A GNSS dual antenna directional state constraint method and system
By employing Kalman filtering estimation and state constraint methods, the problem of carrier phase lockout under severe obstruction conditions is solved, improving heading stability and orientation accuracy while reducing algorithm complexity.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG ZHIYANG ELECTRIC
- Filing Date
- 2023-07-13
- Publication Date
- 2026-06-02
AI Technical Summary
In multipath environments with severe obstruction, carrier phase loss occurs frequently, making it difficult to fix ambiguity and obtain high-precision heading information. Existing methods are either too complex or rely on unstable floating-point solutions, which affects estimation accuracy.
Baseline state information is estimated using Kalman filtering. By calculating the baseline state gain and the corrected state variance-covariance matrix, state constraints are applied using high-precision baseline vector information, independent of the observation equation and the LAMBDA algorithm.
It improves the accuracy of floating-point ambiguity, stabilizes heading estimation, reduces algorithm complexity, enhances the success rate of ambiguity fixation, and improves orientation accuracy in multipath environments.
Smart Images
Figure CN116859432B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of satellite navigation technology for unmanned aerial vehicle (UAV) flight control systems, and specifically relates to a GNSS dual-antenna orientation state constraint method and system. Background Technology
[0002] Currently, dual-antenna directional technology can obtain high-precision and highly reliable baseline vector fixation solutions in open, unobstructed environments, thereby obtaining high-precision heading information. However, with rapid societal development, environments like urban canyons are becoming increasingly common. Mobile vehicles in these scenarios cause severe obstruction, and glass buildings can also lead to multipath interference with satellite signals. Therefore, in severely obstructed multipath scenarios, carrier phase loss occurs frequently, making ambiguity fixation and obtaining high-precision heading information difficult. Generally, appropriate constraint methods are needed to improve the success rate of fixation and optimize the heading accuracy obtained in harsh environments.
[0003] Existing methods incorporate baseline constraints into the LAMBDA technique, specifically a least-squares ambiguity decorrelation search algorithm based on additional baseline length constraints. This method is complex and relies on highly reliable floating-point ambiguity solutions. If the floating-point ambiguity accuracy is low, it directly affects the radius of the search ellipsoid, increasing its size. This not only impacts the success rate of ambiguity fixation but also reduces ambiguity search efficiency. Another approach is to integrate baseline constraints into the observation equations, performing filtering estimation together with double-difference carrier phase observations and double-difference pseudorange observations to obtain a floating-point solution for the baseline vector. However, this method has the drawback that large errors in the predicted baseline vector can affect the estimation accuracy of state parameters such as position and floating-point ambiguity. Summary of the Invention
[0004] To address the aforementioned technical problems, this invention proposes a GNSS dual-antenna orientation state constraint method and system. This can improve the orientation accuracy of dual antennas in multipath environments and enhance the heading stability of mobile vehicles in complex environments with severe obstruction.
[0005] To achieve the above objectives, the present invention adopts the following technical solution:
[0006] A GNSS dual-antenna directional state constraint method includes the following steps:
[0007] The state variance and covariance matrix is updated using the baseline state information estimated by Kalman filtering; when the Kalman filter converges, it is determined whether the baseline state information estimated by the Kalman filter is usable;
[0008] When baseline state information is available, calculate the state constraint gain of the dual antennas; based on the state constraint gain, correct the baseline state information estimated by Kalman filtering, and correct the state variance covariance matrix.
[0009] Furthermore, before updating the state variance covariance by estimating the baseline state information through Kalman filtering, the method also includes calculating the baseline length between the master and slave antennas estimated by Kalman filtering.
