Differential positioning method and device for moving station

Abstract

The invention relates to a moving station differential positioning method and device. The method comprises the following steps: carrying out pseudo-range observation calculation according to a first navigation signal received by a base station to obtain an observation residual sequence and a receiver clock error sequence; performing residual error calculation on the observation residual error sequence according to a median of the receiver clock error sequence to obtain a residual error sequence, and performing gross error elimination on an observation residual error in the observation residual error sequence and a receiver clock error in the receiver clock error sequence according to a mean value and a standard deviation of the residual error sequence to obtain a target observation residual error sequence and a target clock error sequence; obtaining a comprehensive correction number according to the target observation residual error sequence and the target clock error sequence, and obtaining positioning information of the moving station according to the comprehensive correction number and a second navigation signal received by the moving station; the residual error sequence is calculated through the clock error median of the receiver, iterative gross error elimination is carried out in combination with a statistical method, abnormal value interference in observation data is effectively eliminated, and the method has the advantages of improving robustness and enhancing positioning precision and reliability.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of satellite positioning technology, and in particular to a rover differential positioning method and device. BACKGROUND

[0002] As a core enhancement means of modern satellite navigation systems, differential positioning technology is widely used in high-precision measurement, autonomous driving, aerospace, and other fields. The core idea of this technology is to use a reference station with accurate three-dimensional coordinates to calculate correction numbers and send them to nearby rovers in real time. After receiving the correction numbers, the rover corrects its own observation data, effectively eliminating the influence of common errors such as satellite clock bias, orbit deviation, and atmospheric delay. However, traditional differential positioning methods have several technical limitations, which restrict their full performance.

[0003] In the differential positioning process, the observation data of common-view satellites of the reference station and the rover may contain gross errors. These gross errors are usually caused by factors such as satellite signal multipath effects, receiver hardware failures, or environmental interference. The existence of gross errors can cause the error correction numbers calculated by the reference station to be biased, which in turn affects the positioning accuracy of the rover. In application scenarios such as autonomous driving and unmanned aerial vehicle control, which have extremely high safety requirements, such errors can lead to serious accident risks. For example, in the autonomous driving scenario, a slight deviation in positioning accuracy can cause the vehicle to deviate from the planned path, even leading to a collision accident; in unmanned aerial vehicle control, gross errors can cause the navigation system of the unmanned aerial vehicle to make mistakes, affecting flight safety.

[0004] Therefore, it is necessary to propose a rover differential positioning method and device to solve the technical problem that the existing differential positioning does not consider the existence of gross errors in common-view satellites, resulting in low final positioning accuracy. SUMMARY

[0005] Therefore, it is necessary to propose a rover differential positioning method and device to solve the technical problem that the existing differential positioning does not consider the existence of gross errors in common-view satellites, resulting in low final positioning accuracy.

[0006] To solve the above problems, in a first aspect, the present application provides a rover differential positioning method, comprising: obtaining a first navigation signal of a plurality of satellites received by a reference station and a second navigation signal of the plurality of satellites received by a rover; calculating pseudo-range observations according to the first navigation signal to obtain an observation residual sequence and a receiver clock bias sequence; The residual error sequence is calculated according to the median of the receiver clock difference sequence, and the observed residual error in the observed residual error sequence and the receiver clock difference in the receiver clock difference sequence are eliminated according to the mean and standard deviation of the residual error sequence, so as to obtain a target observed residual error sequence and a target clock difference sequence. The integrated correction number is obtained according to the target observed residual error sequence and the target clock difference sequence, and the positioning information of the rover station is obtained according to the integrated correction number and the second navigation signal.

[0007] In a possible implementation, the pseudo-range observation calculation according to the first navigation signal includes: The basic data, satellite information of a plurality of satellites and reference station information are acquired; The cycle slip of the carrier phase of the current epoch in the first navigation signal is detected based on a preset detection method, and the frequencies of two stable and cycle-slip-free carrier signals are determined as target frequencies; The pseudo-range of all frequencies in the first navigation signal is smoothed according to the target frequencies, so as to obtain a smoothed pseudo-range observation value of each satellite; The pseudo-range residual error calculation is performed according to the basic data, the satellite information, the reference station information and the pseudo-range observation values of the plurality of satellites, so as to obtain an observed residual error sequence and a receiver clock difference sequence.

[0008] In a possible implementation, the basic data includes the speed of light, ionospheric delay and tropospheric delay; the satellite information includes satellite clock difference and satellite coordinates of each satellite; the reference station information includes reference station coordinates; and the pseudo-range residual error calculation according to the basic data, the satellite information, the reference station information and the pseudo-range observation values of the plurality of satellites includes: The satellite-geodetic distance of each satellite is obtained according to the satellite coordinates and the reference station coordinates; The ionospheric delay and the tropospheric delay are calculated according to a preset ionospheric model and a preset tropospheric model, respectively, so as to obtain ionospheric delay correction values and tropospheric delay correction values; The satellite-geodetic distance and the pseudo-range observation value of each satellite are calculated according to the ionospheric delay correction values, the tropospheric delay correction values, the satellite clock difference and the speed of light, so as to obtain an observed residual error of each satellite, and an observed residual error sequence is obtained according to the observed residual errors of the plurality of satellites; The corresponding receiver clock difference is obtained according to the observed residual error of each satellite, and a receiver clock difference sequence is obtained according to the receiver clock differences of the plurality of satellites.

[0009] In a possible implementation, the residual calculation on the observation residual sequence according to the median of the receiver clock difference sequence comprises: confirming the median of the receiver clock difference sequence as a receiver clock difference reference value; calculating each observation residual in the observation residual sequence according to the receiver clock difference reference value to obtain a target residual of each satellite, and obtaining a residual sequence according to the target residuals of the plurality of satellites.

