Real-time combination quality control method for GNSS accurate positioning in severe environment and control system thereof

Abstract

The invention belongs to the technical field of navigation satellites, and discloses a real-time combination quality control method and a real-time combination quality control system for GNSS (Global Navigation Satellite System) accurate positioning in a severe environment. The method comprises the following steps: 1, establishing a GNSS double-difference mathematical model; step 2, based on the GNSS double-difference mathematical model in the step 1, establishing an improved processing strategy for a multipath effect of phase and code observation; and step 3, based on the processing strategy improved in the step 2, abnormal value processing in a posterior stage is realized by using an improved DIA method considering the maximum number of data snooping times. According to a quality control method used in the prior art, the processing work of an abnormal value and a multipath effect is mainly in a posterior stage, and related processing cannot be carried out in advance, so that the difference value of GNSS accurate positioning in a severe environment is reduced.

Description

Technical Field

[0001] This invention belongs to the field of navigation satellite technology, specifically relating to a real-time combined quality control method and control system for GNSS precise positioning in harsh environments. Background Technology

[0002] The rapid development of Global Navigation Satellite Systems (GNSS) has brought many new opportunities and challenges to precise positioning technology. GNSS precise positioning technology is widely used in natural disaster monitoring, vehicle driving, and other fields. In practical applications, users may face harsh conditions, such as canyons and urban environments. However, in these harsh environments, relatively severe multipath effects and outliers often occur, which significantly reduce the accuracy and ambiguity resolution of GNSS precise positioning. Therefore, further emphasis should be placed on developing quality control methods for GNSS precise positioning in such harsh environments.

[0003] To date, there has been considerable research on quality control methods for precise GNSS positioning. Outliers and multipath errors are two major problems. Previous work on outlier handling can be broadly categorized into two types: robust estimation methods and outlier detection methods. Specifically, robust estimation methods primarily rely on equivalent weighting functions to detect and eliminate outliers in GNSS observations. However, the finite breakdown point of parameter estimation limits their practical application in harsh environments, especially under the relatively strong influence of outliers. For outlier detection methods, hypothesis testing theory is used to construct these methods based on assumptions about the type, quantity, and location of outliers. The widely used Detection, Identification, and Adaptation (DIA) method was first proposed by Bararda and further discussed by Teunissen, Zaminpardaz, and Teunissen. However, simply handling outliers cannot meet the practical application requirements in harsh environments, especially under relatively strong multipath effects. For handling multipath effects, hardware improvements can be made, such as antenna design and receiver enhancements, but this increases hardware costs. Alternatively, data processing methods can be applied. Several filtering methods, such as Empirical Mode Decomposition (EMD) and wavelet analysis, can mitigate multipath effects, but they are not suitable for real-time applications. Furthermore, some multipath mitigation methods have been developed based on the repeatability of multipath paths. The most popular methods include the classical stellar filtering method and the hemispherical mesh model correction method, which have been proven effective in mitigating multipath effects. However, none of these methods fully meet the requirements of real-time and kinematic applications.

[0004] Unfortunately, previous work on quality control methods (i.e., handling outliers and multipath effects) has primarily focused on the posterior stage. Specifically, robust estimation methods and outlier detection methods rely on posterior residual checks. Furthermore, filtering methods for multipath handling are employed in the post-processed solution, utilizing stellar filtering and hemispherical mesh models based on the posterior residuals. In harsh environments, due to more severe multipath effects and outliers, obtaining truly reliable posterior residuals may be difficult. However, there has been limited attention paid to real-time combined quality control methods that simultaneously consider both prior and posterior knowledge. Summary of the Invention

[0005] This invention provides a real-time combined quality control method and control system for GNSS precise positioning in harsh environments. The existing quality control methods, namely the handling of outliers and multipath effects, are mainly in the post-hoc stage and cannot be performed in advance to reduce the difference in GNSS precise positioning under harsh environments.

