GNSS (Global Navigation Satellite System) non-combined double-difference multi-system multi-frequency dynamic base line real-time resolving method and system

Abstract

The invention relates to the technical field of satellite navigation positioning processing, in particular to a GNSS (Global Navigation Satellite System) non-combined double-difference multi-system multi-frequency dynamic base line real-time resolving method and system, and the method comprises the steps: carrying out the preprocessing of GNSS original observed quantities of two dynamic observation stations; a non-combined double-difference observation model is constructed, and the model forms double-difference observation quantity in the same satellite system based on non-combined carrier and pseudo code observation data; carrying out parameter estimation by adopting extended Kalman filter (EKF), wherein the parameter estimation comprises time updating and observation updating; converting the estimated single-difference ambiguity into double-difference ambiguity, and fixing the ambiguity; and outputting a moving-moving baseline resolving result which comprises a high-precision relative position and a relative speed. According to the invention, through non-combined double-difference multi-system multi-frequency data processing, high-precision real-time calculation of a dynamic base line and a dynamic base line is realized, and the measurement precision of the relative position and speed between dynamic carriers is remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of satellite navigation and positioning processing, in particular to a GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solving method and system. BACKGROUND

[0002] In the traditional GNSS real-time relative positioning, the position of the reference station is usually known in advance and fixed, and the mobile station obtains the reference station data through the data communication link, so as to realize dynamic position solving. However, in special application occasions such as ship formation navigation, unmanned aerial vehicle formation flight, shipborne aircraft landing, aircraft in-flight refueling, etc., the absolute position information is not the focus of attention, and the high-precision real-time relative position between dynamic carriers is more important. Under such conditions, there is no concept of reference station, and both of the two stations involved will be in motion, and GNSS dynamic baseline real-time solving is usually required. In fact, dynamic baseline real-time solving as a special mode of GNSS data processing has been concerned by researchers since the early stage of satellite navigation development. Initially, dynamic baseline solving was mainly applied to GNSS attitude determination, and the dynamic baseline formed by three non-collinear antennas was solved to realize the attitude measurement of the moving carrier. Under this condition, the baseline length is fixed and can be used as a constraint to speed up the ambiguity convergence.

[0003] In recent years, with the continuous development of Beidou, Galileo and other systems, the multi-system multi-frequency era has quietly arrived. In the research of precise point positioning, traditional GNSS real-time relative positioning, etc., multi-system multi-frequency processing has become a research hotspot and gradually become the main direction of technical development. However, in dynamic baseline real-time solving, how to fully exert the advantages of multi-system multi-frequency data is still less involved in related research. If the ionosphere-free double-difference model commonly used in traditional baseline processing is directly used, it will involve complex data combination theory, and with the increase of multi-system multi-frequency data, the corresponding optimal combination method will also change, which is not conducive to real-time solving processing. In addition, dynamic baseline real-time solving has its own characteristics, and sufficient consideration is also needed in multi-system multi-frequency data processing. SUMMARY

[0004] In order to solve the problem that the prior art is difficult to fully exert the advantages of multi-system multi-frequency data in GNSS dynamic baseline real-time solving, and the traditional ionosphere-free double-difference model is not conducive to GNSS dynamic baseline real-time solving due to complex combination, the present application provides a GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solving method, which can conveniently integrate multi-system multi-frequency data, weaken errors, facilitate ambiguity fixing, realize GNSS dynamic baseline high-precision real-time solving, and support ship formation and other scene applications.

[0005] To achieve the above object, the technical scheme adopted is:

[0006] The application provides a GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solving method, which comprises the following steps:

[0007] The GNSS original observation of two dynamic stations is preprocessed;

[0008] A non-combination double-difference observation model is constructed, which is based on non-combination carrier and pseudo-code observation data and forms double-difference observation in the same satellite system;

[0009] Extended Kalman filter (EKF) is used for parameter estimation, including time update and observation update;

[0010] The estimated single-difference ambiguity is converted into double-difference ambiguity, and the ambiguity is fixed;

[0011] The dynamic baseline solving result is output, including high-precision relative position and relative velocity.