[0010] Furthermore, the process of calculating the baseline length between the master and slave antennas estimated by the Kalman filter includes:
[0011] Let the main antenna M and the slave antenna S be mounted on the moving carrier, with the line containing the main antenna M and the slave antenna S parallel to the direction of the carrier's movement; where the position of the main antenna M in the ECEF coordinate system is r. m =(x m ,y m ,z m ), where r m The position vector of the main antenna M in the ECEF coordinate system, x m y m z m These represent the coordinate components of the main antenna M in the ECEF coordinate system along the X, Y, and Z axes, respectively.
[0012] The position vector of antenna S in the ECEF coordinate system is These represent the estimated coordinate components of the antenna along the X, Y, and Z axes of the ECEF coordinate system, respectively; therefore, the carrier phase differential Kalman filter estimates the direction vector between the master and slave antennas.
[0013] Therefore, the baseline length between the master and slave antennas estimated by Kalman filtering is the predicted baseline length.
[0014] Furthermore, the process of estimating baseline state information and updating state variance and covariance using Kalman filtering is as follows:
[0015] When the corresponding measurement noise is ε, the estimation error Where l is the actual baseline length of the dual antennas;
[0016] The state variance covariance matrix obtained by carrier phase differential Kalman filtering estimation is represented in blocks as follows: in This represents a 3x3 position covariance matrix containing 3 state parameters; Represents a k×k velocity, acceleration, and floating-point ambiguity covariance matrix containing k state parameters; and This represents the correlation coefficient matrix between position and velocity, acceleration, and floating-point ambiguity.
[0017] Furthermore, the method for determining the convergence of the Kalman filter is as follows:
[0018]
[0019] Where K represents the scaling factor, T1 is the trace of the position covariance matrix; if T1≤0, the Kalman filter state has converged; if T1>0, it exits.
[0020] Furthermore, the method available for determining the baseline status information is as follows:
[0021] T2 = |v| - α;
[0022] 'a' represents the judgment threshold; if T2 < 0, it means that the baseline information error estimated by the Kalman filter meets the requirements; if T2 ≥ 0, then exit.
[0023] Furthermore, the process of calculating the state constraint gain of the dual antennas is as follows:
[0024] Based on the direction vector between the master and slave antennas Calculate the unit vector from the main antenna M to the secondary antenna S.
[0025] State constraint gain Constant value C = e·σ 1,1 2 ·e T +ε 2 G is a column vector of size (3+k)×1.
[0026] Furthermore, the process of estimating baseline state information via Kalman filtering based on the state constraint gain, and correcting the state variance covariance matrix, includes:
[0027] use Perform state vector correction; where The state vector from antenna S is obtained by carrier phase differential Kalman filtering estimation; it is expressed as: These represent the estimated position, velocity, and acceleration vectors from antenna S, respectively. This is the estimated floating-point ambiguity state vector; This is the corrected floating-point ambiguity state vector;
[0028] Calculate the corrected baseline state direction vector
[0029] Then, feedback correction is applied to the state variance-covariance matrix P1 to obtain the state constraint-corrected variance-covariance matrix P2: P2 = P1 - G·e·[σ 1,1 2 σ 1,2 2 ].
[0030] Furthermore, the modified state variance-covariance matrix further includes:
[0031] The corrected floating-point ambiguity state vector from antenna S The corrected variance P2 is applied to the fixation of LAMBDA integer ambiguity.
[0032] This invention also proposes a GNSS dual-antenna directional state constraint system, including a preprocessing module and a correction module;
[0033] The preprocessing module is used to update the state variance covariance matrix using the baseline state information estimated by Kalman filtering; when the Kalman filter converges, it determines whether the baseline state information estimated by the Kalman filter is available;
[0034] The correction module is used to calculate the state constraint gain of the dual antennas when the baseline state information is available; to correct the baseline state information estimated by Kalman filtering based on the state constraint gain; and to correct the state variance covariance matrix.
