, a baseline vector calculation module, a vehicle approximate attitude angle calculation module, an approximate attitude rotation matrix establishment module, a true attitude rotation matrix decomposition module, an attitude resolution model construction module, and a vehicle true attitude angle output module. The coordinate calculation module calculates the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observations, and then calculates the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna. The observation equation construction module constructs the double-difference pseudorange and carrier phase observation equations between the reference antenna and each auxiliary antenna on the vehicle based on GNSS multi-antenna multi-frequency observation data. The baseline vector calculation module constructs the wide-lane carrier phase observation equation based on the double-difference pseudorange and carrier phase observation equations and obtains the double-difference wide-lane ambiguity of each baseline, then fixes the double-difference wide-lane ambiguity of each baseline and obtains the baseline vector of each baseline in the ECEF coordinate system. The vehicle approximate attitude angle calculation module transforms the baseline vectors of each baseline in the ECEF coordinate system to the navigation coordinate system using the rotation matrix from the ECEF coordinate system to the navigation coordinate system, thereby obtaining the true attitude angles of each baseline. The coordinates of the baselines in the navigation coordinate system are used to calculate the approximate attitude angles of the vehicle. The approximate attitude rotation matrix establishment module is used to construct an approximate attitude rotation matrix from the approximate vehicle coordinate system to the navigation coordinate system using the approximate attitude angles of the vehicle. The real attitude rotation matrix decomposition module is used to construct a real attitude rotation matrix from the real vehicle coordinate system to the navigation coordinate system and decompose the real rotation matrix into an approximate attitude rotation matrix and an attitude update rotation matrix. Then, the attitude update rotation matrix is ​​decomposed to obtain the attitude decomposition result. The attitude solution model construction module uses... An attitude calculation model is constructed based on the attitude decomposition results. The least squares method is then used to calculate the attitude calculation model to obtain the floating-point solution of the first frequency band carrier phase ambiguity and the variance-covariance matrix of the first frequency band carrier phase ambiguity. The carrier true attitude angle output module is used to fix the first frequency band carrier phase ambiguity based on the floating-point solution of the first frequency band carrier phase ambiguity and the variance-covariance matrix of the first frequency band carrier phase ambiguity using the LAMBDA algorithm and obtain the carrier attitude angle correction. The carrier attitude angle correction is then combined with the carrier approximate attitude angle to obtain the carrier true attitude angle.

[0075] The beneficial effects of this invention are as follows: The accurate attitude determination method and system based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution first calculates the coordinates of the reference antenna in the ECEF coordinate system using reference antenna observations, then calculates the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna, then constructs the double-difference pseudorange and carrier phase observation equations between the reference antenna and each auxiliary antenna on the carrier based on GNSS multi-antenna multi-frequency observation data, then constructs the wide-lane carrier phase observation equations based on the double-difference pseudorange and carrier phase observation equations and obtains the double-difference wide-lane ambiguity of each baseline, then fixes the double-difference wide-lane ambiguity of each baseline and obtains the baseline vector of each baseline in the ECEF coordinate system, then transforms the baseline vector of each baseline in the ECEF coordinate system to the navigation coordinate system based on the rotation matrix from the ECEF coordinate system to the navigation coordinate system to obtain the coordinates of each baseline in the navigation coordinate system, and finally uses the coordinates of each baseline in the navigation coordinate system to calculate... The approximate attitude angle of the carrier is calculated. Then, an approximate attitude rotation matrix from the approximate carrier coordinate system to the navigation coordinate system is constructed using the approximate attitude angle. Simultaneously, a real attitude rotation matrix is ​​constructed from the real carrier coordinate system to the navigation coordinate system, and the real rotation matrix is ​​decomposed into an approximate attitude rotation matrix and an attitude update rotation matrix. The attitude update rotation matrix is ​​then decomposed to obtain the attitude decomposition result. Subsequently, an attitude solution model is constructed based on the attitude decomposition result. The attitude solution model is then calculated using the least squares method to obtain the floating-point solution of the first frequency band carrier phase ambiguity and the variance-covariance matrix of the first frequency band carrier phase ambiguity. Finally, based on the floating-point solution of the first frequency band carrier phase ambiguity and the variance-covariance matrix of the first frequency band carrier phase ambiguity, the LAMBDA algorithm is used to fix the first frequency band carrier phase ambiguity and obtain the carrier attitude angle correction. The carrier attitude angle correction is then combined with the approximate carrier attitude angle to obtain the real carrier attitude angle.This invention effectively realizes that the precise attitude determination method and system have the function of performing high-precision attitude determination by solving GNSS multi-antenna multi-frequency ambiguities step by step and using a combination of direct and least squares methods. Furthermore, it improves the solution efficiency of GNSS multi-frequency attitude determination by utilizing the advantage that the ambiguities of multi-frequency combined observations are easy to fix. By constructing a wide-lane carrier phase observation equation based on GNSS multi-antenna multi-frequency observation data and fixing the wide-lane ambiguities, the baseline vectors of each baseline in the ECEF coordinate system can be calculated. Then, each baseline vector is transformed from the ECEF coordinate system to the navigation coordinate system, and the baseline vectors are used to solve the carrier phase ambiguity problem. The approximate attitude angles of the body allow us to obtain the rotation matrix from the navigation coordinate system to the approximate vehicle coordinate system. Then, based on the rotation matrix between the navigation and approximate vehicle coordinate systems, the vectors of each baseline in the vehicle coordinate system, the rotation matrix from the ECEF coordinate system to the navigation coordinate system, and the GNSS observations from each antenna, an attitude calculation model can be constructed under the constraint of wide-lane observations with fixed ambiguity, using the attitude angle correction and ambiguity as unknowns. Finally, by using LAMBDA to fix the first frequency ambiguity, accurate attitude angle corrections can be obtained. This not only makes reasonable use of multi-frequency GNSS observations but also improves the accuracy of attitude calculation. Attached Figure Description