[0010] In a possible implementation, the outlier elimination of the observation residual in the observation residual sequence and the receiver clock difference in the receiver clock difference sequence according to the mean and standard deviation of the residual sequence comprises: calculating the mean and standard deviation of the residual sequence; confirming an out-of-limit residual of the target residual in the residual sequence that satisfies a pre-set condition according to the mean and the standard deviation; eliminating the target residual with the largest difference from the mean in the out-of-limit residual of the residual sequence to obtain a first residual sequence, and eliminating the observation residual corresponding to the observation residual sequence and the receiver clock difference corresponding to the receiver clock difference sequence according to the first residual sequence to obtain a target observation residual sequence and a target clock difference sequence.

[0011] In a possible implementation, after the outlier elimination of the observation residual in the observation residual sequence and the receiver clock difference in the receiver clock difference sequence according to the mean and standard deviation of the residual sequence, the method further comprises: after the elimination of the observation residual and the receiver clock difference, eliminating the corresponding satellite observation value in the first navigation signal according to the eliminated observation residual to obtain an effective satellite observation value; updating the receiver clock difference reference value according to the mean of the target clock difference sequence to obtain a target receiver clock difference reference value; calculating each observation residual in the target observation residual sequence according to the target receiver clock difference reference value to obtain a target residual sequence; eliminating the observation residual in the target observation residual sequence and the receiver clock difference in the target clock difference sequence according to the mean and standard deviation of the target residual sequence to obtain a second residual sequence, a cyclic observation residual sequence and a cyclic clock difference sequence, and updating the effective satellite observation value according to the eliminated observation residual to obtain a cyclic effective satellite observation value; When all residuals in the second residual sequence do not satisfy the preset condition or the number of the cyclic valid satellite observations is less than a preset minimum satellite number, the cyclic observation residual sequence and the cyclic clock error sequence are determined as a target observation residual sequence and a target clock error sequence.

[0012] In a possible implementation, the obtaining of the comprehensive correction number according to the target observation residual sequence and the target clock error sequence comprises: determining an optimal receiver clock error reference value according to the target clock error sequence; calculating the observation residual in the target observation residual sequence according to the optimal receiver clock error reference value to obtain the comprehensive correction number of each satellite.

[0013] In a possible implementation, the obtaining of the positioning information of the rover station according to the comprehensive correction number and the second navigation signal comprises: calculating the comprehensive correction number of each satellite and the pseudo-range observation value in the second navigation signal based on an observation equation to obtain a calculation result; the calculation result comprises a relationship between a geometric distance between a satellite and a rover station and a rover station receiver clock error; calculating all satellite calculation results based on a least square method to obtain a position and a clock error of the rover station; the positioning information comprises the position and the clock error.

[0014] In a possible implementation, the observation equation is:

[0015] In the formula, is a pseudo-range observation value of a satellite by a rover station in a second navigation signal; is a light speed; is a geometric distance between a satellite and a rover station; and respectively represent a receiver clock error of a rover station and a clock error of a satellite; and respectively represent a receiver clock error of a rover station and a clock error of a satellite; and respectively represent a model correction value of ionospheric delay and tropospheric delay at a rover station; is a comprehensive correction number of a satellite provided by a reference station; is an observation noise and other un-modeled errors. In a possible implementation, the observation equation is:

[0016] In a second aspect, the present application further provides a rover station differential positioning device, comprising: a signal receiving module configured to acquire a first navigation signal of a plurality of satellites received by a reference station and a second navigation signal of the plurality of satellites received by a rover station;​ a pseudo-range calculation module configured to perform pseudo-range observation calculation according to the first navigation signals to obtain an observation residual sequence and a receiver clock error sequence; a gross error elimination module configured to perform residual calculation on the observation residual sequence according to a median of the receiver clock error sequence to obtain a residual sequence, and eliminate gross errors of observation residuals in the observation residual sequence and receiver clock errors in the receiver clock error sequence according to a mean and a standard deviation of the residual sequence to obtain a target observation residual sequence and a target clock error sequence; an information positioning module configured to obtain a comprehensive correction number according to the target observation residual sequence and the target clock error sequence, and obtain positioning information of the rover station according to the comprehensive correction number and the second navigation signals.

[0017] The present application has the advantages that: the first navigation signals of multiple satellites received by a reference station and the second navigation signals of multiple satellites received by a rover station are obtained; pseudo-range observation calculation is performed according to the first navigation signals to obtain an observation residual sequence and a receiver clock error sequence; residual calculation is performed on the observation residual sequence according to a median of the receiver clock error sequence to obtain a residual sequence, and gross errors of observation residuals in the observation residual sequence and receiver clock errors in the receiver clock error sequence are eliminated according to a mean and a standard deviation of the residual sequence to obtain a target observation residual sequence and a target clock error sequence; a comprehensive correction number is obtained according to the target observation residual sequence and the target clock error sequence, and positioning information of the rover station is obtained according to the comprehensive correction number and the second navigation signals; the residual sequence is calculated through the median of the receiver clock error, and iterative gross error elimination is performed in combination with a statistical method, so that abnormal value interference in the observation data is effectively eliminated, and the advantages of improved robustness, enhanced positioning accuracy and reliability are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 An embodiment flowchart of a rover station differential positioning method provided by the present application is shown in the figure; Figure 2 An embodiment flowchart of step S102 in the rover station differential positioning method provided by the present application is shown in the figure; Figure 1 An embodiment flowchart of step S204 in the rover station differential positioning method provided by the present application is shown in the figure; Figure 3 Figure 2 An embodiment flowchart of step S103 in the rover station differential positioning method provided by the present application is shown in the figure; Figure 4 An embodiment flowchart of step S103 in the rover station differential positioning method provided by the present application is shown in the figure; Figure 1 An embodiment flowchart of step S103 in the rover station differential positioning method provided by the present application is shown in the figure; Figure 5 Figure 1 An embodiment flowchart of step S103 in the rover station differential positioning method provided by the present application is shown in the figure; Figure 6 An embodiment structure diagram of a rover station differential positioning device provided by the present application is shown in the figure.​​ Detailed Implementation

[0019] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0020] like Figure 1 As shown, a specific embodiment of the present invention discloses a rover differential positioning method, comprising: S101. Acquire the first navigation signals from multiple satellites received by the base station and the second navigation signals from multiple satellites received by the rover station.