[0035] The effects described in the invention are merely those of the embodiments, and not all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects:
[0036] This invention proposes a GNSS dual-antenna orientation state constraint method and system. The method includes calculating the baseline length between the master and slave antennas estimated by Kalman filtering; updating the state variance-covariance matrix using the baseline state information estimated by Kalman filtering; determining the usability of the baseline state information estimated by Kalman filtering when the Kalman filtering converges; calculating the state constraint gain of the dual antennas when the baseline state information is usable; correcting the baseline state information estimated by Kalman filtering based on the state constraint gain; and correcting the state variance-covariance matrix. Based on this GNSS dual-antenna orientation state constraint method, a GNSS dual-antenna orientation state constraint system is also proposed. This invention uses an independent baseline length constraint algorithm based on posterior information estimated by Kalman filtering. This algorithm has low complexity and is flexible in use. It not only improves the stability of the estimated state parameters and avoids the influence of excessively erroneous baseline vector information on other state parameters, but also improves the accuracy of floating-point ambiguity, making floating-point ambiguity less prone to divergence and increasing the success rate of subsequent LAMBDA ambiguity fixing. In summary, this invention can improve the orientation accuracy of dual antennas in multipath environments and enhance the heading stability of mobile vehicles in complex environments with severe obstruction.
[0037] This invention treats baseline length constraint as an independent constraint algorithm. It is neither integrated into the observation equation for joint estimation nor incorporated into the LAMBDA algorithm. Instead, it establishes a separate constraint equation based on the post-hoc high-precision baseline vector information estimated by Kalman filtering, and performs feedback correction on the state parameters to achieve constraint on the state vector, making it more flexible to use. It avoids the influence of excessively erroneous baseline vector information on other state parameters. At the same time, it also constrains floating-point ambiguity, making it less prone to divergence and thus improving the accuracy of floating-point ambiguity. Attached Figure Description
[0038] like Figure 1 This is a flowchart of a GNSS dual-antenna directional state constraint method proposed in Embodiment 1 of the present invention;
[0039] like Figure 2 This is a schematic diagram of a GNSS dual-antenna directional state constraint system proposed in Embodiment 2 of the present invention. Detailed Implementation
[0040] To clearly illustrate the technical features of this solution, the invention will be described in detail below through specific embodiments and in conjunction with the accompanying drawings. The following disclosure provides many different embodiments or examples for implementing different structures of the invention. To simplify the disclosure of the invention, components and arrangements of specific examples are described below. Furthermore, reference numerals and / or letters may be repeated in different examples. This repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed. It should be noted that the components illustrated in the drawings are not necessarily drawn to scale. Descriptions of well-known components, processing techniques, and processes are omitted in this invention to avoid unnecessarily limiting the invention.
[0041] Example 1
[0042] The GNSS dual-antenna directional state constraint method proposed in Embodiment 1 of this invention adds an independent baseline constraint between Kalman filter state estimation and LAMBDA ambiguity fixation, indirectly constraining the Kalman filter state. First, the apologetic high-precision baseline information from the Kalman filter state estimation is used as the predicted baseline length, which serves as the input information for the state constraint. Second, the difference between the predicted baseline length and the actual baseline length is calculated as the error information for the state constraint. Then, the constraint gain is calculated based on the state variance covariance information and unit baseline vector obtained from the Kalman filter estimation. Finally, the state parameters and the corresponding state variance covariance matrix are corrected.
[0043] like Figure 1 This is a flowchart of a GNSS dual-antenna directional state constraint method proposed in Embodiment 1 of the present invention;
[0044] In step S100, the process begins;
[0045] In step S110, the baseline length between the master and slave antennas estimated by the Kalman filter is calculated.
[0046] Before updating the state variance and covariance by estimating the baseline state information using Kalman filtering, the baseline length between the master and slave antennas estimated by Kalman filtering is also calculated.