[0006] This invention is achieved through the following technical solution:

[0007] A real-time integrated quality control method for precise GNSS positioning in harsh environments, the control method specifically includes the following steps.

[0008] Step 1: Establish a GNSS double-difference mathematical model;

[0009] Step 2: Based on the GNSS double-difference mathematical model in Step 1, an improved processing strategy for the multipath effect of phase and code observations was established;

[0010] Step 3: Based on the improved processing strategy in Step 2, use the improved DIA method that takes into account the maximum number of data snooping attempts to implement outlier handling in the posterior stage.

[0011] Furthermore, step 1 specifically involves the observation i of frequency using the dual differential code (DD code) between receivers and between satellites, and the carrier phase, which is typically expressed as follows:

[0012]

[0013]

[0014] Where ▽ and Δ are the operators for differences between satellites and between receivers, respectively; the superscripts k and s represent the reference satellite and the ordinary satellite, respectively; the subscripts m and n represent the reference station and the space station, respectively; P and Φ represent code and phase observations, respectively; ρ and λ represent the receiving satellite range and wavelength, respectively; N represents integer ambiguity; I and T represent ionospheric and tropospheric errors, respectively; U and u represent code and phase multipath errors, respectively; ∈ and ε represent the observation noise of code and phase observations, respectively.

[0015] Furthermore, the DD mathematical model of the GNSS is read as follows:

[0016]

[0017]

[0018] in,

[0019]

[0020]

[0021]

[0022]

[0023]

[0024] A s This represents the baseline component design matrix in three directions (s×3); x = [dx, dy, dz] T B f =diag(λ1,…λ) f ) represents N; e f and I s Let f represent a row vector with all 1s and an s×s identity matrix; f represent the frequency exponent; D represent the covariance matrix of GNSS observations; σ0 represent the unit weight difference coefficient; and Q represent the auxiliary factor matrix of GNSS observations.

[0025] Furthermore, step 2 specifically involves the multipath effect of phase and code observations, as shown in the following equation:

[0026]

[0027]

[0028]

[0029]

[0030] Where i and j (i < j) represent frequency exponents; n represents the sum of frequencies; D1 and D2 represent the statistical information of the first and second detections; T1 and T2 represent the first and second detection thresholds; It is the difference between the measured values ​​C / N0(θ) of frequencies i and j.

[0031] Furthermore, single-frequency observations are performed on the satellite when D1(θ)≥αT1(θ);

[0032] When only one satellite is conducting multi-frequency observations and D1(θ)≥αT1(θ) and D2(θ)≥βT2(θ), both phase and code observations are deleted.

[0033] At this point, if only one detector exceeds the threshold, the phase and code observations will not be deleted.

[0034] Furthermore, step 3, which uses an improved DIA method that considers the maximum number of data snooping attempts to handle outliers in the posterior phase, specifically includes the following steps:

[0035] Step 3.1: Testing, performing global testing to diagnose the compatibility of the mathematical model;

[0036] Step 3.2: Identification. When the above detection procedure is rejected, the LS estimation is considered unusable and an identification procedure should be performed by searching among alternative hypotheses.

[0037] Step 3.3: Adaptation. After identifying suspicious error statements, the remaining observations are used to perform LS estimation again, and then a relatively reliable parameter solution is obtained.

[0038] Furthermore, step 3.1 specifically involves the following: as shown in formulas (5)-(8), two judgment criteria (detectors) are generated during the calculation process. Under different judgment criteria, anomaly detection will be performed on the code measurement value and carrier phase measurement value, and the observation value will be eliminated. After eliminating a portion of the observations, the least squares method will result in a situation where there is no solution due to missing rank, which is used to detect the occurrence of this situation.

[0039] Furthermore, step 3.2 specifically involves the following: if the LS estimation is deemed unusable, it indicates that some signals are multipath signals. By replacing these multipath signals with other signals and then performing LS estimation verification, the observations that are interfered with by multipath can be identified.