[0012] According to the GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solving method, further, the GNSS original observation comprises multi-system multi-frequency carrier observation data and pseudo-code observation data, and the multi-system comprises a Beidou system and a Galileo system; the preprocessing comprises format unification, time system conversion and space coordinate system unification.

[0013] According to the GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solving method, further, the non-combination double-difference observation model comprises a double-difference carrier observation model and a double-difference pseudo-code observation model, and the double-difference carrier observation model expression is:

[0014]

[0015] The double-difference pseudo-code observation model expression is:

[0016]

[0017] Wherein, m1 and m2 are the numbers of two dynamic stations respectively; i and j are the numbers of two GNSS satellites respectively; g is the signal frequency number; Respectively represent dynamic baseline double-difference carrier and double-difference pseudo-code observation; λ g Represents the wavelength of g frequency carrier; Represents the single-difference ambiguity between stations; λ1 represents the carrier wavelength of the first frequency signal; Respectively represent the single-difference ionospheric delay between stations in the vertical direction of the first frequency signal of i satellite and j satellite; Respectively represent the projection function in the direction of i satellite and j satellite; denotes the double-difference geometric range between stations; ε Φ ,ε P denote the double-difference carrier and double-difference pseudorange observation noise, respectively.

[0018] According to the GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solution method, further, the expression of the to-be-estimated parameter vector of the parameter estimation is:

[0019]

[0020] wherein, are the position and velocity vectors of the m2 station; denotes the first frequency signal vertical ionospheric delay inter-station single-difference vector, and n denotes the number of visible satellites at the current epoch; is a single-difference ambiguity vector, and l denotes the number of frequencies of multi-frequency data, and the element of the vector is denotes the single-difference ambiguity of h frequency carriers observed by two dynamic stations m1 and m2 on p satellites.

[0021] According to the GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solution method, further, the specific process of the time update includes:

[0022] First, the predicted value vector of the to-be-estimated parameter at the current time is calculated according to the uniform motion model, and the formula is:

[0023]

[0024] wherein, denotes the to-be-estimated parameter estimation value vector at the t k-1 time; denotes the to-be-estimated parameter predicted value vector at the t k time; denotes a state transition matrix;

[0025] Then, the covariance matrix of the to-be-estimated parameter predicted value at the current time is calculated, and the formula is:

[0026]

[0027] wherein, denotes the covariance matrix of the to-be-estimated parameter estimation value vector at the t k-1 time; denotes the covariance matrix of the to-be-estimated parameter predicted value vector at the t k time; denotes a state noise matrix.

[0028] According to the GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solution method, further, the state transition matrix The expression is:

[0029]

[0030] Wherein, t k ,t k-1 Respectively represent the current observation time and the last epoch observation time; L represents the frequency number of multi-frequency data; N represents the number of visible satellites in the current epoch; I is a unit matrix, and 0 is a zero matrix.

[0031] According to the GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solving method, further, the specific process of the observation update comprises:

[0032] Firstly, the gain matrix is calculated, and the formula is:

[0033]

[0034] Wherein, K k Indicates the gain matrix, Indicates the design matrix, W k Is the weight inverse matrix of double-difference observation;

[0035] Then, the estimated parameter value vector at the current time is calculated, and the formula is:

[0036]

[0037] Wherein, Indicates the estimated parameter value vector at t k , y k Indicates the observation vector formed by all double-difference carrier waves and double-difference pseudorandom codes at t k , and Indicates the observation function.

[0038] Finally, the covariance matrix of the estimated parameter value vector at the current time is calculated, and the formula is:

[0039]

[0040] Wherein, Indicates the covariance matrix of the estimated parameter value vector at t k , and I is a unit matrix.

[0041] According to the GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solving method, further, the ambiguity fixing specifically comprises:

[0042] Firstly, the single-difference ambiguity float solution is converted into a double-difference ambiguity float solution;

[0043] Then, the LAMBDA method is used to solve the integer least square problem, and the double-difference ambiguity integer solution is obtained.