[0047] The process of calculating the baseline length between the master and slave antennas estimated by Kalman filtering includes:
[0048] Let the main antenna M and the slave antenna S be mounted on the moving carrier, with the line containing the main antenna M and the slave antenna S parallel to the direction of the carrier's movement; where the position of the main antenna M in the ECEF coordinate system is r. m =(x m ,y m ,z m ), where r m The position vector of the main antenna M in the ECEF coordinate system, x m y m z m These represent the coordinate components of the main antenna M in the ECEF coordinate system along the X, Y, and Z axes, respectively.
[0049] The position vector of antenna S in the ECEF coordinate system is These represent the estimated coordinate components of the antenna along the X, Y, and Z axes of the ECEF coordinate system, respectively; therefore, the carrier phase differential Kalman filter estimates the direction vector between the master and slave antennas.
[0050] Therefore, the baseline length between the master and slave antennas estimated by Kalman filtering is the predicted baseline length.
[0051] The state variance covariance matrix obtained by carrier phase differential Kalman filtering estimation is represented in blocks as follows: in This represents a 3x3 position covariance matrix containing 3 state parameters; Represents a k×k velocity, acceleration, and floating-point ambiguity covariance matrix containing k state parameters; and This represents the correlation coefficient matrix between position and velocity, acceleration, and floating-point ambiguity.
[0052] In step S120, when the corresponding measurement noise is ε, the estimated value error is... Where l is the actual baseline length of the dual antennas;
[0053] In step S130, it is determined whether the Kalman filter has converged. If it has not converged, step S170 is executed. If it has converged, step S140 is executed.
[0054] The method for determining the convergence of a Kalman filter is as follows:
[0055]
[0056] Where K represents the scaling factor, K>0, and needs to be set based on experience; T1 is the trace of the position covariance matrix; if T1≤0, it indicates that the Kalman filter state has converged, and the next step can be performed. If T1>0, the state constraint is exited, and no further operations are performed.
[0057] The scaling factor K is essentially the maximum allowable percentage of error in the location variance. For example, K = 0.5 means that the location variance needs to be considered convergent within 0.5 times the baseline length.
[0058] In step S140, it is determined whether the baseline status information is available. If it is not available, step S170 is executed. If it is available, step S150 is executed.
[0059] The methods available for determining baseline status information are:
[0060] T2 = |v| - α;
[0061] 'a' represents the judgment threshold, which needs to be set based on experience. If T2 < 0, it means that the baseline information error estimated by the Kalman filter meets the requirements, and the next step of processing can be carried out. If T2 ≥ 0, the state constraint is exited and no further operations are performed.
[0062] In step S150, the constraint state gain is calculated. The state constraint gain is determined by taking into account the Kalman filter state variance and baseline length measurement noise.
[0063] The process of calculating the state constraint gain of the two antennas is as follows:
[0064] Based on the direction vector between the master and slave antennas Calculate the unit vector from the main antenna M to the secondary antenna S.
[0065] State constraint gain Constant value C = e·σ 1,1 2 ·e T +ε 2 G is a column vector of size (3+k)×1.
[0066] In step S160, the baseline state information is estimated by Kalman filtering based on the state constraint gain correction, and the state variance covariance matrix is corrected.
[0067] Specifically, this includes: adopting Perform state vector correction; where The state vector from antenna S is obtained by carrier phase differential Kalman filtering estimation; it is expressed as: These represent the estimated position, velocity, and acceleration vectors from antenna S, respectively. This is the estimated floating-point ambiguity state vector; This is the corrected floating-point ambiguity state vector;
[0068] Calculate the corrected baseline state direction vector
[0069] Then, feedback correction is applied to the state variance-covariance matrix P1 to obtain the state constraint-corrected variance-covariance matrix P2: P2 = P1 - G·e·[σ 1,1 2 σ 1,2 2 ].
[0070] In step S170, the corrected floating-point ambiguity state vector from antenna S is... The corrected variance P2 is applied to the fixation of LAMBDA integer ambiguity.