[0021] The rover differential positioning method provided in this application embodiment can be applied to a rover differential positioning system. The rover differential positioning system can be a software system running on a terminal device. The terminal device can be a server, tablet computer, augmented reality (AR) / virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), mobile phone, etc. This application embodiment does not impose any restrictions on the specific type of terminal device.

[0022] The primary function of the base station is as a reference point with known precise coordinates. It receives satellite signals, calculates differential correction information, and then transmits this information to the rover in real time via a data link. By transmitting differential correction information in real time, the base station helps the rover improve its positioning accuracy. The rover is used for actual unknown point positioning. It simultaneously receives satellite signals and combines them with the differential correction information obtained from the base station to calculate the high-precision three-dimensional coordinates of the current point. The rover is the device that performs the actual measurement, using the data provided by the base station for high-precision positioning. The base station and the rover communicate in real time via a data link. The rover receives the differential correction information sent by the base station and performs joint calculations with its own satellite signal data to obtain high-precision three-dimensional coordinates. This differential positioning (DGNSS) technology can significantly improve positioning accuracy and meet the needs of high-precision measurement. The rover differential positioning system can acquire the first navigation signals from multiple satellites received by the base station and the second navigation signals from multiple satellites received by the rover.

[0023] S102. Perform pseudorange observation calculations based on the first navigation signal to obtain the observation residual sequence and the receiver clock error sequence.

[0024] The pseudo-range observation calculation refers to eliminating observation noise through carrier phase smoothing pseudo-range technology. Pseudo-range noise is usually large and is easily affected by multipath effect and measurement noise, and can only provide meter-level positioning accuracy, which is difficult to meet the demand of high-precision application. The single-frequency phase smoothing pseudo-range method can smooth the noise in the pseudo-range by using the high-precision characteristics of the carrier phase, but is only applicable to the scene where the ionospheric delay change amount between epochs is small, and continuous single-frequency smoothing will lead to ionospheric error accumulation divergence. In view of this, the double-frequency phase smoothing pseudo-range is used to replace the traditional pseudo-range observation value in the embodiment, so as to suppress the measurement noise of the traditional pseudo-range and overcome the error accumulation of the single-frequency phase smoothing pseudo-range, thereby significantly improving the accuracy of the observation data.

[0025] S103, residual calculation is performed on the observation residual sequence according to the median of the receiver clock difference sequence to obtain a residual sequence, and coarse error elimination is performed on the observation residual in the observation residual sequence and the receiver clock difference in the receiver clock difference sequence according to the mean and standard deviation of the residual sequence to obtain a target observation residual sequence and a target clock difference sequence.

[0026] The median of the receiver clock difference sequence refers to taking the middle value after sorting the clock difference values calculated by multiple satellites, which can be realized by a fast selection algorithm. This parameter can resist the interference of individual satellite clock difference abnormal values on the overall clock difference estimation of the system. The coarse error elimination refers to identifying abnormal data based on statistical principles. From the perspective of mathematical statistics, under the premise that the bias terms such as satellite clock difference, atmospheric delay and orbit error are reasonably corrected, the observation data should obey a normal distribution with the mean being the theoretical expectation value and the variance being determined by the observation noise . In the same epoch, the receiver clock difference obtained from the pseudo-range observation values of different satellites should be consistent, and its fluctuation range should be limited within a statistically acceptable narrow interval. Based on the characteristics of the normal distribution, the receiver clock differences of any two satellites should fall within the interval with a high probability , where is the confidence factor, usually taken as 2-3. If the receiver clock difference of a satellite deviates significantly from the interval, the corresponding pseudo-range observation value may have a gross error. The observation residual in the observation residual sequence and the receiver clock difference in the receiver clock difference sequence are subjected to coarse error elimination according to the mean and standard deviation of the residual sequence, and the observation residual and the corresponding receiver clock difference that are out of the confidence interval are eliminated to obtain a target observation residual sequence and a target clock difference sequence.

[0027] S104, comprehensive correction numbers are obtained according to the target observation residual sequence and the target clock difference sequence, and positioning information of the rover station is obtained according to the comprehensive correction numbers and the second navigation signal.

[0028] The comprehensive correction number refers to the error correction quantity of fusing the target observation residual and the target clock difference sequence, and eliminating the influence of the orbit error and the atmospheric delay model residual.

[0029] Specifically, the reference station first acquires navigation signals of multiple satellites, and obtains high-precision pseudo-range observations through carrier phase smoothing technology. The theoretical satellite-ground distance is calculated in combination with the satellite coordinates and the reference station coordinates, and the observation residual is obtained by comparing the pseudo-range observation. The receiver clock bias is calculated using the pseudo-range observation, the satellite-ground distance, the satellite clock bias, the ionospheric delay and the tropospheric delay. The median is used to determine the reference clock bias reference value, the satellite residual is recalculated and the residual sequence is constructed. According to the mean and standard deviation of the residual sequence, the observation residual in the observation residual sequence and the receiver clock bias in the receiver clock bias sequence are removed. After removing the abnormal values, the reference value of the clock bias is iteratively optimized based on the remaining data until the residual distribution does not meet the pre-set condition or the number of satellite observations is less than the pre-set minimum satellite number. Finally, the optimized observation residual and clock bias are used to generate a comprehensive correction number, and the rover station applies the correction number to correct the pseudo-range observation value, and the precise position is solved by the least square method.