[0040] Furthermore, step 3.3 specifically involves using LS estimation on the remaining normal observations after excluding the multipath signals identified in step 3.2 based on the multipath signals identified in step 3.2, to calculate a relatively accurate positioning result.

[0041] A real-time integrated quality control system for precise GNSS positioning in harsh environments, the control system employing the aforementioned real-time integrated quality control method for precise GNSS positioning in harsh environments, the control system comprising...

[0042] The model building module is used to build a GNSS double-difference mathematical model;

[0043] The processing strategy module is used to establish an improved processing strategy for the multipath effects of phase and code observations based on the GNSS double-difference mathematical model.

[0044] The outlier handling module is used to handle outliers in the posterior phase using an improved DIA method that takes into account the maximum number of data snooping attempts, based on the improved processing strategy of the processing strategy module.

[0045] The beneficial effects of this invention are:

[0046] This invention establishes a method for handling front-end multipath effects, proposes an improved multipath processing strategy for phase and code observations, and improves the reliability of posterior residuals.

[0047] This invention combines the proposed front-end multipath effect processing method with the improved DIA algorithm, which can minimize the influence of errors in harsh environments and obtain reliable results.

[0048] This invention can enhance positioning performance and improve accuracy by identifying and processing multipath / non-line-of-sight observations in complex observation environments. Attached Figure Description

[0049] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0050] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.

[0051] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0052] It should also be understood that the terminology used in this application specification is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this application specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0053] The following is in conjunction with the appendix to this application specification. Figure 1The technical solutions in the embodiments of this application are clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0054] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0055] Example 1

[0056] This security door provides a real-time combined quality control method for GNSS precise positioning in harsh environments. First, based on the multipath effect of phase observations for GNSS precise positioning, a GNSS double-difference mathematical model is proposed and processed, which can further improve the reliability of the posterior residual. Then, in the posterior stage, an improved DIA method that considers the maximum number of data snoopings is used to handle outliers in harsh environments.

[0057] like Figure 1 As shown, a real-time combined quality control method for precise GNSS positioning in harsh environments is disclosed. The control method specifically includes the following steps:

[0058] Step 1: Establish a GNSS double-difference mathematical model;

[0059] Step 2: Based on the GNSS double-difference mathematical model in Step 1, an improved processing strategy for the multipath effect of phase and code observations was established;

[0060] Step 3: Based on the improved processing strategy in Step 2, use the improved DIA method that takes into account the maximum number of data snooping attempts to implement outlier handling in the posterior stage.

[0061] A real-time integrated quality control method for precise GNSS positioning in harsh environments, wherein step 1, establishing a GNSS double-difference mathematical model, specifically comprises:

[0062] Inter-receiver and inter-satellite double differential (DD) codes are introduced to eliminate clock errors and hardware delays, and to mitigate atmospheric delays. The inter-receiver and inter-satellite double differential codes, i.e., DD codes, and the carrier phase versus frequency observation i are typically represented as follows:

[0063]

[0064]

[0065] Where ▽ and Δ are the operators for differences between satellites and between receivers, respectively; the superscripts k and s represent the reference satellite and the ordinary satellite, respectively; the subscripts m and n represent the reference station and the space station, respectively; P and Φ represent code and phase observations, respectively; ρ and λ represent the receiving satellite range and wavelength, respectively; N represents integer ambiguity; I and T represent ionospheric and tropospheric errors, respectively; U and u represent code and phase multipath errors, respectively; ∈ and ε represent the observation noise of code and phase observations, respectively.