[0044] Finally, the remaining state parameters are corrected according to the double-difference ambiguity integer solution.

[0045] According to the GNSS non-combination double-difference multi-system multi-frequency kinematic baseline real-time solution method, further, the single-difference ambiguity float solution is converted into the double-difference ambiguity float solution through the conversion matrix G: respectively represent the t k instant parameter estimation value vector and the covariance matrix thereof, P k respectively represent the double-difference float solution and the covariance matrix thereof after conversion; wherein the expression of the conversion matrix G is:

[0046]

[0047] wherein, I (6+n)×(6+n) is a unit matrix; the expression of D is:

[0048]

[0049] Further, the application further provides a GNSS non-combination double-difference multi-system multi-frequency kinematic baseline real-time solution system for realizing the above-mentioned GNSS non-combination double-difference multi-system multi-frequency kinematic baseline real-time solution method, comprising:

[0050] A preprocessing module is configured to preprocess GNSS original observation data of two dynamic stations.

[0051] An observation model construction module is configured to construct a non-combination double-difference observation model, which is based on non-combination carrier and pseudo-code observation data and forms double-difference observation data within the same satellite system.

[0052] A parameter estimation module is configured to perform parameter estimation by using an extended Kalman filter (EKF), including time update and observation update.

[0053] An ambiguity fixing module is configured to convert the estimated single-difference ambiguity into double-difference ambiguity and perform ambiguity fixing.

[0054] A result output module is configured to output kinematic baseline solution results, including high-precision relative position and relative velocity.

[0055] By using the above technical solution, the following beneficial effects are achieved:

[0056] 1. Adapt to multi-system multi-frequency data, improve solution flexibility and robustness

[0057] The non-combination double difference observation model is adopted, and original multi-system (such as Beidou, Galileo) multi-frequency carrier and pseudo-code observation data are directly used, without complex ionosphere elimination combination, and the optimal combination method adaptation problem caused by system or frequency increase is avoided. Meanwhile, through double difference processing, common errors such as satellite clock error and receiver clock error are effectively weakened, the redundancy advantage of multi-system multi-frequency data is fully utilized, and the robustness of solving in a dynamic environment is improved.

[0058] 2. Precise modeling error, improve parameter estimation accuracy

[0059] The ionospheric delay error is considered in the model and is taken as an estimated parameter, avoiding the over-simplification of ionospheric error in the traditional ionosphere elimination combination, and being especially suitable for the scene in the active ionosphere area. Combined with the design of inter-station single difference ambiguity, the dimension of ambiguity parameter is reduced, laying a foundation for subsequent ambiguity fixing, and then improving the estimation accuracy of position, velocity and other core parameters.

[0060] 3. Optimizing real-time solving process to meet the timeliness requirement in dynamic scenes

[0061] Based on the extended Kalman filter (EKF) time update and observation update mechanism of uniform motion model, recursive real-time solving of the estimated parameters is realized. The integer least square problem of double difference ambiguity is solved efficiently by LAMBDA method, the ambiguity fixing time is shortened, and the real-time requirement of the whole solving process in dynamic baseline scenes such as ship formation and unmanned aerial vehicle cooperation is ensured.

[0062] 4. Strong universality, expand application scenarios

[0063] The application does not depend on specific GNSS system or frequency configuration, and is compatible with existing and newly added satellite navigation systems and signal frequency bands, and is suitable for relative positioning scenes between various dynamic stations (such as intelligent transportation, precise surveying and mapping, aerospace, etc.), and has wide engineering application value. BRIEF DESCRIPTION OF DRAWINGS

[0064] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings of the embodiments of the application will be briefly introduced below. The drawings are only used to show some embodiments of the application, and the application is not limited to the drawings.

[0065] Figure 1 is a flowchart of the GNSS non-combination double difference multi-system multi-frequency dynamic baseline real-time solving method of the embodiments of the application. DETAILED DESCRIPTION

[0066] The following clearly and completely describes example schemes of embodiments of the present application with reference to the accompanying drawings of specific embodiments of the present application. Unless otherwise defined, technical terms or scientific terms used in the present application should be understood as having their usual meanings understood by those having ordinary skills in the art.