[0030] Compared with the prior art, the embodiment provided by the embodiment acquires the first navigation signals of multiple satellites received by the reference station and the second navigation signals of multiple satellites received by the rover station; performs pseudo-range observation calculation according to the first navigation signals to obtain an observation residual sequence and a receiver clock bias sequence; performs residual calculation on the observation residual sequence according to the median of the receiver clock bias sequence to obtain a residual sequence, and removes the observation residual in the observation residual sequence and the receiver clock bias in the receiver clock bias sequence according to the mean and standard deviation of the residual sequence to obtain a target observation residual sequence and a target clock bias sequence; obtains a comprehensive correction number according to the target observation residual sequence and the target clock bias sequence, and obtains positioning information of the rover station according to the comprehensive correction number and the second navigation signals; the residual sequence is calculated through the median of the receiver clock bias, and the iteration of the coarse error removal is performed in combination with the statistical method, so that the abnormal value interference in the observation data is effectively eliminated, and the advantages of improving the robustness, enhancing the positioning accuracy and reliability are achieved.

[0031] Through the above technical solution, the application effectively eliminates the influence of the coarse error in the observation data of the reference station, and improves the calculation accuracy of the comprehensive correction number. After the rover station corrects the observation value using the high-precision correction number, the systematic error caused by the coarse error in the positioning result is significantly reduced. The scheme can be applied to environments with significant multipath effects such as urban canyons and viaducts, and can provide more reliable real-time positioning services for autonomous vehicles, unmanned aerial vehicles and the like.

[0032] In some embodiments of the application, as shown in Figure 2 S102 includes: S201, acquiring basic data and satellite information and reference station information of multiple satellites.

[0033] The basic data may include data such as the speed of light, ionospheric delay, and tropospheric delay. Satellite information may include data such as satellite clock errors and the coordinates of each satellite, which are used to provide basic parameters for error correction of subsequent pseudorange observations. The reference station information includes reference station coordinates, used to calculate the geometric distance between the satellite and the reference station.

[0034] S202. Based on a preset detection method, the cycle slip of the carrier phase in the current epoch of the first navigation signal is detected, and the frequencies of the two stable carrier signals without cycle slips are determined as the target frequency.

[0035] The preset detection method refers to detecting cycle slips through carrier phase change rate or dual-frequency combination methods, such as the TurboEdit method to determine carrier phase continuity. A cycle slip is a sudden change in the integer count of a carrier phase observation, which leads to increased measurement error. Carrier phase refers to the phase value measured by the receiver when tracking the satellite signal carrier; its stability directly affects the pseudorange smoothing effect. The target frequency refers to two frequency bands selected from multiple signal frequencies that are cycle-slip-free and phase-stable, such as the L1 and L2 bands of GPS. The TurboEdit method is used to detect whether a cycle slip exists in the carrier phase of the current epoch. When both carrier signals are detected to be stable and cycle-slip-free, their target frequencies are set as follows: and .

[0036] S203. Based on the target frequency, smooth the pseudorange of all frequencies in the first navigation signal to obtain the smoothed pseudorange observation value for each satellite.

[0037] Smoothing refers to filtering pseudorange observations using carrier phase, for example, by employing a Hatch filtering algorithm to reduce pseudorange noise. This is done when determining the target frequency. and Then, the pseudoranges of all frequencies can be smoothed, as shown in formula (1): (1) In the formula, , and Corresponding to frequencies , and wavelength, Represents any frequency, and It can be equal to or , ; for epochs Frequency primitive pseudorange; and They are respectively and reference station j smoothed pseudo-range; , and , respectively and epoch and frequency carrier phase observation.

[0038] S204, pseudo-range residual calculation is performed according to the basic data, satellite information, reference station information and pseudo-range observation values of multiple satellites, to obtain an observation residual sequence and a receiver clock error sequence.

[0039] Specifically, at the reference station end, first, satellite ephemeris, reference station coordinates, basic data and other basic information are obtained, and then cycle slip detection is performed on the carrier phase data of the current epoch. For example, when it is detected that the L1 frequency band carrier phase jumps more than a preset threshold (for example, 0.5 cycles), it is determined that there is a cycle slip in the frequency band, and then the frequency band with cycle slip is excluded. After selecting two stable frequencies without cycle slip, the carrier phase is used to smooth the pseudo-range. For example, a dual-frequency smoothing algorithm is used to combine the observation values of the L1 and L2 frequency bands to generate more accurate pseudo-range observation values. Finally, the satellite coordinates and the reference station coordinates are used to calculate the satellite-ground distance, and the ionospheric delay model, the tropospheric delay model and the satellite clock error are combined to calculate the observation residual and the receiver clock error for each satellite to form sequence data for subsequent processing.

[0040] Compared with the prior art, the traditional method does not perform cycle slip detection on the carrier phase, and directly uses the original observation value for pseudo-range smoothing, which may cause the frequency band containing cycle slip error to participate in the calculation. For example, when an unrecognized cycle slip occurs in a certain frequency band, the carrier phase data of the frequency band is already inaccurate, and the pseudo-range smoothing based on this will introduce systematic bias. The present scheme selects stable frequency bands without cycle slip through a preset detection method, ensures that the carrier phase data participating in the smoothing processing is reliable, and thus improves the accuracy of the pseudo-range observation value. In addition, the dual-frequency combination smoothing processing can effectively suppress the accumulation of ionospheric delay errors, and has higher anti-interference ability compared with the single-frequency smoothing method.

[0041] Through the above technical scheme, the present application can effectively identify and exclude the carrier signal with cycle slip, avoid the distortion of the pseudo-range observation value caused by phase jump, and at the same time, use dual-frequency smoothing processing to suppress the accumulation and divergence of ionospheric delay errors, to provide more accurate input data for subsequent outlier rejection and positioning solution, thereby improving the overall reliability of the differential positioning.

[0042] In some embodiments of the present application, the satellite information includes the speed of light, satellite clock error, ionospheric delay, tropospheric delay and satellite coordinates of each satellite; the reference station information includes reference station coordinates; and the basic data includes satellite ephemeris, satellite clock error, ionospheric delay, tropospheric delay and reference station coordinates.Figure 3 As shown, step S204 includes: S301, according to the satellite coordinates and the reference station coordinates, the satellite- ground distance of each satellite is obtained.