[0066] A real-time integrated quality control method for precise GNSS positioning in harsh environments, wherein the DD mathematical model of the GNSS is read as follows:

[0067]

[0068]

[0069] in,

[0070]

[0071]

[0072]

[0073]

[0074]

[0075] A s This represents the baseline component design matrix in three directions (s×3); x = [dx, dy, dz] T B f =diag(λ1,…λ) f ) represents N; e f and I s Let f be a row vector with all 1s and an s×s identity matrix; f be the frequency exponent; D be the covariance matrix of GNSS observations; σ0 be the unit weight difference coefficient; and Q be the auxiliary factor matrix of GNSS observations. To further improve the efficiency of correct DD ambiguity resolution, integer least squares (ILS) methods are frequently used, such as least squares disambiguation decorrelation adjustment (LAMBDA) and improved LAMBDA methods.

[0076] A real-time combined quality control method for precise GNSS positioning in harsh environments is disclosed. Step 2 establishes an improved processing strategy for multipath effects in phase and code observations. Specifically, in harsh environments, severe multipath effects frequently occur, which seriously affect positioning accuracy and reliability. Therefore, it is necessary to establish a processing method for multipath effects in the previous stage. This method can also improve the reliability of the posterior residual of the outlier detection method.

[0077] The carrier-to-noise ratio (C / N0) and satellite elevation θ theoretically exist in low multipath environments; therefore, the functional relationship between satellite orbits of the same type, C / N0 and θ, can be pre-established for a given receiving system and satellite, i.e., the C / N0 template function.

[0078] Specifically, the template function at frequency i of C / N0 is established using reference data observed in a low multipath environment.

[0079] To ensure the effectiveness of the proposed method, it is crucial to determine a highly reliable template function.

[0080] The improved processing strategy for multipath effects in phase and code observations is specifically as follows:

[0081] The multipath effect of phase and code observations is shown in the following equation.

[0082]

[0083]

[0084]

[0085]

[0086] Where i and j (i < j) represent frequency exponents; n represents the sum of frequencies; D1 and D2 represent the statistical information of the first and second detections; T1 and T2 represent the first and second detection thresholds; It is the difference between the measured values ​​C / N0(θ) of frequencies i and j.

[0087] A real-time combined quality control method for precise GNSS positioning in harsh environments is proposed. Multipath errors not only affect C / N0 measurements but also exhibit a different frequency than the C / N0 measurement itself. Therefore, the specific strategy for handling multipath effects is as follows:

[0088] When D1(θ)≥αT1(θ), single-frequency observations are performed on the satellite;

[0089] When only one satellite is conducting multi-frequency observations and D1(θ)≥αT1(θ) and D2(θ)≥βT2(θ), both phase and code observations are deleted.

[0090] At this point, if only one detector exceeds the threshold, the phase and code observations will not be deleted. This is a more stringent decision to avoid the reduction of geometric strength in harsh environments. The coefficients α and β are two constant factors, which are suggested to be defined as 3 or 4 from the perspective of normal distribution. This processing strategy is very effective in this case, especially under relatively severe multipath effects.

[0091] The term "relatively severe" specifically refers to the fact that multipath effects are relatively pronounced when the observation location is surrounded by calm water surfaces, tall trees, flat building facades, or other complex obstructions. This contrasts with open, minimally obstructed or unobstructed observation environments.

[0092] A real-time integrated quality control method for precise GNSS positioning in harsh environments, wherein step 3 uses an improved DIA method that considers the maximum number of data snooping attempts to handle outliers in the posterior stage, specifically including the following steps:

[0093] Step 3.1: Testing, performing global testing to diagnose the compatibility of the mathematical model;

[0094] Step 3.2: Identification. When the above detection procedure is rejected, the LS estimation is considered unusable and an identification procedure should be performed by searching among alternative hypotheses.

[0095] Step 3.3: Adaptation. After identifying suspicious error statements, the remaining observations are used to perform LS estimation again, and then a relatively reliable parameter solution is obtained.

[0096] However, in practical applications, the identification step often involves deciding among multiple alternative hypotheses after the detection step, especially in harsh environments. Furthermore, in harsh environments, the residuals of normally distributed observations are easily contaminated by the strong influence of various outliers. Therefore, normal observations may be excluded during data observation. Here, we set an adjustment threshold T. count This is the maximum number of data observations, and efforts should be made to avoid making erroneous decisions during normal observations in harsh environments.