[0067] Embodiments of the present application disclose a GNSS non-combination double-difference multi-system multi-frequency moving baseline real-time solving method, a method flow is as shown in Figure 1 The method specifically includes the following steps:

[0068] Step S101, pre-processing GNSS original observation data of two dynamic stations, the GNSS original observation data including multi-system multi-frequency carrier observation data and pseudo-code observation data, the multi-system including a Beidou system and a Galileo system, etc. The pre-processing includes format unification, time system conversion and space coordinate system unification.

[0069] Step S102, constructing a non-combination double-difference observation model, the model being based on non-combination carrier and pseudo-code observation data, and forming double-difference observation data in the same satellite system.

[0070] To facilitate multi-system multi-frequency data fusion processing and avoid a complex data combination process, here, non-combination carrier and pseudo-code observation data are used, at the same time, the advantages of moving baseline solving are fully given, double differences are formed to weaken various error influences, and ambiguity fixing is facilitated. Considering that both stations are in motion, to avoid frequent replacement of reference stars and influence solving effect caused by constantly changing estimated parameters, inter-station single-difference ambiguity is selected as an estimated parameter, and after parameter estimation, double-difference ambiguity is re-formed for fixing processing. It should be noted that the moving baseline discussed in the present scheme is usually a medium-short baseline, under this condition, most errors can be basically ignored after double difference, but considering that the moving baseline may have a large dynamic range, and ionospheric delay error itself also changes relatively complexly, to improve solving precision, ionospheric delay error is considered in parameter estimation. In summary, the non-combination double-difference observation model includes a double-difference carrier observation model and a double-difference pseudo-code observation model.

[0071] The double-difference carrier observation model expression is:

[0072]

[0073] The double-difference pseudo-code observation model expression is:

[0074]

[0075] Wherein, m1 and m2 are respectively numbers of two dynamic stations; i and j are respectively numbers of two GNSS satellites; g is a signal frequency number. respectively, denote the double-difference carrier and code observations, and it is noted that the double-difference observations are only formed between the same satellite system to reduce the complexity of processing when processing multi-system data. g denotes the wavelength of the g frequency carrier; denotes the single-difference ambiguity between stations; denotes the wavelength of the first frequency signal; denotes the single-difference ionospheric delay in the vertical direction between the first frequency signals of the i satellite and the j satellite; denotes the projection function in the direction of the i satellite and the j satellite, respectively; denotes the double-difference geometric distance between stations and satellites; Φ ,ε P denote the double-difference carrier and code observation noise, respectively.

[0076] Step S103, parameter estimation is performed using the extended Kalman filter (EKF), including time update and observation update.

[0077] In order to facilitate real-time calculation, the EKF is used for parameter estimation. Considering that the relative position relationship between two stations is concerned in the dynamic baseline solution, one station (such as m1) is fixed as the single-point positioning result, and the position and velocity of the other station (such as m2) are taken as the parameters to be estimated, so that the relative baseline solution can be reflected by the estimated position and velocity of the station m2. In this case, the expression of the parameter estimation parameter vector to be estimated is:

[0078]

[0079] wherein, denote the position and velocity vectors of the station m2, respectively; denotes the single-difference ionospheric delay vector in the vertical direction of the first frequency signal, and n denotes the number of visible satellites at the current epoch; denotes the single-difference ambiguity vector, and l denotes the number of frequencies of multi-frequency data, and the element denotes the single-difference ambiguity on the h frequency carrier observed by the two dynamic stations m1 and m2 on the p satellite.

[0080] (1) Time update

[0081] The time update calculates the predicted value vector of the parameter to be estimated at the current time using the uniform motion model, and the formula is:

[0082]

[0083] wherein, denotes the estimated value vector of the parameter to be estimated at the time t k-1 , and denotes the predicted value vector of the parameter to be estimated at the time t k , denotes the state transition matrix.