[0043] Wherein, the satellite- ground distance refers to the geometric distance between the satellite and the reference station, which can be realized by calculating the Euclidean distance of the three-dimensional space coordinate difference of the satellite coordinates and the reference station coordinates, and is used to represent the geometric length of the signal propagation path. Since the coordinates of the reference station are accurately known, the satellite- ground distance can be determined by using the spatial geometric relationship, as shown in formula (2): (2) In the formula, is the reference station coordinates, is the satellite coordinates, is the satellite- ground distance.

[0044] S302, according to the preset ionospheric model and the preset tropospheric model, the ionospheric delay and the tropospheric delay are calculated respectively, and the ionospheric delay correction value and the tropospheric delay correction value are obtained.

[0045] Wherein, the ionospheric delay and the tropospheric delay can be confirmed according to the actual situation, in order to need to correct the ionospheric delay and the tropospheric delay, and obtain the ionospheric delay correction value and the tropospheric delay correction value. The ionospheric delay correction value refers to the correction amount of the signal propagation delay caused by the ionosphere, which can be calculated by Klobuchar model, and is used to eliminate the influence of ionospheric refraction effect on the pseudo range observation. The tropospheric delay correction value refers to the correction amount of the signal delay caused by the troposphere, which can be calculated by Saastamoinen model, and is used to correct the path delay caused by atmospheric refraction.

[0046] S303, according to the ionospheric delay correction value, the tropospheric delay correction value, the satellite clock difference, the light speed, the satellite- ground distance and the pseudo range observation value of each satellite are calculated one by one, the observation residual of each satellite is obtained, and the observation residual sequence is obtained according to the observation residual of multiple satellites.

[0047] Wherein, the satellite clock difference refers to the deviation between the satellite atomic clock and the navigation system standard time, which can be calculated by polynomial fitting through the clock difference parameter in the navigation message, and is used to eliminate the time synchronization error of the satellite end. The pseudo range observation equation in actual application can be converted to obtain the calculation equation of the observation residual, as shown in formula (3): (3) In the formula, is the observation residual of each satellite, is the satellite bsmoothed pseudo-range observation value; is the pseudo-range between the satellite and the reference station; is the speed of light; is the satellite clock bias.

[0048] After the observation residuals of all satellites are calculated, the observation residual sequence can be obtained.

[0049] S304, according to the observation residual of each satellite, the corresponding receiver clock bias is obtained, and the receiver clock bias sequence is obtained according to the receiver clock biases of multiple satellites.

[0050] Wherein, ignoring the influence of orbit error, ionospheric and tropospheric model residual error and observation noise, the receiver clock bias of the reference station can be approximately obtained As shown in formula (4): (4) Specifically, when calculating the differential correction at the reference station end, first, based on the accurate known conditions of satellite coordinates and reference station coordinates, the geometric distance from each satellite to the reference station is calculated. Then, the ionospheric delay and tropospheric delay on the signal propagation path are respectively modeled and corrected by ionospheric model and tropospheric model, for example, Klobuchar model is used to eliminate ionospheric delay, and Saastamoinen model is used to eliminate tropospheric delay. Then, the satellite clock bias parameter and the speed of light parameter are jointly solved with the geometric distance and the pseudo-range observation value, and the observation residual of each satellite is solved through the pseudo-range observation equation. The observation residual reflects the residual error that has not been corrected by the model, including receiver clock bias, multipath effect and other error sources. The observation residual is calculated for each satellite and a sequence is formed, and the receiver clock bias sequence is generated according to the corresponding relationship between the residual and the receiver clock bias, which provides a data basis for subsequent outlier rejection.

[0051] Through the above technical solution, the present application can accurately separate the influence of different error sources on the pseudo-range observation value, for example, after the atmospheric delay error is corrected by the model, the receiver clock bias and the atmospheric delay model residual error are mainly reserved in the observation residual, so that the subsequent outlier detection process can more accurately identify abnormal observation values. This overcomes the limitation of mixing all error sources in traditional differential positioning, and through targeted processing of different characteristic error sources, the speed of spatial error correlation decreasing with distance is significantly slowed down.

[0052] In some embodiments of the present application, step S103 comprises: The median of the receiver clock bias sequence is confirmed as the receiver clock bias reference value.

[0053] The median of the receiver clock error sequence refers to a value in the middle position after all receiver clock error values are arranged in order of size, and can be realized by using a sorting algorithm combined with odd-even quantity judgment logic. The value can effectively resist the influence of individual abnormal clock error on the reference value. Assuming that there are effective satellites at the current epoch, the receiver clock error corresponding to each satellite is obtained by solving the pseudo-range observation equation, and the receiver clock error sequence is obtained according to the pseudo-range observation values of multiple satellites. To enhance robustness and avoid the influence of extreme values, the sequence is first sorted according to the value size, and the median is selected as the receiver clock error reference value. .

[0054] The observation residual in the observation residual sequence is calculated one by one according to the receiver clock error reference value, and the target residual of each satellite is obtained, and the residual sequence is obtained according to the target residuals of multiple satellites.

[0055] The residual sequence refers to a sequence formed by difference calculation of the observation residual of each satellite and the receiver clock error reference value, and can be realized by using element-by-element subtraction operation combined with data container storage. The receiver clock error reference value is brought into formula (5) to calculate the observation residual in the observation residual sequence one by one, and the target residual of each satellite is obtained, as shown in formula (5): (5) In the formula, Residual i represents the target residual of the i th satellite; is the effective satellite number at the current epoch.

