A positioning method to improve the RTK fixation rate of geodetic GNSS receivers

Abstract

This invention discloses a positioning method to improve the RTK fixation rate of a geodesic GNSS receiver. After carrier phase and preprocessing, a three-difference solution is obtained. When the calculated post-hoc residual is less than 0.025, the ambiguity jump is detected by combining the post-hoc residual, and the predicted coordinates X(t+1) = X(t) + Δ are calculated. x (t / t+1); Determine whether to perform carrier distance least squares. If not, or if carrier distance least squares fails, output the predicted coordinates as the output coordinates. If carrier distance least squares succeeds, fix the coordinates X(t+1) = X(t+1) + Δ x As the output coordinates, this application achieves preprocessing using the three-difference solution to solve the pre-verification residual and LLI, and solves for the position correction dx. This eliminates the need to determine whether the ambiguity has been fixed, thus maximizing positioning accuracy and fixation rate.

Description

Technical Field

[0001] This invention relates to the field of real-time differential positioning (RTK), and in particular to a positioning method for improving the fixation rate of geodesic GNSS receivers in RTK. Background Technology

[0002] In recent years, a surge of large-scale RTK (Real-Time Kinematic) network construction has emerged in China, with companies such as Qianxun, Sixpoint, and China Mobile appearing to build a nationwide network capable of providing stable and reliable large-scale differential positioning services. Consequently, massive application demands have arisen in fields such as shared bicycles, smart cars, drone-based agricultural applications, and outdoor robots. Leveraging its unique absolute centimeter-level positioning capabilities, RTK technology has achieved large-scale market application in these fields.

[0003] In RTK technology, the ability to achieve stable ambiguity fixation is a crucial factor in determining whether positioning accuracy can reach the centimeter level. Ambiguity is obtained through carrier double difference and then restored to integer values ​​using methods such as rounding or least-squares integer solutions (ILS). In practical applications, when mobile vehicles carrying GNSS (Global Navigation Satellite System) equipment receive satellite signals, the received signals are often interfered with by various environmental factors such as tall buildings, trees, and overpasses. This can cause carrier phase loss, half-cycle or full-cycle cycle slips, leading to problems such as RTK failure, incorrect fixation, and decreased accuracy.

[0004] To address this issue, many scholars have proposed various cycle slip detection methods to identify cycle slips and combine them with partial ambiguity fixing methods. This eliminates the need to include ambiguities containing cycle slips in the fixing process, preventing ambiguity from causing the ILS search ratio value to be too low and resulting in the inability to fix the slip. In addition, some scholars have proposed cycle slip repair techniques to directly repair ambiguities. However, this still requires a new ILS search, which can easily lead to the inability to fix the slip due to re-verification.

[0005] For example, search result 1: "Research on Real-time Dynamic GPS Cycle Slip Repair Method Based on Three-Difference Observations" published in the Journal of Wuhan University (Information Science Edition). This paper introduces the cycle slip detection method, but it is limited to how to identify cycle slips and does not propose how to use the repaired ambiguity to improve the RTK fixation rate.

[0006] For example, search result 2: "Carrier half-cycle repair method and its RTK integer ambiguity fixing method" (202210597076.1). This method only mentions identifying and compensating for the carrier half-cycle from the perspective of the baseband carrier loop; it does not mention the perspective of the positioning algorithm; nor does it mention how to achieve continuous RTK fixing.

[0007] For example, search result 3: "A factor graph RTK positioning method based on ambiguity detection and identification repair" (202510516063.0). This patent introduces the ambiguity after cycle slip repair as factor information of FGO, but does not mention how it plays a role in the RTK FIX solution. It still uses lambda to search and fix it again, which often fails to fix it due to the small number of satellites or the low ratio.

[0008] For example, search result 4: "A relative positioning device and its carrier integer ambiguity resolution method" (201510874632.5). This patent only introduces cycle slip detection and ambiguity fixing respectively, but does not mention how to connect ambiguity repair and RTK stabilization fixing.