[0084] The covariance matrix of the predicted value of the parameter to be estimated at the current time is calculated, and the formula is:

[0085]

[0086] wherein, denotes the covariance matrix of the estimated value of the parameter to be estimated at the time t k-1 ; denotes the covariance matrix of the predicted value of the parameter to be estimated at the time t k , denotes the state noise matrix.

[0087] The expression of the state transition matrix is:

[0088]

[0089] wherein, t k , t k-1 respectively denote the current observation time and the last epoch observation time; l denotes the number of frequencies of the multi-frequency data; n denotes the number of visible satellites at the current epoch; I is a unit matrix, and 0 is a zero matrix.

[0090] (2) Observation update

[0091] Based on the observation model formula (1) and (2), the observation update is performed, and the model is as follows:

[0092] First, the gain matrix is calculated, and the formula is:

[0093]

[0094] wherein, K k denotes the gain matrix, denotes the design matrix, such as formula (13); W k is the weight inverse matrix of the double-difference observation.

[0095] Then, the estimated value vector of the parameter to be estimated at the current time is calculated, and the formula is:

[0096]

[0097] wherein, denotes the estimated value vector of the parameter to be estimated at the time t k ; denotes the observation vector composed of all double-difference carriers and double-difference codes at the time t k , k denotes the observation function, such as formula (10).

[0098] The covariance matrix of the estimated parameter vector at the current time is calculated, and the formula is:

[0099]

[0100] wherein, denotes the covariance matrix of the estimated parameter vector at the time t k , and I is the unit matrix.

[0101] h(x) = (…, h Φ,g T ,…, h P,g T ,) (10)

[0102] In the formula, h Φ,g h P,g respectively represent the double-difference carrier and double-difference code observation functions corresponding to the gth frequency, and the definitions are shown in formulas (11) and (12).

[0103]

[0104] In the formula, the definitions of the symbols are shown in formula (1).

[0105]

[0106] In the formula, the definitions of the symbols are shown in formula (2).

[0107]

[0108] In the formula, γ1, γ2, …, γ l and λ1, λ2, …, λ l have the same definitions as formulas (1) and (2); D, E, M I have the definitions shown in formulas (14), (15) and (16).

[0109]

[0110] In the formula, denotes the direction cosine vector between the qth satellite and the mobile station m2.

[0111]

[0112] In the formula, denotes the projection function of the qth satellite.

[0113] In step S104, the estimated single-difference ambiguity is converted into a double-difference ambiguity, and the ambiguity is fixed.

[0114] By the above parameter estimation process, the floating point solution of the to-be-estimated parameters can be obtained. In order to further improve the estimation accuracy and accelerate the convergence time, the single-difference ambiguity obtained by estimation is converted into double-difference ambiguity, and the ambiguity is fixed, in the following way:

[0115] Firstly, the estimated parameters are converted into double-difference mode:

[0116]

[0117] wherein, respectively represent the to-be-estimated parameter estimation vector and its covariance matrix at time t k ; the floating point solution obtained by the parameter estimation in step S103; P k and P N respectively represent the double-difference floating point solution and its covariance matrix after conversion; represents the double-difference ambiguity floating point solution; the other to-be-estimated parameters have the same meaning as in formula (3); Q R , Q RN , Q NR respectively represent the variance matrix of the double-difference ambiguity floating point solution and other to-be-estimated parameters, and the covariance matrix between them; G is the conversion matrix, which is defined as follows:

[0118]

[0119] wherein, I (6+n)×(6+n) is an identity matrix, and the meaning of D is shown in formula (14).

[0120] Therefore, we have:

[0121]

[0122] wherein, represents the double-difference ambiguity floating point solution, Q N represents the covariance matrix of the double-difference ambiguity floating point solution, represents the double-difference ambiguity integer solution, and N∈Z represents that N is an integer.

[0123] According to the LAMBDA method, the integer least square problem shown in formula (20) is solved to obtain the double-difference ambiguity integer solution. Then, the fixed solution of other parameters is obtained by formula (21):

[0124]

[0125] wherein, respectively represent the position, velocity vector of m2 stations and the floating point solution of the ionospheric delay inter-station single-difference vector, which are obtained by estimation according to formula (8).