[0056] Specifically, in the process of calculating the reference station, the receiver clock error sequence may have abnormal values caused by multipath effect, atmospheric delay residual and hardware delay. By selecting the median as the reference value, the deviation of the overall calculation caused by a single abnormal clock error can be avoided. The observation residual of each satellite needs to be deducted from the reference value to eliminate the influence of the receiver clock error, and a comparable residual sequence is formed. For example, when the receiver clock error sequence contains 5 values, the value in the third position after ascending arrangement is selected as the median, and the observation residual of all satellites is subtracted from the median value to generate a new residual sequence for subsequent analysis.

[0057] Compared with the prior art, the median is used as the reference value, which can effectively eliminate the interference of abnormal clock error. For example, when the clock error jumps at a certain epoch, the median can still remain stable. This processing method makes the generated residual sequence more reliable, and provides an accurate data basis for subsequent outlier detection.

[0058] In some embodiments of the present application, as Figure 4 ​​As shown, step S103 further includes: S401, calculate the mean and standard deviation of the residual sequence.

[0059] Wherein, the mean of the residual sequence and the standard deviation refers to the quantitative analysis of the central tendency and dispersion of the observed residual by statistical method, which can be realized by arithmetic mean method and Bessel formula, and is used to construct the baseline parameters of data screening.

[0060] S402, according to the mean and standard deviation, confirm the target residual in the residual sequence that meets the pre-set confidence condition.

[0061] Wherein, the confidence interval can be set as (typically ), the confidence interval refers to the data screening range constructed based on the assumption of normal distribution, which can be realized by the method of mean plus or minus three times the standard deviation, and is used to identify abnormal observation values deviating from the normal distribution. The pre-set confidence condition can be , so that the target residual in the residual sequence and the calculation result of the mean can be judged one by one according to the mean and the standard deviation, if yes, it means that the target residual is not in the confidence interval, and the target residual is the over-limit residual.

[0062] S403, the target residual with the largest difference value from the mean in the over-limit residual of the residual sequence is removed to obtain the first residual sequence, and the corresponding observed residual in the observed residual sequence and the corresponding receiver clock error in the receiver clock error sequence are removed according to the first residual sequence to obtain the target observed residual sequence and the target clock error sequence.

[0063] Wherein, since the receiver clock error reference value is the median, the theoretical mean of the residual sequence should be 0, and the confidence interval can be . Based on this, the gross error determination process is as follows: check whether there is a target residual in the residual sequence that exceeds the confidence interval , which can be determined by the pre-set confidence condition, confirm the over-limit residual of the target residual in the residual sequence that meets the pre-set confidence condition , then calculate the absolute value of the difference between each over-limit residual and the mean, remove the target residual with the largest absolute value in the residual sequence, i.e. , to obtain the first residual sequence, and remove the corresponding observed residual in the observed residual sequence OMC and the corresponding receiver clock error in the receiver clock error sequence according to the over-limit residual removed in the first residual sequence, to obtain the target observed residual sequence and the target clock error sequence. If there is no over-limit residual in the residual sequence, i.e. the confidence interval Or the number of satellite observations is less than the preset minimum satellite number, then terminate iteration, output the current clock difference reference value.

[0064] Further, if there is a confidence interval belonging to the confidence interval, indicating that the observation residual does not meet the preset confidence condition, then directly proceed to step S104.

[0065] Compared with the prior art, the conventional method usually only uses a single threshold or static statistics for outlier rejection, without considering the influence of systematic bias on the screening strategy. Setting the preset confidence condition as the rejection basis avoids the risk of misjudgment in the presence of systematic bias and improves the identification sensitivity of abnormal values in normal scenarios.

[0066] Through the above technical solution, the present application solves the problem of correction number distortion caused by outliers in the satellite observation data of the reference station, improves the reliability of the residual sequence through a dynamic screening strategy, and further makes the integrated correction number obtained by the rover station more accurate.

[0067] In some embodiments of the present application, as shown in Figure 5 After step S103, it further includes: S501, after the observation residual and the receiver clock difference are rejected, the corresponding satellite observation value in the first navigation signal is rejected according to the rejected observation residual, and the effective satellite observation value is obtained.

[0068] Among them, the effective satellite observation value refers to the effective observation data set reserved after rejecting the satellite observation value with outliers, which can be realized by a method based on residual threshold judgment, and is used to exclude the influence of abnormal satellite signals on subsequent calculation.

[0069] S502, updating the receiver clock difference reference value according to the average value of the target clock difference sequence, to obtain the target receiver clock difference reference value.

[0070] Among them, the average value of the receiver clock difference in the remaining target clock difference sequence is calculated, and the target receiver clock difference reference value is confirmed .

[0071] S503, calculating the observation residual in the target observation residual sequence one by one according to the target receiver clock difference reference value, to obtain the target residual sequence.

[0072] Among them, the target receiver clock difference reference value is substituted into formula (5) to calculate the observation residual in the target observation residual sequence one by one, which is used to update the residual sequence to obtain the target residual sequence .

[0073] S504, according to the mean and standard deviation of the target residual sequence, the observation residual in the target observation residual sequence and the receiver clock error in the target clock error sequence are eliminated, the second residual sequence, the cycle observation residual sequence and the cycle clock error sequence are obtained, and the effective satellite observation value is updated according to the eliminated observation residual, and the cycle effective satellite observation value is obtained.

[0074] Wherein, through the updated target residual sequence and target observation residual sequence, steps S401 to S404 are re-performed, through the above-mentioned cycle process, the second residual sequence, and the cycle observation residual sequence and the cycle clock error sequence eliminated according to the second residual sequence are obtained, and the cycle effective satellite observation value is obtained by updating the effective satellite observation value according to the eliminated observation residual.

[0075] S505, when all residuals in the second residual sequence do not satisfy the preset confidence condition or the number of cycle effective satellite observation values is less than the preset minimum satellite number, the cycle observation residual sequence and the cycle clock error sequence are determined as the target observation residual sequence and the target clock error sequence.