[0097] A real-time combined quality control method for GNSS precise positioning in harsh environments, wherein step 3.1 is as follows: as in formula (5)-(8), two judgment criteria (detectors) are generated during the calculation process. Under different judgment criteria, anomaly detection will be performed on the code measurement value and carrier phase measurement value, and the observation value will be eliminated. After eliminating a part of the observation, the least squares method will result in a situation of no solution due to missing rank, which is the so-called compatibility of mathematical models, and is used to detect the occurrence of this situation.

[0098] A real-time integrated quality control method for precise GNSS positioning in harsh environments, specifically step 3.2, involves identifying which part of the signal is useful and which part is interference caused by multipath effects. If LS estimation is deemed unusable, it indicates that some signals are multipath signals. By replacing these multipath signals with other signals (observations) and then performing LS estimation verification, the observations interfered with by multipath can be identified.

[0099] A real-time combined quality control method for precise GNSS positioning in harsh environments, wherein step 3.3 specifically involves, based on the multipath signals identified in step 3.2, after excluding these observations, using LS estimation on the remaining normal observations to calculate a relatively accurate positioning result.

[0100] Assume the receiver's true location is (x, y, z), and the location of each satellite is known. For each satellite, we can measure the pseudorange (i.e., the distance) between the receiver and the satellite. If we consider n satellites, each satellite's location is (x, y, z). i ,y i ,z i The pseudorange is d. i For each satellite, we have:

[0101]

[0102] Since the equation is nonlinear, a Taylor expansion is performed near an initial estimation location (x0, y0, z0) to achieve local linearization. The linearized equation can be expressed as:

[0103] Aδx=b

[0104] Here, δx = [δx, δy, δz, δt] T The errors are position and time, A is the design matrix, and b is the observation vector.

[0105] Construct the design matrix and observation vector: Each row of the design matrix A corresponds to a satellite and contains the partial derivative of that satellite with respect to the initial estimated position. The observation vector b contains the pseudorange of each satellite minus the calculated distance from the initial estimated position to each satellite.

[0106] Solving the least squares problem: The goal of the least squares problem is to find δx such that ||Aδx-b|| 2 Minimum. This can be achieved by solving the following equation:

[0107] (A T A)δx=A T b

[0108] Furthermore, update the position and time estimates, repeating the iteration until the required accuracy is met.

[0109] Example 2

[0110] The present invention also provides a real-time integrated quality control system for GNSS precise positioning in harsh environments, using the real-time integrated quality control method for GNSS precise positioning in harsh environments as described in Embodiment 1 above. The control system includes a model building module, a processing strategy module, and an outlier processing module.

[0111] The model building module is used to build a GNSS double-difference mathematical model;

[0112] The processing strategy module is used to establish an improved processing strategy for the multipath effect of phase and code observations based on the GNSS double-difference mathematical model.

[0113] The outlier handling module is used to perform outlier handling in the posterior stage using an improved DIA method that takes into account the maximum number of data snooping attempts, based on the improved processing strategy of the processing strategy module.

[0114] A real-time integrated quality control system for precise GNSS positioning in harsh environments, wherein the model building module establishes a GNSS double-difference mathematical model, specifically as follows:

[0115] Inter-receiver and inter-satellite double differential (DD) codes are introduced to eliminate clock errors and hardware delays, and to mitigate atmospheric delays. The inter-receiver and inter-satellite double differential codes, i.e., DD codes, and the carrier phase versus frequency observation i are typically represented as follows:

[0116]

[0117]

[0118] Where ▽ and Δ are the operators for differences between satellites and between receivers, respectively; the superscripts k and s represent the reference satellite and the ordinary satellite, respectively; the subscripts m and n represent the reference station and the space station, respectively; P and Φ represent code and phase observations, respectively; ρ and λ represent the receiving satellite range and wavelength, respectively; N represents integer ambiguity; I and T represent ionospheric and tropospheric errors, respectively; U and u represent code and phase multipath errors, respectively; ∈ and ε represent the observation noise of code and phase observations, respectively.