[0126] Step S105, outputting the dynamic baseline solution result, including high-precision relative position, relative velocity, ionospheric delay difference and double-difference ambiguity integer solution.

[0127] Corresponding to the above method, the embodiment of the application also discloses a GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solution system, which comprises:

[0128] A preprocessing module is configured to preprocess GNSS original observation data of two dynamic stations.

[0129] An observation model construction module is configured to construct a non-combination double-difference observation model, which is based on non-combination carrier and pseudo-code observation data and forms double-difference observation data in the same satellite system.

[0130] A parameter estimation module is configured to perform parameter estimation by using an extended Kalman filter (EKF), including time update and observation update.

[0131] An ambiguity fixing module is configured to convert the estimated single-difference ambiguity into double-difference ambiguity and fix the ambiguity.

[0132] A result output module is configured to output the dynamic baseline solution result, including high-precision relative position and relative velocity.

[0133] Unless otherwise specified, the components, steps, numerical expressions and values described in the embodiments do not limit the scope of the application.

[0134] The embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method part.

[0135] The units and method steps of each example described in combination with the embodiments disclosed in the present application can be realized by electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example are generally described in the above description. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation does not exceed the scope of the application.

[0136] Those skilled in the art can understand that all or part of the steps in the foregoing method can be instructed by programs to the related hardware, and the programs can be stored in a computer readable storage medium, such as a read-only memory, a magnetic disk or an optical disk. Alternatively, all or part of the steps of the foregoing embodiments can also be implemented using one or more integrated circuits, and accordingly, each module / unit in the foregoing embodiments can be implemented in the form of hardware or in the form of a software functional module. The present application is not limited to any specific form of combination of hardware and software.

[0137] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present application, which are used to illustrate the technical solutions of the present application, rather than limit the same. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can make modifications to the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some of the technical features within the technical scope disclosed by the present application. Such modifications or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A real-time solution method for GNSS non-combined double-difference multi-system multi-frequency kinematic baseline, characterized in that, The method comprises the following steps: preprocessing GNSS raw observations of two dynamic stations; constructing a non-combination double-difference observation model based on non-combination carrier and pseudo-code observation data to form double-difference observations within the same satellite system; performing parameter estimation by using extended Kalman filtering (EKF), including time update and observation update; transforming estimated single-difference ambiguities into double-difference ambiguities and fixing the ambiguities; outputting dynamic baseline solution results, including high-precision relative position and relative velocity.

2. The GNSS non-combined double difference multi-system multi-frequency kinematic baseline real-time solution method according to claim 1, characterized in that, The GNSS raw observations include multi-system and multi-frequency carrier observation data and pseudo-code observation data, and the multi-system includes the Beidou system and the Galileo system; the preprocessing includes format unification, time system conversion and spatial coordinate system unification.

3. The GNSS non-combined double difference multi-system multi-frequency kinematic baseline real-time solution method according to claim 1, characterized in that, The non-combination double-difference observation model includes a double-difference carrier observation model and a double-difference pseudo-code observation model, and the double-difference carrier observation model has the expression: The double-difference pseudo-code observation model has the expression: where m1, m2 are the indices of the two dynamic stations respectively; i, j are the indices of the two GNSS satellites respectively; g is the signal frequency index; respectively represent the double-difference carrier and double-difference code observations of the baseline between the two dynamic stations; λ g represents the wavelength of the g frequency carrier; represents the single-difference ambiguity between the two stations; λ1 represents the wavelength of the first frequency signal; respectively represent the single-difference ionospheric delay in the vertical direction for the first frequency signal of the i satellite and the j satellite; respectively represent the projection function in the direction of the i satellite and the j satellite; represents the double-difference geometric distance between the station and the satellite; ε Φ ,ε P respectively represent the double-difference carrier and double-difference code observation noise.