[0076] Figure 1 This is an overall flowchart of the accurate attitude determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution according to the present invention;

[0077] Figure 2 This is a schematic diagram illustrating the operating principle of the precise attitude determination method system based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution, as presented in this invention.

[0078] Figure 3 This is a schematic diagram of the arrangement of four antennas in the GNSS of this invention;

[0079] Figure 4 This is a satellite image of the experimental site in an embodiment of the present invention;

[0080] Figure 5 This is a comparison diagram of the attitude determination results in the embodiments of the present invention. Detailed Implementation

[0081] The present invention will now be further described with reference to the accompanying drawings.

[0082] like Figure 1 As shown, the accurate attitude determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution of the present invention includes the following steps:

[0083] Step A involves calculating the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observations, and then calculating the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna. The specific steps are as follows.

[0084] Step A1: Calculate the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observations. The antenna of the GNSS receiver on the carrier includes one reference antenna and three auxiliary antennas. Specifically, the coordinates of the reference antenna in the ECEF coordinate system are calculated based on the reference antenna observations using real-time dynamic differential RTK positioning, single-point positioning, precise single-point positioning and INS combined positioning.

[0085] Step A2: Calculate the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna, as shown in formula (1).

[0086] (1)

[0087] in, This is the rotation matrix from the ECEF coordinate system to the navigation coordinate system. The latitude of the reference antenna. The longitude of the reference antenna.

[0088] Step B involves constructing the double-difference pseudorange and carrier phase observation equations between the reference antenna and each auxiliary antenna on the carrier based on GNSS multi-antenna multi-frequency observation data, as shown in formula (2).

[0089] (2)

[0090] in, As a reference antenna, As an auxiliary antenna, For the station star double difference operator, The pseudorange observation values ​​for the nth frequency band are those of the reference antenna a and the auxiliary antenna i. The nth frequency band carrier phase observation values ​​of reference antenna a and auxiliary antenna i are given. The single-difference line-of-sight matrix of the reference antenna. This is a diagonal matrix whose diagonal values ​​are the wavelengths of the nth frequency band. It is a zero matrix. Baseline Baseline vector coordinates in the ECEF coordinate system The double-difference ambiguity of the nth frequency band between the reference antenna a and the auxiliary antenna i. This is to account for observation noise error.

[0091] Step C: Construct the wide-lane carrier phase observation equation based on the double-difference pseudorange and carrier phase observation equations, and obtain the double-difference wide-lane ambiguity of each baseline. Then, fix the double-difference wide-lane ambiguity of each baseline and obtain the baseline vector of each baseline in the ECEF coordinate system. The specific steps are as follows:

[0092] Step C1: Construct the wide-lane carrier phase observation equation based on the double-difference pseudorange and carrier phase observation equation, and obtain the double-difference wide-lane ambiguity of each baseline, as shown in formula (3).

[0093] (3)

[0094] in, Baseline Double-difference wide-lane carrier phase observations of the combination of frequency band 1 and frequency band n. This is a diagonal matrix whose diagonal values ​​are the combined wavelengths of the first and nth frequency bands. Baseline Wide-lane carrier phase ambiguity in combination of the first frequency band and the nth frequency band;

[0095] Step C2 involves fixing the double-difference wide-lane ambiguity of each baseline and obtaining the baseline vector of each baseline in the ECEF coordinate system. Specifically, this is achieved by using the wide-lane carrier phase observation equation combined with the elevation angle weighting model and the least squares method to obtain the floating-point solution of the double-difference wide-lane ambiguity of each baseline and the variance-covariance matrix of the double-difference wide-lane ambiguity of each baseline. Then, the LAMBDA algorithm or Bootstrapping algorithm is used to fix the double-difference wide-lane ambiguity of each baseline and obtain the baseline vector of each baseline in the ECEF coordinate system. .