[0009] For example, search result 5: "Method for rapid fixing of real-time dynamic nova ambiguity" (201811630314.4). This patent describes calculating the position change by using the fixed ambiguity to calculate the new ambiguity, but does not mention how to fix the ambiguity when there is a half-cycle ambiguity. In addition, this invention solves the position correction number dx by using the fixed ambiguity, which has requirements on the number of fixed ambiguities and the conditions are relatively strict.

[0010] Although the aforementioned patents and papers propose cycle slip detection and repair techniques, these typically refer to directly compensating for ambiguity without altering the ambiguity variance through filtering. This still requires re-fixing via ILS search and performing a fixation check, conditions that could potentially cause the check to fail. Furthermore, re-searching via ILS may result in ambiguity search errors, leading to pseudo-fixation. Additionally, regarding patent (201811630314.4), this patent uses the already fixed ambiguity to calculate the position correction number dx and then reverse-calculates the new ambiguity. Assuming the previous epoch had 5 fixed ambiguities, and the current epoch has 2 half-cycle slips, this method cannot successfully calculate the position correction number dx and will exit directly. Summary of the Invention

[0011] The purpose of this invention is to provide a positioning method that improves the RTK fixation rate of geodetic GNSS receivers, enabling better adaptation to field working environments and search, long-term temperature data acquisition, extended power supply duration, and avoidance of power outages and data loss due to high power consumption.

[0012] To achieve the above technical objectives, this invention provides a positioning method for improving the RTK fixation rate of a geodesic GNSS receiver. After carrier phase and preprocessing, a three-difference solution is obtained, and the post-hoc residual is calculated. When <0.025, combined with the post-test residual detection ambiguity jump The predicted coordinates are X(t+1) = X(t) + Δ x (t / t+1); Determine whether to perform carrier distance least squares. If not, or if carrier distance least squares fails, output the predicted coordinates as the output coordinates. If carrier distance least squares succeeds, fix the coordinates X(t+1) = X(t+1) + Δ x Output as output coordinates.

[0013] This invention provides a positioning method to improve the fixation rate of geodesic GNSS receivers using RTK. It preprocesses the position correction dx using the triple-difference solution pre-verification residual and LLI, eliminating the need to determine if ambiguity has been fixed. This invention solves the problem of inability to fix the position when carrier cycles slip during scene switching, especially with half-cycle repair, maintaining fixation. Furthermore, by combining previous epoch fixed information with the current epoch triple-difference solution result, it maximizes positioning accuracy and fixation rate. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] like Figure 1 As shown, this invention provides a positioning method to improve the RTK fixation rate of a geodesic GNSS receiver. After carrier phase and preprocessing, a three-difference solution is obtained, and the post-hoc residual is calculated. When <0.025, combined with the post-test residual detection ambiguity jump The predicted coordinates are X(t+1) = X(t) + Δ x (t / t+1); Determine whether to perform carrier distance least squares. If not, or if carrier distance least squares fails, output the predicted coordinates as the output coordinates. If carrier distance least squares succeeds, fix the coordinates X(t+1) = X(t+1) + Δ x Output as output coordinates.

[0017] This invention provides a positioning method to improve the fixation rate of geodesic GNSS receivers using RTK. It preprocesses the position correction dx using the triple-difference solution pre-verification residual and LLI, eliminating the need to determine if ambiguity has been fixed. This invention solves the problem of inability to fix the position when carrier cycles slip during scene switching, especially with half-cycle repair, maintaining fixation. Furthermore, by combining previous epoch fixed information with the current epoch triple-difference solution result, it maximizes positioning accuracy and fixation rate.

[0018] As a further improvement, the preprocessing includes: constructing double-difference observation equations for inter-station and inter-satellite differences:

[0019]

[0020] in, For wavelength, For frequency number, The symbol indicates a double difference, with single difference between stations and double difference between satellites. For carrier phase observations, These are pseudorange observations. For the distance between the stars, For ambiguity, For ionospheric error, For tropospheric error, To observe noise; considering the close distance between the rover and the base station, it is assumed that... and Approximately equal to 0.