4. The GNSS non-combined double difference multi-system multi-frequency kinematic baseline real-time solution method according to claim 1, characterized in that, The expression of the parameter estimation to-be-estimated parameter vector is: wherein, are the position and velocity vector of the m2 station respectively; represents the single-difference vector of the vertical ionospheric delay of the first frequency signal between stations, and n represents the number of visible satellites at the current epoch; is the single-difference ambiguity vector, and l represents the number of frequencies of the multi-frequency data, and the element represents the single-difference ambiguity of the h frequency carrier observed by two dynamic stations m1 and m2 on p satellites.

5. The GNSS non-combined double difference multi-system multi-frequency kinematic baseline real-time solution method according to claim 1, characterized in that, The specific process of the time update includes: firstly, calculating a to-be-estimated parameter prediction value vector at the current time according to a uniform motion model, and the formula is: wherein represents t k-1 the parameter estimate vector at the time instant to be estimated; represents t k the parameter forecast vector at the time instant to be estimated; represents the state transition matrix; then, calculating a covariance matrix of the to-be-estimated parameter prediction value at the current time, and the formula is: wherein represents t k-1 the covariance matrix of the parameter estimate vector to be estimated at time t; represents t k the covariance matrix of the parameter prediction vector to be estimated at time t; represents the state noise matrix.

6. The GNSS non-combined double difference multi-system multi-frequency kinematic baseline real-time solution method according to claim 5, characterized in that, The state transition matrix The expression is: where t k , t k-1 represent the current observation time and the last epoch observation time, respectively; l represents the number of frequencies of multi-frequency data; n represents the number of visible satellites at the current epoch; I is an identity matrix, and 0 is a zero matrix.

7. The GNSS non-combined double difference multi-system multi-frequency kinematic baseline real-time solution method according to claim 5, characterized in that, The specific process of the observation update includes: firstly, calculating a gain matrix, and the formula is: where K k represents a gain matrix, represents a design matrix, W k is a weight inverse matrix of double difference observations; then, calculating a to-be-estimated parameter estimation value vector at the current time, and the formula is: wherein, denotes t k the parameter estimate vector to be estimated at time t, y k denotes t k the observation vector consisting of all double-difference code and double-difference carrier at time t, denotes the observation function; finally, calculating a covariance matrix of the to-be-estimated parameter estimation value vector at the current time, and the formula is: wherein denotes t k the covariance matrix of the parameter estimate vector to be estimated at time instant t, I being the identity matrix.

8. The GNSS non-combined double difference multi-system multi-frequency kinematic baseline real-time solution method according to claim 1, characterized in that, The ambiguity fixing specifically includes: firstly, transforming single-difference ambiguity float solutions into double-difference ambiguity float solutions; then, solving an integer least square problem by using the LAMBDA method to obtain double-difference ambiguity integer solutions; finally, correcting the remaining state parameters according to the double-difference ambiguity integer solutions.

9. The GNSS non-combined double difference multi-system multi-frequency kinematic baseline real-time solution method according to claim 8, characterized in that, The single-difference ambiguity float solution is converted into a double-difference ambiguity float solution by a conversion matrix G: respectively represent the t k instant parameter estimation value vector and its covariance matrix, P′ k respectively represent the converted double-difference float solution and its covariance matrix; wherein the expression of the conversion matrix G is: where I (6+n)×(6+n) is the identity matrix; and D is given by: 10.A GNSS non-combined double-difference multi-system multi-frequency kinematic baseline real-time solution system, characterized in that, The method for implementing the GNSS non-combination double-difference multi-system multi-frequency dynamic baseline real-time solution method according to any one of claims 1-9 comprises: a preprocessing module configured to preprocess GNSS raw observations of two dynamic stations; an observation model construction module configured to construct a non-combination double-difference observation model based on non-combination carrier and pseudo-code observation data to form double-difference observations within the same satellite system; a parameter estimation module configured to perform parameter estimation by using extended Kalman filtering (EKF), including time update and observation update; an ambiguity fixing module configured to transform estimated single-difference ambiguities into double-difference ambiguities and fix the ambiguities; a result output module configured to output dynamic baseline solution results, including high-precision relative position and relative velocity.

Images