[0096] Step D involves transforming the baseline vectors of each baseline in the ECEF coordinate system to the navigation coordinate system using the rotation matrix from the ECEF coordinate system to the navigation coordinate system, thereby obtaining the coordinates of each baseline in the navigation coordinate system. Then, the approximate attitude angles of the vehicle are calculated using the coordinates of each baseline in the navigation coordinate system. The specific steps are as follows.

[0097] Step D1: Based on the rotation matrix from the ECEF coordinate system to the navigation coordinate system, transform the baseline vectors of each baseline in the ECEF coordinate system to the navigation coordinate system to obtain the coordinates of each baseline in the navigation coordinate system, as shown in formula (4).

[0098] (4)

[0099] in, This represents the baseline vector of each baseline in the navigation coordinate system;

[0100] Step D2: Calculate the approximate attitude angles of the vehicle using the coordinates of each baseline in the navigation coordinate system, wherein the approximate attitude angles of the vehicle include the approximate heading angles of the vehicle. Carrier general pitch angle and the approximate roll angle of the carrier The specific steps are as follows:

[0101] Step D21, based on a baseline parallel to the longitudinal axis of the carrier. Calculate the approximate heading angle of the vehicle. and approximate pitch angle of the carrier Specifically, as shown in formulas (5) and (6) respectively,

[0102] (5)

[0103] (6)

[0104] in, Baseline The baseline vector in the navigation coordinate system;

[0105] Step D22, using baseline Combined with approximate heading angle and approximate pitch angle Calculate the approximate roll angle of the carrier. Specifically, as shown in formula (7),

[0106] (7)

[0107] in, Baseline The baseline vector in the navigation coordinate system.

[0108] Step E: Construct the approximate attitude rotation matrix from the approximate carrier coordinate system to the navigation coordinate system using the approximate carrier attitude angle, as shown in formula (8).

[0109] (8)

[0110] in, This is a rough attitude rotation matrix from the approximate vehicle coordinate system to the navigation coordinate system. For approximate attitude coordinate system, For cosine, It is the sine.

[0111] Step F involves constructing a true attitude rotation matrix from the real vehicle coordinate system to the navigation coordinate system, and then decomposing the true rotation matrix into a preliminary attitude rotation matrix and an attitude update rotation matrix. The attitude update rotation matrix is ​​then further decomposed to obtain the attitude decomposition result. The specific steps are as follows:

[0112] Step F1: Construct the true attitude rotation matrix from the real vehicle coordinate system to the navigation coordinate system, and decompose the true rotation matrix into a preliminary attitude rotation matrix and an attitude update rotation matrix, as shown in formula (9).

[0113] (9)

[0114] in, This is the true attitude rotation matrix from the real vehicle coordinate system to the navigation coordinate system. This is the transformation matrix from the real carrier coordinate system to the approximate carrier coordinate system. For the true pitch angle of the carrier Approximate pitch angle of the carrier difference, For the true roll angle of the carrier Approximate roll angle of the carrier difference, For the carrier's true heading angle Approximate heading angle of the vehicle difference;

[0115] Step F2 involves decomposing the pose update rotation matrix and obtaining the pose decomposition result, as shown in formula (10).

[0116] ;

[0117] (10)

[0118] in, The difference between the actual posture of the carrier and its approximate posture. It is a 3×3 identity matrix. It is a differential rotation matrix.

[0119] As shown in Figure 3, in step G, an attitude calculation model is constructed based on the attitude decomposition results. Then, the least squares method is used to calculate the attitude calculation model to obtain the floating-point solution of the carrier phase ambiguity in the first frequency band and the variance-covariance matrix of the carrier phase ambiguity in the first frequency band. The specific steps are as follows.

[0120] Step G1: Construct an attitude calculation model based on the attitude decomposition results. The specific steps are as follows.

[0121] Step G11, establish baseline The attitude calculation model is shown in Equation (11).

[0122] , ;

[0123] ;

[0124] ;

[0125] (11)

[0126] in, These are the pseudorange observations for all frequency bands of the reference antenna a and the auxiliary antenna i. Baseline The double-difference wide-lane carrier phase observations for the first frequency band and all frequency band combinations, where E is the double-difference operator matrix. For combined wavelengths, It is a column vector, specifically n×1. , Use the ECEF coordinate system;

[0127] Step G12, for the baseline The attitude calculation model is parameterized and a baseline is obtained. The parameterized attitude calculation model is shown in Equation (12).