[0071] The GNSS dual-antenna orientation state constraint method proposed in Embodiment 1 of this invention employs an independent baseline length constraint algorithm based on posterior information estimated by Kalman filtering. This algorithm boasts low complexity and flexible application, improving not only the stability of the estimated state parameters and preventing the influence of excessively erroneous baseline vector information on other state parameters, but also enhancing the accuracy of floating-point ambiguities, making them less prone to divergence and increasing the success rate of subsequent LAMBDA ambiguity fixing. In summary, this method improves the orientation accuracy of dual antennas in multipath environments and enhances the heading stability of mobile vehicles in complex environments with severe obstruction.
[0072] The GNSS dual-antenna directional state constraint method proposed in Embodiment 1 of this invention is an independent constraint algorithm. It is neither integrated into the observation equation for joint estimation nor incorporated into the LAMBDA algorithm. Instead, it is based on the post-hoc high-precision baseline vector information estimated by Kalman filtering, and establishes constraint equations separately to provide feedback correction for state parameters, thereby constraining the state vector. This method is more flexible in use and avoids the influence of excessively erroneous baseline vector information on other state parameters. It also constrains floating-point ambiguity, making it less prone to divergence and thus improving the accuracy of floating-point ambiguity.
[0073] Example 2
[0074] Based on the GNSS dual-antenna orientation state constraint method proposed in Embodiment 1 of this invention, Embodiment 2 of this invention also proposes a GNSS dual-antenna orientation state constraint system, such as... Figure 2 This is a schematic diagram of a GNSS dual-antenna directional state constraint system proposed in Embodiment 2 of the present invention. The system includes: a preprocessing module and a correction module.
[0075] The preprocessing module is used to update the state variance-covariance matrix using the baseline state information estimated by Kalman filtering; when the Kalman filter converges, it determines whether the baseline state information estimated by the Kalman filter is available.
[0076] The correction module is used to calculate the state constraint gain of the dual antennas when baseline state information is available; correct the baseline state information estimated by Kalman filtering based on the state constraint gain; and correct the state variance covariance matrix.
[0077] The preprocessing module performs the following steps:
[0078] Let the main antenna M and the slave antenna S be mounted on the moving carrier, with the line containing the main antenna M and the slave antenna S parallel to the direction of the carrier's movement; where the position of the main antenna M in the ECEF coordinate system is r. m =(x m ,y m ,z m ), where r m The position vector of the main antenna M in the ECEF coordinate system, x m y m z m These represent the coordinate components of the main antenna M in the ECEF coordinate system along the X, Y, and Z axes, respectively.
[0079] The position vector of antenna S in the ECEF coordinate system is These represent the estimated coordinate components of the antenna along the X, Y, and Z axes of the ECEF coordinate system, respectively; therefore, the carrier phase differential Kalman filter estimates the direction vector between the master and slave antennas.
[0080] Therefore, the baseline length between the master and slave antennas estimated by Kalman filtering is the predicted baseline length.
[0081] The state variance covariance matrix obtained by carrier phase differential Kalman filtering estimation is represented in blocks as follows: in This represents a 3x3 position covariance matrix containing 3 state parameters; Represents a k×k velocity, acceleration, and floating-point ambiguity covariance matrix containing k state parameters; and This represents the correlation coefficient matrix between position and velocity, acceleration, and floating-point ambiguity.
[0082] When the corresponding measurement noise is ε, the estimation error Where l is the actual baseline length of the dual antennas;
[0083] To determine whether a Kalman filter has converged, the following methods are used:
[0084]
[0085] Where K represents the scaling factor, K>0, and needs to be set based on experience; T1 is the trace of the position covariance matrix; if T1≤0, it indicates that the Kalman filter state has converged, and the next step can be performed. If T1>0, the state constraint is exited, and no further operations are performed.
[0086] The scaling factor K is essentially the maximum allowable percentage of error in the location variance. For example, K = 0.5 means that the location variance needs to be considered convergent within 0.5 times the baseline length.