[0119] A real-time integrated quality control system for precise GNSS positioning in harsh environments, wherein the DD mathematical model of the GNSS is read as follows:

[0120]

[0121]

[0122] in,

[0123]

[0124]

[0125]

[0126]

[0127]

[0128] A s This represents the baseline component design matrix in three directions (s×3); x = [dx, dy, dz] T B f =diag(λ1,…λ) f ) represents N; e f and I s Let f be a row vector with all 1s and an s×s identity matrix; f be the frequency exponent; D be the covariance matrix of GNSS observations; σ0 be the unit weight difference coefficient; and Q be the auxiliary factor matrix of GNSS observations. To further improve the efficiency of correct DD ambiguity resolution, integer least squares (ILS) methods are frequently used, such as least squares disambiguation decorrelation adjustment (LAMBDA) and improved LAMBDA methods.

[0129] A real-time integrated quality control system for precise GNSS positioning in harsh environments is disclosed. The processing strategy module establishes an improved processing strategy for multipath effects in phase and code observations. Specifically, in harsh environments, severe multipath effects frequently occur, which seriously affect positioning accuracy and reliability. Therefore, it is necessary to establish a method for processing multipath effects in the previous stage. This method can also improve the reliability of the posterior residual of the outlier detection method.

[0130] The carrier-to-noise ratio (C / N0) and satellite elevation θ theoretically exist in low multipath environments; therefore, the functional relationship between satellite orbits of the same type, C / N0 and θ, can be pre-established for a given receiving system and satellite, i.e., the C / N0 template function.

[0131] Specifically, the template function at frequency i of C / N0 is established using reference data observed in a low multipath environment.

[0132] To ensure the effectiveness of the proposed method, it is crucial to determine a highly reliable template function.

[0133] The improved processing strategy for multipath effects in phase and code observations is specifically as follows:

[0134] The multipath effect of phase and code observations is shown in the following equation.

[0135]

[0136]

[0137]

[0138]

[0139] Where i and j (i < j) represent frequency exponents; n represents the sum of frequencies; D1 and D2 represent the statistical information of the first and second detections; T1 and T2 represent the first and second detection thresholds; It is the difference between the measured values ​​C / N0(θ) of frequencies i and j.

[0140] A real-time integrated quality control system for precise GNSS positioning in harsh environments addresses the issue that multipath errors not only affect C / N0 measurements but also exhibit a different frequency than the actual C / N0 measurement. Therefore, the specific strategy for handling multipath effects is as follows:

[0141] When D1(θ)≥αT1(θ), single-frequency observations are performed on the satellite;

[0142] When only one satellite is conducting multi-frequency observations and D1(θ)≥αT1(θ) and D2(θ)≥βT2(θ), both phase and code observations are deleted.

[0143] At this point, if only one detector exceeds the threshold, the phase and code observations will not be deleted. This is a more stringent decision to avoid the reduction of geometric strength in harsh environments. The coefficients α and β are two constant factors, which are suggested to be defined as 3 or 4 from the perspective of normal distribution. This processing strategy is very effective in this case, especially under relatively severe multipath effects.

[0144] A real-time integrated quality control system for precise GNSS positioning in harsh environments, wherein the outlier handling module uses an improved DIA method that considers the maximum number of data snooping attempts to handle outliers in the posterior stage, specifically including the following steps:

[0145] Step 3.1: Testing, performing global testing to diagnose the compatibility of the mathematical model;

[0146] Step 3.2: Identification. When the above detection procedure is rejected, the LS estimation is considered unusable and an identification procedure should be performed by searching among alternative hypotheses.

[0147] Step 3.3: Adaptation. After identifying suspicious error statements, the remaining observations are used to perform LS estimation again, and then a relatively reliable parameter solution is obtained.