[0128] , , , ;

[0129] (12)

[0130] in, Baseline Double difference observations For the line-of-sight matrix, Baseline The ambiguity parameter design matrix, For wavelength matrix, It is a column vector;

[0131] Step G13, based on the baseline Establishing a baseline for the parameterized attitude solution model The attitude error equation is shown in equation (13).

[0132] ;

[0133] ;

[0134] (13)

[0135] in, Baseline The difference vector between the calculated value and the observed value. Baseline The design matrix for attitude angle update parameters, Update parameters for attitude angles;

[0136] Step G14, based on baseline The attitude error equations of baseline ab, baseline ac and baseline ad are constructed and together constitute the attitude solution model, as shown in formula (14).

[0137] (14)

[0138] in, , and These are the vectors of differences between the calculated and observed values ​​of baselines ab, ac, and ad, respectively. , and These are the attitude angle update parameter design matrices for baselines ab, ac, and ad, respectively. , and Design matrices for the ambiguity parameters of baselines ab, ac, and ad, respectively. , and These are the first-band ambiguity parameters for baselines ab, ac, and ad, respectively.

[0139] Step G2 involves using the least squares method to calculate the attitude solution model and obtain the floating-point solution of the carrier phase ambiguity in the first frequency band and the variance-covariance matrix of the carrier phase ambiguity in the first frequency band.

[0140] Step H: Based on the floating-point solution of the carrier phase ambiguity of the first frequency band and the variance-covariance matrix of the carrier phase ambiguity of the first frequency band, the LAMBDA algorithm is used to fix the carrier phase ambiguity of the first frequency band and obtain the carrier attitude angle correction. Then, the carrier attitude angle correction is combined with the carrier approximate attitude angle to obtain the carrier true attitude angle, as shown in formula (15).

[0141] (15).

[0142] like Figure 2As shown, a precise attitude determination system based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution includes a coordinate calculation module, an observation equation construction module, a baseline vector calculation module, a carrier approximate attitude angle calculation module, an approximate attitude rotation matrix establishment module, a true attitude rotation matrix decomposition module, an attitude resolution model construction module, and a carrier true attitude angle output module. The coordinate calculation module calculates the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observations, and then calculates the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna. The observation equation construction module constructs the double-difference pseudorange and carrier phase observation equations between the reference antenna and each auxiliary antenna on the carrier based on GNSS multi-antenna multi-frequency observation data. The baseline vector calculation module constructs the wide-lane carrier phase observation equation based on the double-difference pseudorange and carrier phase observation equations and obtains the double-difference wide-lane ambiguity of each baseline, then fixes the double-difference wide-lane ambiguity of each baseline and obtains the baseline vector of each baseline in the ECEF coordinate system. The carrier approximate attitude angle calculation module transforms the baseline vectors of each baseline in the ECEF coordinate system to the navigation coordinate system based on the rotation matrix from the ECEF coordinate system to the navigation coordinate system, thereby obtaining the true attitude angle. The coordinates of each baseline in the navigation coordinate system are obtained, and then the approximate attitude angle of the vehicle is calculated using the coordinates of each baseline in the navigation coordinate system. The approximate attitude rotation matrix establishment module is used to construct an approximate attitude rotation matrix from the approximate vehicle coordinate system to the navigation coordinate system using the approximate attitude angle of the vehicle. The real attitude rotation matrix decomposition module is used to construct a real attitude rotation matrix from the real vehicle coordinate system to the navigation coordinate system and decompose the real rotation matrix into an approximate attitude rotation matrix and an attitude update rotation matrix. Then, the attitude update rotation matrix is ​​decomposed to obtain the attitude decomposition result. The attitude solution model construction module... The attitude decomposition module is used to construct an attitude calculation model based on the attitude decomposition results. Then, the least squares method is used to calculate the attitude calculation model to obtain the floating-point solution of the first frequency band carrier phase ambiguity and the variance-covariance matrix of the first frequency band carrier phase ambiguity. The carrier true attitude angle output module is used to fix the first frequency band carrier phase ambiguity based on the floating-point solution of the first frequency band carrier phase ambiguity and the variance-covariance matrix of the first frequency band carrier phase ambiguity using the LAMBDA algorithm and obtain the carrier attitude angle correction. Then, the carrier attitude angle correction is combined with the carrier approximate attitude angle to obtain the carrier true attitude angle.

[0143] To better illustrate the effectiveness of the present invention, an embodiment of using the precise attitude determination method of the present invention to determine the attitude of measured data is described below.