[0076] Wherein, the preset confidence condition refers to the state that all values in the residual sequence are in the preset statistical confidence interval, which can be set as The preset minimum satellite number refers to the minimum satellite number threshold required to maintain positioning calculation, for example, set to 4 satellites, which is used to prevent positioning failure caused by excessive elimination. When all residuals in the second residual sequence do not satisfy the preset confidence condition (i.e. , that is, the residuals are all located in the confidence interval) or the cycle effective satellite observation value is less than the preset minimum satellite number, the cycle observation residual sequence and the cycle clock error sequence are determined as the target observation residual sequence and the target clock error sequence.

[0077] Specifically, after the first gross error elimination, the satellite observation value set is updated to the effective satellite observation value, and at this time the receiver clock error reference value is corrected based on the average value of the target clock error sequence. The corrected clock error reference value is used to recalculate the observation residual in the target observation residual sequence to generate a new target residual sequence. Based on the statistical parameters of the new residual sequence, the gross error elimination operation is performed again to update the cycle observation residual sequence and the cycle clock error sequence, while synchronously updating the satellite observation value set. This process is executed in a loop until any of the following conditions is met: the second residual sequence as a whole does not satisfy the preset confidence condition or the number of available satellites is lower than the preset minimum satellite number. For example, through multiple rounds of iterative processing, the optimal observation data set is gradually approached, and it is ensured that the residual gross error is completely excluded.

[0078] Compared with the prior art, the traditional method usually only performs single rough error elimination and cannot eliminate the problems of false elimination or missed detection caused by deviation of clock difference reference value. The scheme effectively solves the problem of systematic deviation of residual error distribution after single elimination by dynamically updating the clock difference reference value and performing multiple rounds of residual calculation. For example, when a small clock difference deviation is left after the first elimination, the updated reference value can correct the deviation, so that the observation residual calculated in the second time more accurately reflects the true observation error. In addition, the prior art does not set an iteration termination condition, which may cause infinite loop or premature termination. The scheme balances between calculation efficiency and accuracy by pre-setting a signal condition and a satellite number threshold.

[0079] Through the above technical scheme, the completeness and accuracy of rough error detection can be significantly improved, and the problem of residual error accumulation caused by incomplete single elimination can be avoided. In the scene where the reference station has multipath interference or instantaneous signal loss, the clock difference reference value is updated by iteration, which can effectively distinguish between true observation error and pseudo residual error caused by clock difference deviation, thereby improving the calculation accuracy of the combined correction number.

[0080] In some embodiments of the present application, step S104 comprises: determining an optimal receiver clock difference reference value according to the target clock difference sequence; calculating the observation residual in the target observation residual sequence one by one according to the optimal receiver clock difference reference value to obtain the combined correction number of each satellite.

[0081] The optimal receiver clock difference reference value refers to the representative value with the highest stability in the clock difference data retained after multiple iterations of rough error elimination. Specifically, the median or mean of the target clock difference sequence can be used to achieve this. The receiver clock difference in the target clock difference sequence is the optimal receiver clock difference obtained by processing. The determination of the reference value can effectively reduce the influence of abnormal clock difference on subsequent calculation. The combined correction number refers to the correction parameter that fuses the observation residual and the receiver clock difference information. Specifically, the linear combination of the observation residual and the optimal receiver clock difference reference value can be used to achieve this. This parameter is used to eliminate the systematic error in the pseudorange observation value of the rover station. The observation value remaining after rough error elimination is considered to be reliable and effective, and can be used to generate the broadcast correction number. The selected value and the current optimal receiver clock difference reference value are substituted into formula (6) to calculate the combined correction number of the orbit error and the atmospheric delay model residual, and the formula (6) is as follows: (6) In the formula, is the first The integrated correction number of the Beidou satellite. Thus, the receiver clock error and the atmospheric delay error model correction value can be effectively separated, and only the orbit error and the atmospheric delay model residual error is broadcast to the mobile station as a correction number, which can provide accurate compensation for the orbit error and the atmospheric delay for real-time positioning calculation of the mobile station.

[0082] Specifically, after obtaining the target observation residual sequence and the target clock error sequence by completing the outlier rejection, first, the target clock error sequence is statistically analyzed, for example, the mean value thereof is calculated as the optimal receiver clock error reference value. Subsequently, for each satellite target observation residual, the optimal receiver clock error reference value is associated and calculated to obtain the integrated correction number of the corresponding satellite. This process integrates the reliable clock error and observation residual data screened, ensures that the calculation of the correction number is only based on effective information, and avoids the interference of outliers.

[0083] Compared with the prior art, the traditional method usually directly uses the mean value of the original clock error sequence as the reference value, without considering the stability of the data after outlier rejection. However, the reference value generated by the target clock error sequence screened by multiple iterations in the present scheme can significantly reduce the deviation of the correction number caused by abnormal clock error or residual error. For example, in the scene with multipath effect, the traditional method may generate incorrect correction number due to the unremoved clock error data disturbed, while the present scheme effectively excludes the influence of the disturbed data through the screening mechanism of the target clock error sequence.

[0084] In some embodiments of the present application, step S104 comprises: Based on the observation equation, the integrated correction number of each satellite and the pseudo-range observation value in the second navigation signal are calculated to obtain a calculation result; the calculation result includes the relationship between the geometric distance between the satellite and the mobile station and the receiver clock error of the mobile station; Based on the least square method, the calculation results of all satellites are calculated to obtain the position and clock error of the mobile station; the positioning information includes the position and clock error.