[0021] As a further improvement, parameters are estimated based on the Kalman filter method, and then implemented using the ILS method with additional set check conditions. time The integer is fixed. and fixed solutions with centimeter-level accuracy. .

[0022] As a further improvement, the triple-difference solution includes: constructing a triple-difference equation by differencing different observations over consecutive epochs: ;

[0023] As a further improvement, observations with cycle slips need to be pre-selected to eliminate the negative impact of abnormal observation information on the accuracy of the triple-difference solution. If the actual application scenario is GNSS+INS, the a priori position change between epochs is directly calculated using the INS velocity and substituted into the observation residual equation. The Median or MAD method is used to identify large residuals and the observations corresponding to these residuals are then removed. If the actual application scenario is only GNSS, LLI information is used to pre-select observations with cycle slips.

[0024] When the remaining observations do not have cycle slips and the number of observations is greater than 3, the triple-difference observation equation is linearized to obtain:

[0025] ;

[0026] Where l, m, and n are linearization coefficients; These are the parameters representing the changes in location to be estimated. Indicates the distance between the satellite and the ground.

[0027] As a further improvement, the least squares algorithm is used to estimate the positional change parameters to be estimated. Calculate the post-test residuals ,in, It is the number of observations. yes The post-hoc residuals corresponding to each observation , yes The weight matrix corresponding to each observation;

[0028] Calculate post-test residuals When <0.025, position change parameter The solution accuracy is considered to be sufficiently high.

[0029] As a further improvement, the position change parameters are then... As prior information, calculate the posterior residuals corresponding to the observations that were initially eliminated through prior residuals or LLI:

[0030] ,get ;

[0031] when If <0.75, then it is considered that there is no integer jump. If the value is less than 0.15, then the carrier phase is considered to have no cycle slip. Otherwise, recalculate. ,right Rounding up ,if If the value is less than 0.15, it is considered that the carrier phase has undergone a half-cycle flip, and the ambiguity of the stored post-hoc residual detection changes abruptly. ;

[0032] when When >=0.75, then for Rounding up ,if If the value is less than 0.15, it is considered that the carrier phase has skipped an integer cycle, and the stored post-hoc residual detection ambiguity jumps. .

[0033] As a further improvement, assuming that a fixed solution has been achieved in the previous epoch but not in the current epoch, the ambiguity that was fixed in the previous epoch is then calculated by adding the position change of the current epoch calculated using the three-difference solution to the position of the previous epoch. Get the current position At the same time, the ambiguity of the current epoch can be obtained by using the ambiguity that has been fixed in the previous epoch. ;

[0034] Ambiguity of the current epoch Substituting into the double-difference observation equation Further position at the current moment Linearization yields Formula 7 as follows:

[0035] Where l', m', and n' are linearization coefficients;

[0036] When the number of observations satisfying equation 7 is greater than 3, it is constructed in matrix form:

[0037]

[0038] in, for 3D residual vector for 3D linearized matrix; , , Pick ; , The number of observations;

[0039] The a posteriori residual can be obtained and calculated using least squares. A value <0.025 indicates the result is considered reliable;

[0040] It can be obtained The fixed solution position at time.

[0041] This invention addresses the problem in the field of GNSS RTK positioning technology, where, when a consumer-grade measurement device is mounted on a mobile vehicle for positioning, environmental factors such as obstruction can cause partial carrier phase loss and ambiguity jumps in half-cycles or whole cycles, resulting in the inability of RTK to maintain a stable position and a decrease in positioning accuracy. The invention provides a new and stable ambiguity fixing strategy that can significantly improve the RTK fixation rate during transitions in adverse scenarios.

[0042] By constructing the double-difference (inter-station and inter-satellite difference) observation equation, as follows:

[0043] (1)

[0044] in, For wavelength, For frequency number, The symbol indicates a double difference, with single difference between stations and double difference between satellites. For carrier phase observations, These are pseudorange observations. For the distance between the stars, For ambiguity, For ionospheric error, For tropospheric error, To observe noise; considering the close distance between the rover and the base station, it can be almost assumed that... and Approximately equal to 0.