[0144] The satellite image of the experimental site in this embodiment is as follows: Figure 4As shown in the diagram; during the experiment, the CUTC0 station was used as the reference antenna a, and auxiliary antennas b, c, and d were CUTB0, CUT00, and CUTA0, respectively, all of which were installed on the roof of the building. The carrier's Y-axis points from CUTC0 to CUTB0, and the Y-axis, together with the CUTC0-CUT00 baseline, forms the XOY plane of the carrier's coordinate system. The X-axis is perpendicular to the carrier's Y-axis and points to the right, while the Z-axis points to the sky and is perpendicular to the XOY plane.

[0145] like Figure 5 The figure shows a comparison of attitude determination results using the attitude determination method of this invention and existing attitude determination methods. The three sub-figures, from top to bottom, represent the yaw angle, pitch angle, and roll angle, respectively. Blue represents the attitude determination result using the attitude determination method of this invention on the measured data, and red represents the attitude determination result using the direct attitude determination method on the measured data. It can be seen that the direct attitude determination method has more waveform glitches and poorer stability and attitude determination accuracy, as well as epochs with estimation errors and poor reliability. The attitude determination stability and accuracy of the attitude determination method of this invention are significantly improved compared to existing attitude determination methods.

[0146] In summary, the accurate attitude determination method and system based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution of the present invention first calculates the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observations, then calculates the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna. Next, based on the GNSS multi-antenna multi-frequency observation data, it constructs the double-difference pseudorange and carrier phase observation equations between the reference antenna and each auxiliary antenna on the carrier. Subsequently, based on the double-difference pseudorange and carrier phase observation equations, it constructs the wide-lane carrier phase observation equations and obtains the double-difference wide-lane ambiguity of each baseline. Then, it fixes the double-difference wide-lane ambiguity of each baseline and obtains the baseline vector of each baseline in the ECEF coordinate system. Then, based on the rotation matrix from the ECEF coordinate system to the navigation coordinate system, it transforms the baseline vectors of each baseline in the ECEF coordinate system to the navigation coordinate system to obtain the coordinates of each baseline in the navigation coordinate system. Finally, it uses the coordinates of each baseline in the navigation coordinate system to calculate the carrier's... The approximate attitude angle of the carrier is first determined, and then an approximate attitude rotation matrix from the approximate carrier coordinate system to the navigation coordinate system is constructed using the approximate carrier attitude angle. Simultaneously, a real attitude rotation matrix is ​​constructed from the real carrier coordinate system to the navigation coordinate system, and the real rotation matrix is ​​decomposed into an approximate attitude rotation matrix and an attitude update rotation matrix. The attitude update rotation matrix is ​​then decomposed to obtain the attitude decomposition result. Subsequently, an attitude calculation model is constructed based on the attitude decomposition result. The attitude calculation model is then calculated using the least squares method to obtain the floating-point solution of the first frequency band carrier phase ambiguity and the variance-covariance matrix of the first frequency band carrier phase ambiguity. Finally, based on the floating-point solution of the first frequency band carrier phase ambiguity and the variance-covariance matrix of the first frequency band carrier phase ambiguity, the LAMBDA algorithm is used to fix the first frequency band carrier phase ambiguity and obtain the carrier attitude angle correction. The carrier attitude angle correction is then combined with the approximate carrier attitude angle to obtain the real carrier attitude angle.This invention effectively realizes that the precise attitude determination method and system have the function of performing high-precision attitude determination by solving GNSS multi-antenna multi-frequency ambiguities step by step and using a combination of direct and least squares methods. Furthermore, it improves the solution efficiency of GNSS multi-frequency attitude determination by utilizing the advantage that the ambiguities of multi-frequency combined observations are easy to fix. By constructing a wide-lane carrier phase observation equation based on GNSS multi-antenna multi-frequency observation data and fixing the wide-lane ambiguities, the baseline vectors of each baseline in the ECEF coordinate system can be calculated. Then, each baseline vector is transformed from the ECEF coordinate system to the navigation coordinate system, and the baseline vectors are used to solve the carrier phase ambiguity problem. The approximate attitude angles of the body allow us to obtain the rotation matrix from the navigation coordinate system to the approximate vehicle coordinate system. Then, based on the rotation matrix between the navigation and approximate vehicle coordinate systems, the vectors of each baseline in the vehicle coordinate system, the rotation matrix from the ECEF coordinate system to the navigation coordinate system, and the GNSS observations from each antenna, an attitude calculation model can be constructed under the constraint of wide-lane observations with fixed ambiguity, using the attitude angle correction and ambiguity as unknowns. Finally, by using LAMBDA to fix the first frequency ambiguity, accurate attitude angle corrections can be obtained. This not only makes reasonable use of multi-frequency GNSS observations but also improves the accuracy of attitude calculation.