[0148] However, in practical applications, the identification step often involves deciding among multiple alternative hypotheses after the detection step, especially in harsh environments. Furthermore, in harsh environments, the residuals of normally distributed observations are easily contaminated by the strong influence of various outliers. Therefore, normal observations may be excluded during data observation. Here, we set an adjustment threshold T. count This is the maximum number of data observations, and efforts should be made to avoid making erroneous decisions during normal observations in harsh environments.

[0149] A real-time combined quality control method for GNSS precise positioning in harsh environments, wherein step 3.1 is as follows: as in formula (5)-(8), two judgment criteria (detectors) are generated during the calculation process. Under different judgment criteria, anomaly detection will be performed on the code measurement value and carrier phase measurement value, and the observation value will be eliminated. After eliminating a part of the observation, the least squares method will result in a situation of no solution due to missing rank, which is the so-called compatibility of mathematical models, and is used to detect the occurrence of this situation.

[0150] A real-time integrated quality control method for precise GNSS positioning in harsh environments, specifically step 3.2, involves identifying which part of the signal is useful and which part is interference caused by multipath effects. If LS estimation is deemed unusable, it indicates that some signals are multipath signals. By replacing these multipath signals with other signals (observations) and then performing LS estimation verification, the observations interfered with by multipath can be identified.

[0151] A real-time combined quality control method for precise GNSS positioning in harsh environments, wherein step 3.3 specifically involves, based on the multipath signals identified in step 3.2, after excluding these observations, using LS estimation on the remaining normal observations to calculate a relatively accurate positioning result.

[0152] Assume the receiver's true location is (x, y, z), and the location of each satellite is known. For each satellite, we can measure the pseudorange (i.e., the distance) between the receiver and the satellite. If we consider n satellites, each satellite's location is (x, y, z). i ,y i ,z i The pseudorange is d. i For each satellite, we have:

[0153]

[0154] Since the equation is nonlinear, a Taylor expansion is performed near an initial estimation location (x0, y0, z0) to achieve local linearization. The linearized equation can be expressed as:

[0155] Aδx=b

[0156] Here, δx = [δx, δy, δz, δt] T The errors are position and time, A is the design matrix, and b is the observation vector.

[0157] Construct the design matrix and observation vector: Each row of the design matrix A corresponds to a satellite and contains the partial derivative of that satellite with respect to the initial estimated position. The observation vector b contains the pseudorange of each satellite minus the calculated distance from the initial estimated position to each satellite.

[0158] Solving the least squares problem: The goal of the least squares problem is to find δx such that ||Aδx-b|| 2 Minimum. This can be achieved by solving the following equation:

[0159] (A T A)δx=A T b

[0160] Furthermore, update the position and time estimates, repeating the iteration until the required accuracy is met.

Claims

1. A real-time combined quality control method for precise GNSS positioning in harsh environments, characterized in that, The control method specifically includes the following steps. Step 1: Establish a GNSS double-difference mathematical model; Step 2: Based on the GNSS double-difference mathematical model in Step 1, an improved processing strategy for the multipath effect of phase and code observations was established; Step 3: Based on the improved processing strategy in Step 2, use the improved DIA method that takes into account the maximum number of data snooping attempts to implement outlier handling in the posterior stage.

2. The real-time combined quality control method for precise GNSS positioning in harsh environments according to claim 1, characterized in that, Specifically, step 1 involves the observation i of frequency using the dual differential code (DD code) between receivers and between satellites, and the carrier phase. This observation is typically represented as follows: Where ▽ and Δ are the operators for differences between satellites and between receivers, respectively; the superscripts k and s represent the reference satellite and the ordinary satellite, respectively; the subscripts m and n represent the reference station and the space station, respectively; P and Φ represent code and phase observations, respectively; ρ and λ represent the receiving satellite range and wavelength, respectively; N represents integer ambiguity; I and T represent ionospheric and tropospheric errors, respectively; U and u represent code and phase multipath errors, respectively; ∈ and ε represent the observation noise of code and phase observations, respectively.