[0045] Parameter estimation based on the Kalman filter method, followed by ILS method with certain check conditions, can achieve [the desired result]. time The integer is fixed. and fixed solutions with centimeter-level accuracy. .

[0046] By differing the observations over different epochs, a three-difference equation can be constructed: (2)

[0047] By transforming the above formula, we can obtain the three-difference observation residuals.

[0048] Because of cycle slips, abnormal observations can negatively impact the accuracy of the triple-difference solution, necessitating the pre-elimination of observations with cycle slips. If the actual application scenario is GNSS+INS (Inertial Navigation System), the prior position changes between consecutive epochs can be directly calculated using the INS velocity. These prior position changes are then substituted into the equations, and the Median or MAD (Median Absolute Deviation) method is used to identify and eliminate observations with large residuals. If the actual application scenario is only GNSS, LLI (Loss of Locking Indicator) information can be used to pre-eliminate observations with cycle slips.

[0049] Assuming the remaining observations do not have cycle slips, then , Equal values ​​can be eliminated in the above equation; when the number of observations is greater than 3, linearizing the above equation yields:

[0050] (3)

[0051] Where l, m, n are linearization coefficients; These are the parameters representing the changes in location to be estimated. These are non-reference stars and reference stars, respectively. Indicates the distance between the satellite and the ground.

[0052] The parameters of the positional change to be estimated can be obtained by using the least squares algorithm. Calculate the post-test residuals ,in, It is the number of observations. yes The post-hoc residuals corresponding to each observation , yes The weight matrix corresponding to each observation; it is generally considered that When <0.025, position change parameter The solution accuracy is high enough;

[0053] Then As prior information, calculate the posterior residuals corresponding to the observations that were initially eliminated through prior residuals or LLI:

[0054] (5)

[0055] It can be obtained ;

[0056] when If <0.75, then no integer jump is considered to exist; if If the value is less than 0.15, then the carrier phase is considered to have no cycle slip. Otherwise, recalculate. ,right Rounding up ,if If the value is less than 0.15, it is considered that the carrier phase has undergone a half-cycle flip and is stored. ;

[0057] when When >=0.75, then for Rounding up ,if If the value is less than 0.15, it is considered that the carrier phase has skipped an integer cycle, and the value is stored. ;

[0058] Assuming a fixed solution was achieved in the previous epoch but not in the current epoch, the ambiguity of the fixed solution in the previous epoch can be calculated by adding the position change of the current epoch calculated using the three-difference solution to the position of the previous epoch. Get the current position At the same time, the ambiguity of the current epoch can be obtained by using the ambiguity that has been fixed in the previous epoch. ;

[0059] Will Substitute into formula (1) and find the position at the current time. After linearization, we get:

[0060] (7),

[0061] Where l', m', and n' are linearization coefficients, when the number of observations satisfying the above equation is greater than 3, it can be constructed in matrix form:

[0062] (8)

[0063] in, for 3D residual vector for 3D linearized matrix; , , Pick ; , The number of observations.

[0064] The result can be obtained using least squares, and the apostolic residuals can be calculated. A value <0.025 indicates the result is considered reliable;

[0065] It can be obtained The fixed solution position at time.

[0066] This invention is based on a technique for continuously fixing half-cycle and integer cycle slip ambiguities using the post-hoc residuals of fixed solutions and triple-difference solutions. In existing similar techniques, searches 1 to 4 only mention cycle slip detection and repair, but do not combine them with fixed solutions; search 5 uses already fixed ambiguities to calculate position corrections dx to inversely calculate new ambiguities. Assuming the previous epoch had 5 fixed ambiguities, and the current epoch has 2 half-cycle slips, this method cannot successfully calculate the position corrections dx and will directly exit the fixed solution, becoming a floating-point solution, etc. This invention, however, can achieve a fixed solution.