[0087] The process of executing the correction module includes:
[0088] Determining whether baseline status information is available involves the following methods:
[0089] T2 = |v| - α;
[0090] 'a' represents the judgment threshold, which needs to be set based on experience. If T2 < 0, it means that the baseline information error estimated by the Kalman filter meets the requirements, and the next step of processing can be carried out. If T2 ≥ 0, the state constraint is exited and no further operations are performed.
[0091] The constraint state gain is calculated. The state constraint gain is determined by taking into account the Kalman filter state variance and baseline length measurement noise.
[0092] The process of calculating the state constraint gain of the two antennas is as follows:
[0093] Based on the direction vector between the master and slave antennas Calculate the unit vector from the main antenna M to the secondary antenna S.
[0094] State constraint gain Constant value C = e·σ 1,1 2 ·e T +ε 2 G is a column vector of size (3+k)×1.
[0095] The state constraint gain correction is based on the estimation of baseline state information through Kalman filtering, and the correction of the state variance covariance matrix.
[0096] Specifically, this includes: adopting Perform state vector correction; where The state vector from antenna S is obtained by carrier phase differential Kalman filtering estimation; it is expressed as: These represent the estimated position, velocity, and acceleration vectors from antenna S, respectively. This is the estimated floating-point ambiguity state vector; This is the corrected floating-point ambiguity state vector;
[0097] Calculate the corrected baseline state direction vector
[0098] Then, feedback correction is applied to the state variance-covariance matrix P1 to obtain the state constraint-corrected variance-covariance matrix P2: P2 = P1 - G·e·[σ 1,1 2 σ 1,2 2 ].
[0099] After the correction module completes its execution, it also includes: transferring the corrected floating-point ambiguity state vector from antenna S. The corrected variance P2 is applied to the fixation of LAMBDA integer ambiguity.
[0100] The GNSS dual-antenna orientation state constraint system proposed in Embodiment 2 of this invention employs an independent baseline length constraint algorithm based on posterior information estimated by Kalman filtering. This algorithm boasts low complexity and flexible application, improving not only the stability of the estimated state parameters and preventing the influence of excessively erroneous baseline vector information on other state parameters, but also enhancing the accuracy of floating-point ambiguities, making them less prone to divergence and increasing the success rate of subsequent LAMBDA ambiguity fixing. In summary, this system improves the orientation accuracy of dual antennas in multipath environments and enhances the heading stability of mobile vehicles in complex environments with severe obstruction.
[0101] The description of the relevant parts of the GNSS dual-antenna directional state constraint system provided in Embodiment 2 of this application can be found in the detailed description of the corresponding parts of the GNSS dual-antenna directional state constraint method provided in Embodiment 1 of this application, and will not be repeated here.
[0102] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that the elements inherent in a process, method, article, or apparatus that includes a list of elements are included. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. Additionally, portions of the technical solutions provided in the embodiments of this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.
[0103] While specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art can make other modifications or variations based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.
Claims
1. A GNSS dual-antenna directional state constraint method, characterized in that, Includes the following steps: The state variance and covariance matrix is updated using the baseline state information estimated by Kalman filtering; when the Kalman filter converges, it is determined whether the baseline state information estimated by the Kalman filter is usable; When baseline state information is available, calculate the state constraint gain of the dual antennas; correct the baseline state information estimated by Kalman filtering based on the state constraint gain, and correct the state variance covariance matrix. The process of calculating the state constraint gain of the dual antennas is as follows: Based on the direction vector between the master and slave antennas Calculate the main antenna From antenna unit vector ; State constraint gain ; with the corresponding measurement noise being At that time, constant value , for A column vector of size; Indicates the number of state parameters; Represents the position covariance matrix; A matrix representing the correlation coefficients between velocity, acceleration, floating-point ambiguity, and position; The process of correcting the baseline state information estimated by Kalman filtering based on the state constraint gain, and correcting the state variance covariance matrix, includes: use Perform state vector correction; where The carrier phase differential Kalman filter estimation obtained from the antenna The state vector; represented as: , , , They represent the estimated values from the antenna. Position, velocity, and acceleration vectors This is the estimated floating-point ambiguity state vector; This is the corrected floating-point ambiguity state vector; This indicates the error in the baseline length estimate; Calculate the corrected baseline state direction vector ; Main antenna Position vector in the ECEF coordinate system , , These represent the main antennas. Coordinate components of the X, Y, and Z axes in the ECEF coordinate system; And the state variance covariance matrix Feedback correction is performed to obtain the variance-covariance matrix after state constraint correction. : ; The modified state variance-covariance matrix is followed by: The corrected antenna floating-point ambiguity state vector and the corrected variance-covariance matrix It is applied to the fixing of integer ambiguity in LAMBDA.