3. The real-time combined quality control method for precise GNSS positioning in harsh environments according to claim 2, characterized in that, The DD mathematical model of the GNSS is read as follows: in, A s This represents the baseline component design matrix in three directions (s×3); x = [dx, dy, dz] T B f =diag(λ1,…λ) f ) represents N; e f and I s Let f represent a row vector with all 1s and an s×s identity matrix; f represent the frequency exponent; D represent the covariance matrix of GNSS observations; σ0 represent the unit weight difference coefficient; and Q represent the auxiliary factor matrix of GNSS observations.

4. The real-time combined quality control method for precise GNSS positioning in harsh environments according to claim 3, characterized in that, Step 2 specifically involves the multipath effect of phase and code observations, as shown in the following equation. Where i and j (i < j) represent frequency exponents; n represents the sum of frequencies; D1 and D2 represent the statistical information of the first and second detections; T1 and T2 represent the first and second detection thresholds; It is the difference between the measured values ​​C / N0(θ) of frequencies i and j.

5. The real-time combined quality control method for precise GNSS positioning in harsh environments according to claim 4, characterized in that, When D1(θ)≥αT1(θ), single-frequency observations are performed on the satellite; When only one satellite is conducting multi-frequency observations and D1(θ)≥αT1(θ) and D2(θ)≥βT2(θ), both phase and code observations are deleted. At this point, if only one detector exceeds the threshold, the phase and code observations will not be deleted.

6. The real-time combined quality control method for precise GNSS positioning in harsh environments according to claim 1, characterized in that, Step 3, which uses an improved DIA method that considers the maximum number of data snooping attempts to handle outliers in the posterior phase, specifically includes the following steps: Step 3.1: Testing, performing global testing to diagnose the compatibility of the mathematical model; Step 3.2: Identification. When the above detection procedure is rejected, the LS estimation is considered unusable and an identification procedure should be performed by searching among alternative hypotheses. Step 3.3: Adaptation. After identifying suspicious error statements, the remaining observations are used to perform LS estimation again, and then a relatively reliable parameter solution is obtained.

7. The real-time combined quality control method for precise GNSS positioning in harsh environments according to claim 6, characterized in that, Specifically, step 3.1 involves formulas (5)-(8) generating two judgment criteria during the calculation process. Under different judgment criteria, anomaly detection will be performed on the code measurement value and carrier phase measurement value, and the observation values ​​will be eliminated. After eliminating a portion of the observations, the least squares method will result in a situation where there is no solution due to missing rank, which is used to detect the occurrence of this situation.

8. The real-time combined quality control method for precise GNSS positioning in harsh environments according to claim 6, characterized in that, Specifically, step 3.2 involves the following steps: if the LS estimation is deemed unusable, it indicates that some signals are multipath signals. By replacing these multipath signals with other signals and then performing the LS estimation verification, the observations that are interfered with by multipath interference can be identified.

9. The real-time combined quality control method for precise GNSS positioning in harsh environments according to claim 6, characterized in that, Specifically, step 3.3 involves using LS estimation on the remaining normal observations after excluding the multipath signals identified in step 3.2 to calculate a relatively accurate positioning result based on the multipath signals identified in step 3.

2.

10. A real-time combined quality control system for precise GNSS positioning in harsh environments, characterized in that, The control system uses the real-time combined quality control method for precise GNSS positioning in harsh environments as described in any one of claims 1-9, and the control system includes, The model building module is used to build a GNSS double-difference mathematical model; The processing strategy module is used to establish an improved processing strategy for the multipath effects of phase and code observations based on the GNSS double-difference mathematical model. The outlier handling module is used to handle outliers in the posterior phase using an improved DIA method that takes into account the maximum number of data snooping attempts, based on the improved processing strategy of the processing strategy module.

Images