[0147] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1.A precise positioning method based on GNSS multi-antenna multi-frequency ambiguity step-by-step solution, characterized in that: The method comprises the following steps, Step A, calculating the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observation values, and then calculating the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna; Step B, constructing the double-difference pseudo-range and carrier phase observation equations between the reference antenna and each auxiliary antenna on the carrier according to the GNSS multi-antenna multi-frequency observation data; Step C, constructing the wide-lane carrier phase observation equation according to the double-difference pseudo-range and carrier phase observation equation, obtaining the double-difference wide-lane ambiguity of each baseline, fixing the double-difference wide-lane ambiguity of each baseline, and obtaining the baseline vector of each baseline in the ECEF coordinate system; Step D, converting the baseline vector of each baseline in the ECEF coordinate system to the navigation coordinate system according to the rotation matrix from the ECEF coordinate system to the navigation coordinate system, thereby obtaining the coordinates of each baseline in the navigation coordinate system, and then calculating the carrier rough attitude angle using the coordinates of each baseline in the navigation coordinate system; Step E, constructing the rough carrier coordinate system to the navigation coordinate system using the carrier rough attitude angle to obtain the rough attitude rotation matrix; Step F, constructing the true attitude rotation matrix from the true carrier coordinate system to the navigation coordinate system, decomposing the true rotation matrix into the rough attitude rotation matrix and the attitude update rotation matrix, and then decomposing the attitude update rotation matrix to obtain the attitude decomposition result; Step G, constructing an attitude solution model based on the attitude decomposition result, and then calculating the first frequency band carrier phase ambiguity float solution and the first frequency band carrier phase ambiguity variance-covariance matrix by using the least squares method on the attitude solution model; Step H, fixing the first frequency band carrier phase ambiguity by using the LAMBDA algorithm based on the first frequency band carrier phase ambiguity float solution and the first frequency band carrier phase ambiguity variance-covariance matrix, obtaining the carrier attitude angle correction, and then combining the carrier attitude angle correction with the carrier rough attitude angle to obtain the true carrier attitude angle. 2.The precise positioning method based on GNSS multi-antenna and multi-frequency ambiguity step-by-step solution according to claim 1, characterized in that: Step A, calculating the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observation values, and then calculating the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna, the specific steps are as follows, Step A1, calculating the coordinates of the reference antenna in the ECEF coordinate system using the reference antenna observation values, wherein the antennas of the GNSS receiver on the carrier include one reference antenna and three auxiliary antennas, and the coordinates of the reference antenna in the ECEF coordinate system are calculated by using the real-time kinematic differential RTK positioning, single point positioning, precise point positioning, and INS combined positioning according to the reference antenna observation values; Step A2, calculating the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna, which is specifically shown in formula (1), （1） wherein, is a rotation matrix from ECEF to navigation coordinate system, is a latitude of the reference antenna, is a longitude of the reference antenna. 3.The precise pose determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution according to claim 2, characterized in that: Step B, constructing the double-difference pseudo-range and carrier phase observation equations between the reference antenna and each auxiliary antenna on the carrier according to the GNSS multi-antenna multi-frequency observation data, which is specifically shown in formula (2), （2） wherein, is a reference antenna, is an auxiliary antenna, is a station-star double-difference operator, is the nth frequency band pseudorange observation value of the reference antenna a and the auxiliary antenna i, is the nth frequency band carrier phase observation value of the reference antenna a and the auxiliary antenna i, is the single-difference line-of-sight matrix of the reference antenna, is a diagonal matrix with the nth frequency band wavelength as the diagonal value, is a zero matrix, is a baseline is the baseline vector coordinate in the ECEF coordinate system, is the nth frequency band double-difference ambiguity of the reference antenna a and the auxiliary antenna i, is an observation noise error. 4.The precise positioning method based on GNSS multi-antenna multi-frequency ambiguity step-by-step solution according to claim 3, characterized in that: Step C, constructing the wide-lane carrier phase observation equation according to the double-difference pseudo-range and carrier phase observation equation, obtaining the double-difference wide-lane ambiguity of each baseline, fixing the double-difference wide-lane ambiguity of each baseline, and obtaining the baseline vector of each baseline in the ECEF coordinate system, the specific steps are as follows, Step C1, a wide-lane carrier phase observation equation is constructed according to a double-difference pseudo-range and carrier phase observation equation, and double-difference wide-lane ambiguities of each baseline are obtained, and the specific formula is shown in formula (3), （3） wherein, is the baseline a double-difference wide-lane carrier-phase observation of the first frequency band combined with the nth frequency band, is a diagonal matrix with diagonal values of the wavelength of the first frequency band combined with the nth frequency band, is the baseline a wide-lane carrier-phase ambiguity of the first frequency band combined with the nth frequency band; Step C2, fixing double-difference wide-lane ambiguities of each baseline and obtaining baseline vectors of each baseline in the ECEF coordinate system, specifically, using the wide-lane carrier phase observation equation combined with the elevation angle weighting model to obtain the float solution of double-difference wide-lane ambiguities of each baseline and the variance-covariance matrix of double-difference wide-lane ambiguities of each baseline by the least squares method, and then using the LAMBDA algorithm or the Bootstrapping algorithm to fix double-difference wide-lane ambiguities of each baseline and obtain the baseline vectors of each baseline in the ECEF coordinate system . 