2. The GNSS dual-antenna directional state constraint method according to claim 1, characterized in that, Before updating the state variance and covariance by estimating the baseline state information through Kalman filtering, the method also includes calculating the baseline length between the master and slave antennas estimated by Kalman filtering.
3. The GNSS dual-antenna directional state constraint method according to claim 2, characterized in that, The process of calculating the baseline length between the master and slave antennas estimated by the Kalman filter includes: Command the main antenna and from antenna Mounted on a mobile platform, main antenna and from antenna The line in question is parallel to the direction of the carrier's motion; among them, the main antenna... The position in the ECEF coordinate system is ,in, Main antenna Position vector in the ECEF coordinate system , , These represent the main antennas. Coordinate components of the X, Y, and Z axes in the ECEF coordinate system; From antenna The position vector in the ECEF coordinate system is ; Let X, Y, and Z represent the estimated coordinate components of the antenna in the ECEF coordinate system, respectively; therefore, the carrier phase differential Kalman filter estimates the direction vector between the master and slave antennas. ; Therefore, the baseline length between the master and slave antennas estimated by Kalman filtering is the predicted baseline length. .
4. The GNSS dual-antenna directional state constraint method according to claim 3, characterized in that, The process of estimating baseline state information and updating state variance and covariance using Kalman filtering is as follows: The corresponding measurement noise is At that time, the estimated value error ;in, This represents the actual baseline length of the dual antennas. The state variance covariance matrix obtained by carrier phase differential Kalman filtering estimation is represented in blocks as follows: ,in This represents a 3x3 position covariance matrix containing 3 state parameters; express The magnitude of the velocity, acceleration, and floating-point ambiguity covariance matrix contains One state parameter; A matrix representing the correlation coefficients between position and velocity, acceleration, and floating-point ambiguity; This represents the correlation coefficient matrix between velocity, acceleration, floating-point ambiguity, and position.
5. A GNSS dual-antenna directional state constraint method according to claim 4, characterized in that, The method for determining the convergence of a Kalman filter is as follows: ； in, Indicates the scaling factor. Let the trace be the location covariance matrix; if This indicates that the Kalman filter state has converged; if Then exit.
6. The GNSS dual-antenna directional state constraint method according to claim 4, characterized in that, The methods available for determining baseline status information are: ； Indicates the threshold for judgment; if This indicates that the baseline information error estimated by the Kalman filter meets the requirements; if Then exit.
7. A GNSS dual-antenna orientation state constraint system, used to execute the GNSS dual-antenna orientation state constraint method according to any one of claims 1 to 6, characterized in that, Includes a preprocessing module and a correction module; The preprocessing module is used to update the state variance covariance matrix using the baseline state information estimated by Kalman filtering; when the Kalman filter converges, it determines whether the baseline state information estimated by the Kalman filter is available; The correction module is used to calculate the state constraint gain of the dual antennas when the baseline state information is available; to correct the baseline state information estimated by Kalman filtering based on the state constraint gain; and to correct the state variance covariance matrix.