[0067] This invention is based on RTK fixed solution maintenance technology using triple-difference solution prediction; it can maximize positioning accuracy when carrier distance least squares cannot be performed or the result check fails; this invention can solve the problem of carrier cycle slip caused by scene switching, especially half-cycle repair, which can maintain fixation, and at the same time combine the previous epoch fixed information with the current epoch triple-difference solution result to maximize positioning accuracy and fixation rate.

[0068] It should be understood that the scope of protection sought by this invention is not limited to the non-limiting embodiments, which are merely illustrative examples. The substantive scope of protection claimed in this application is further embodied in the scope provided by the independent claims and their dependent claims.

Claims

1. A positioning method for improving the RTK fixation rate of a geodesic GNSS receiver, characterized in that: After carrier phase preprocessing, the parameters are estimated based on the Kalman filter method, and then the ILS method is applied with additional set check conditions to achieve [the desired result]. time The integer is fixed. and fixed solutions with centimeter-level accuracy. Perform a triple difference solution; Observations with cycle slips need to be removed in advance to eliminate the negative impact of abnormal observation information on the accuracy of the triple difference solution. When the remaining observations do not have cycle slips and the number of observations is greater than 3, the observation residual equation is linearized. The parameters of the position change to be estimated are obtained by using the least squares algorithm. Calculate the post-test residuals When <0.025, position change parameter The solution accuracy is considered sufficiently high; Then change the position parameters As prior information, calculate the posterior residuals corresponding to the observations that were initially eliminated through prior residuals or LLI: get Where l, m, and n are linearization coefficients. For wavelength, For frequency number, The symbol indicates a double difference, with single difference between stations and double difference between satellites. For carrier phase observations, Indicates the distance between the satellite and the ground; when If <0.75, then it is considered that there is no integer jump. If the value is less than 0.15, then the carrier phase is considered to have no cycle slip. Otherwise, recalculate. ,right Rounding up ,if If the value is less than 0.15, it is considered that the carrier phase has undergone a half-cycle flip, and the ambiguity of the stored post-hoc residual detection changes abruptly. ; when When >=0.75, then for Rounding up ,if If the value is less than 0.15, it is considered that the carrier phase has skipped an integer cycle, and the stored post-hoc residual detection ambiguity jumps. ; Assuming a fixed solution was achieved in the previous epoch but not in the current epoch, the ambiguity of the fixed solution from the previous epoch is then calculated by adding the position change of the current epoch, calculated using the three-difference solution, to the position of the previous epoch. Get the current position Simultaneously, the ambiguity of the current epoch is obtained using the ambiguity that has been fixed in the previous epoch. ; Ambiguity of the current epoch Substituting into the double-difference observation equation and further linearizing it at the current position, we obtain Equation 7 as follows: Where l', m', and n' are linearization coefficients; when the number of observations satisfying Equation 7 is greater than 3, it is constructed in matrix form: in, for 3D residual vector for 3D linearized matrix; , , Pick ; , The number of observations; The a posteriori residual can be obtained and calculated using least squares. A value <0.025 indicates the result is considered reliable; get The fixed solution position at time.

2. The positioning method for improving the RTK fixation rate of a geodesic GNSS receiver according to claim 1, characterized in that: The preprocessing includes: constructing double-difference observation equations for inter-station and inter-satellite differences: in, For wavelength, For frequency number, The symbol indicates a double difference, with single difference between stations and double difference between satellites. For carrier phase observations, These are pseudorange observations. For the distance between the stars, For ambiguity, For ionospheric error, For tropospheric error, To observe noise; considering the close distance between the rover and the base station, it is assumed that... and Approximately equal to 0.

3. A positioning method for improving the RTK fixation rate of a geodesic GNSS receiver according to claim 2, characterized in that: The triple-difference solution includes: constructing a triple-difference equation by differencing different observations over consecutive epochs. .

4. A positioning method for improving the RTK fixation rate of a geodesic GNSS receiver according to claim 3, characterized in that: Calculate post-test residuals ,in, It is the number of observations. yes The post-hoc residuals corresponding to each observation , yes The weight matrix corresponding to each observation.

Images