5.The precise pose determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution according to claim 4, characterized in that: Step D, the baseline vector of each baseline in the ECEF coordinate system is converted into the navigation coordinate system according to the rotation matrix from the ECEF coordinate system to the navigation coordinate system, so as to obtain the coordinates of each baseline in the navigation coordinate system, and then the carrier rough attitude angle is calculated according to the coordinates of each baseline in the navigation coordinate system, and the specific steps are as follows, Step D1, the baseline vector of each baseline in the ECEF coordinate system is converted into the navigation coordinate system according to the rotation matrix from the ECEF coordinate system to the navigation coordinate system, so as to obtain the coordinates of each baseline in the navigation coordinate system, and the specific formula is shown in formula (4), （4） wherein is a baseline vector of each baseline in the navigation coordinate system; Step D2, computing a carrier coarse attitude angle using the coordinates of each baseline in the navigation coordinate system, wherein the carrier coarse attitude angle comprises a carrier coarse heading angle , a carrier coarse pitch angle , and a carrier coarse roll angle , in particular as follows, Step D21, according to a baseline parallel to the longitudinal direction of the carrier calculating a carrier coarse heading angle and a carrier coarse pitch angle as shown in Equations (5) and (6), respectively, （5） （6） wherein baseline baseline vector in the navigation coordinate system; Step D22, utilize baseline combined with the coarse heading angle and the coarse pitch angle to calculate the coarse cross-vehicle roll angle as shown in equation (7), （7） wherein Baseline Baseline vector in the navigation coordinate system. 6.The precise pose determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution according to claim 5, characterized in that: Step E, a rough attitude rotation matrix from the rough carrier coordinate system to the navigation coordinate system is constructed according to the carrier rough attitude angle, and the specific formula is shown in formula (8), （8） wherein is a coarse pose rotation matrix of the coarse body coordinate system to the navigation coordinate system, is a coarse pose coordinate system, is a cosine, is a sine. 7.The precise pose determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution according to claim 6, characterized in that: Step F, a true attitude rotation matrix is constructed according to the true carrier coordinate system to the navigation coordinate system, and the true rotation matrix is decomposed into a rough attitude rotation matrix and an attitude update rotation matrix, then the attitude update rotation matrix is decomposed to obtain the attitude decomposition result, and the specific steps are as follows, Step F1, a true attitude rotation matrix is constructed according to the true carrier coordinate system to the navigation coordinate system, and the true rotation matrix is decomposed into a rough attitude rotation matrix and an attitude update rotation matrix, and the specific formula is shown in formula (9), （9） wherein, is a real pose rotation matrix from the real carrier coordinate system to the navigation coordinate system, is a conversion matrix from the real carrier coordinate system to the sketchy carrier coordinate system, is a real pitch angle of the carrier and a sketchy pitch angle of the carrier is a difference between, is a real roll angle of the carrier and a sketchy roll angle of the carrier is a difference between, is a real heading angle of the carrier and a sketchy heading angle of the carrier is a difference between. Step F2, the attitude update rotation matrix is decomposed to obtain the attitude decomposition result, and the specific formula is shown in formula (10), ； （10） wherein, is the difference between the carrier true attitude and the carrier sketchy attitude, is a 3x3 identity matrix, superscript is a differential rotation matrix. 8.The precise pose determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution according to claim 7, characterized in that: Step G, an attitude solving model is constructed based on the attitude decomposition result, and then the least square method is used to calculate the attitude solving model to obtain the first frequency band carrier phase ambiguity float solution and the first frequency band carrier phase ambiguity variance-covariance matrix, and the specific steps are as follows, Step G1, an attitude solving model is constructed based on the attitude decomposition result, and the specific steps are as follows, Step G11, constructing baseline of the pose solution model, as shown in equation (11), ， ； ； ； （11） wherein, is the pseudo-range observation of the reference antenna a and the auxiliary antenna i in all frequency bands, is the baseline is the double-difference wide-lane carrier phase observation of the first frequency band and all frequency band combination, E is the double-difference operator matrix, is the combined wavelength, is a column vector, specifically an n x 1 , is the ECEF coordinate system; Step G12, parameterize the pose-solution model of baseline and obtain the parameterized pose-solution model of baseline , as shown in equation (12), ， ， ， ； （12） wherein, is a baseline of double-difference observations, is a line-of-sight matrix, is a baseline of ambiguity parameter design matrix, is a wavelength matrix, is a column vector; Step G13, establish the baseline of the parameterized attitude solution model according to the parameters of the baseline of the attitude error equation, specifically as shown in equation (13), ； ； （13） wherein is the baseline a difference vector of the calculated values minus the observed values, is the baseline a design matrix of the attitude angle update parameters, is the attitude angle update parameter; Step G14, constructing the attitude error equation based on the baseline The attitude error equations of the baselines ab, ac and ad are constructed based on the attitude error equation of the baseline and jointly constitute the attitude solution model, as shown in formula (14). （14） wherein, , and are the difference vectors between the calculated and observed values of the baseline ab, baseline ac and baseline ad respectively, , and are the attitude angle update parameter design matrices of the baseline ab, baseline ac and baseline ad respectively, , and are the ambiguity parameter design matrices of the baseline ab, baseline ac and baseline ad respectively, , and are the 1st band ambiguity parameters of the baseline ab, baseline ac and baseline ad respectively. Step G2, the least square method is used to calculate the attitude solving model to obtain the first frequency band carrier phase ambiguity float solution and the first frequency band carrier phase ambiguity variance-covariance matrix. 9.The precise pose determination method based on GNSS multi-antenna multi-frequency ambiguity step-by-step resolution according to claim 8, characterized in that: Step H, the LAMBDA algorithm is used to fix the first frequency band carrier phase ambiguity based on the first frequency band carrier phase ambiguity float solution and the first frequency band carrier phase ambiguity variance-covariance matrix to obtain the carrier attitude angle correction, and then the carrier attitude angle correction is combined with the carrier rough attitude angle to obtain the true carrier attitude angle, and the specific formula is shown in formula (15), （15）。 10. A precise positioning system based on GNSS multi-antenna multi-frequency ambiguity step-by-step solution, the precise positioning system specific positioning process is based on the precise positioning method of any one of claims 1-9, characterized in that: The system comprises a coordinate calculation module, an observation equation construction module, a baseline vector calculation module, a carrier rough attitude angle calculation module, a rough attitude rotation matrix establishment module, a true attitude rotation matrix decomposition module, an attitude solving model construction module and a carrier true attitude angle output module, the coordinate calculation module is used to calculate the coordinates of the reference antenna in the ECEF coordinate system by using the reference antenna observation value, and then the rotation matrix from the ECEF coordinate system to the navigation coordinate system at the reference antenna is calculated; The observation equation construction module is used to construct the double-difference pseudo-range and carrier phase observation equation between the reference antenna and each auxiliary antenna on the carrier according to the GNSS multi-antenna multi-frequency observation data; The baseline vector calculation module is used to calculate the baseline vector of each baseline in the ECEF coordinate system according to the GNSS multi-antenna multi-frequency observation data; The baseline vector calculation module is configured to construct a wide-lane carrier phase observation equation according to a double-difference pseudo-range and carrier phase observation equation, and obtain double-difference wide-lane ambiguities of each baseline, and then fix the double-difference wide-lane ambiguities of each baseline and obtain a baseline vector of each baseline in an ECEF coordinate system; The carrier rough attitude angle calculation module is configured to convert the baseline vector of each baseline in the ECEF coordinate system to a navigation coordinate system according to a rotation matrix from the ECEF coordinate system to the navigation coordinate system, so as to obtain coordinates of each baseline in the navigation coordinate system, and then calculate a carrier rough attitude angle by using the coordinates of each baseline in the navigation coordinate system; The rough attitude rotation matrix establishment module is configured to construct a rough attitude rotation matrix from a rough carrier coordinate system to the navigation coordinate system by using the carrier rough attitude angle; The real attitude rotation matrix decomposition module is configured to construct a real attitude rotation matrix from a real carrier coordinate system to the navigation coordinate system, and decompose the real rotation matrix into the rough attitude rotation matrix and an attitude update rotation matrix, and then decompose the attitude update rotation matrix and obtain an attitude decomposition result; The attitude solution model construction module is configured to construct an attitude solution model based on the attitude decomposition result, and then calculate the attitude solution model by using a least square method, so as to obtain a first frequency band carrier phase ambiguity float solution and a first frequency band carrier phase ambiguity variance-covariance matrix; The carrier real attitude angle output module is configured to fix the first frequency band carrier phase ambiguity by using a LAMBDA algorithm based on the first frequency band carrier phase ambiguity float solution and the first frequency band carrier phase ambiguity variance-covariance matrix, so as to obtain a carrier attitude angle correction, and then combine the carrier attitude angle correction with the carrier rough attitude angle, so as to obtain a carrier real attitude angle.

Images