[0085] The observation equation refers to a physical model for describing the mathematical relationship between the pseudorange observation value and the position parameter of the rover station, and can be specifically implemented by using an equation set containing the speed of light, the geometric distance, the clock error parameter, and the atmospheric delay correction term. The equation can systematically eliminate common errors such as model residuals of satellite orbit errors, ionospheric and tropospheric delays by introducing the comprehensive correction number into the pseudorange observation value. The least square method refers to a mathematical optimization method for solving the optimal solution by establishing an objective function for minimizing the error square sum, and can be specifically implemented by using the Gauss-Newton iteration algorithm or the Levenberg-Marquardt algorithm. The method can effectively suppress the interference of random noise on the positioning result by balancing the weight of each satellite observation value. In differential positioning, when the spatial distance between the rover station and the reference station is relatively close, the atmospheric delay shows strong correlation in space, and it can be reasonably assumed that the atmospheric delay residuals between the two stations are approximately equal. In addition, the orbit error of the rover station and the common view satellite can also be considered as being highly consistent. Based on this assumption, after the rover station receives the correction number, the pseudorange observation data of the rover station can be corrected, so as to construct the corrected observation equation. The observation equation is shown in formula (7). (7) In the formula, φ is the pseudorange observation value of the rover station to the satellite in the second navigation signal; c is the speed of light; R is the geometric distance between the satellite and the rover station; and δrc and δsc are the receiver clock error of the rover station and the clock error of the satellite, respectively; and δT and δI are the model correction values of the ionospheric delay and the tropospheric delay at the rover station, respectively; and δ is the comprehensive correction number of the satellite provided by the reference station; and ε is the observation noise and other un-modeled errors.

[0086] Specifically, the observation equation is constructed as a multivariate function containing the rover station receiver clock error, the satellite clock error, the geometric distance, and the atmospheric delay correction. In the calculation process, the comprehensive correction number is substituted into the equation to correct the pseudorange observation value of the second navigation signal, to form a linearized observation equation set, which is an equation set for the relationship between the geometric distance between the satellite and the rover station and the receiver clock error of the rover station. When the number of satellites observed by the rover station is more than 4, the equation set is then solved by the least square method, and the specific process for solving the position and the receiver clock error is as follows: (1) initial coordinate setting: the initial coordinates of the rover station are usually set as ​​​​​​​​​​​​​​(2) Linearization processing: Taylor expansion is performed on the geodetic distance in the observation equation at the initial coordinates, and the observation equation is converted into a linear matrix form. At this time, the unknown parameters include the coordinate correction amounts in the x, y and z directions and the receiver clock correction amount, and then the least square method is used to solve the linearized matrix equation. (3) Update of coordinates and receiver clock: the correction obtained in (2) is added to the initial value to complete the update amount of the coordinates and the receiver clock.

[0087] Compared with the prior art, the error sources are eliminated in a targeted manner by establishing an observation equation containing comprehensive corrections.

[0088] In the embodiment of the application, the receiver clock error and the atmospheric delay error are separated from the comprehensive error, the different characteristic error sources are processed in a targeted manner to slow down the decrease of the spatial error correlation with the increase of the distance. The gross error elimination mechanism based on statistical test is introduced to identify and eliminate the abnormal values in the reference station observation data, so that the quality and reliability of the observation data are improved, the accuracy and reliability of the broadcast correction are ensured, the influence of the observation noise on the atmospheric delay correction is effectively inhibited, and the positioning accuracy of the rover station is improved.

[0089] In order to better implement the rover differential positioning method in the embodiment of the application, on the basis of the rover differential positioning method, the embodiment of the application also provides a rover differential positioning device, as shown in Figure 6 The rover differential positioning device 600 comprises: A signal receiving module 601 is configured to acquire the first navigation signals of a plurality of satellites received by the reference station and the second navigation signals of a plurality of satellites received by the rover station. A pseudo-range calculation module 602 is configured to perform pseudo-range observation calculation according to the first navigation signals to obtain an observation residual sequence and a receiver clock error sequence. A gross error elimination module 603 is configured to perform residual calculation on the observation residual sequence according to the median of the receiver clock error sequence to obtain a residual sequence, and perform gross error elimination on the observation residuals in the observation residual sequence and the receiver clock errors in the receiver clock error sequence according to the mean and standard deviation of the residual sequence to obtain a target observation residual sequence and a target clock error sequence. An information positioning module 604 is configured to obtain comprehensive corrections according to the target observation residual sequence and the target clock error sequence, and obtain the positioning information of the rover station according to the comprehensive corrections and the second navigation signals.

[0090] The flow station differential positioning device 600 provided by the above embodiment can implement the technical solutions described in the flow station differential positioning method embodiments described above, and the principles of the specific implementation of each module or unit can be referred to the corresponding content in the flow station differential positioning method embodiments described above, which will not be described here again.

[0091] The flow station differential positioning method and device provided by the present application are described in detail above, and the principles and implementation modes of the present application are described by applying specific examples. The above embodiment is only used to help understand the method and core idea of the present application. Meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed, and the above description should not be understood as a limitation of the present application.

Claims

1. A flow station differential positioning method, characterized in that, The method comprises the following steps: acquiring a first navigation signal of a plurality of satellites received by a reference station and a second navigation signal of the plurality of satellites received by a rover station; performing pseudo-range observation calculation according to the first navigation signal to obtain an observation residual sequence and a receiver clock error sequence; performing residual calculation on the observation residual sequence according to the median of the receiver clock error sequence to obtain a residual sequence, and performing rough error elimination on the observation residual in the observation residual sequence and the receiver clock error in the receiver clock error sequence according to the mean and standard deviation of the residual sequence to obtain a target observation residual sequence and a target clock error sequence; obtaining a comprehensive correction number according to the target observation residual sequence and the target clock error sequence, and obtaining positioning information of the rover station according to the comprehensive correction number and the second navigation signal.

2. The flow station differential positioning method of claim 1, wherein, The method comprises the following steps: acquiring the basic data, satellite information of a plurality of satellites and reference station information; detecting the cycle slip of the carrier phase of the current epoch in the first navigation signal based on a preset detection method, determining the frequency of two carrier signals that are stable and have no cycle slip as a target frequency; performing smoothing processing on the pseudo-range of all frequencies in the first navigation signal according to the target frequency to obtain a smoothed pseudo-range observation value of each satellite; performing pseudo-range residual calculation according to the basic data, the satellite information, the reference station information and the pseudo-range observation value of the plurality of satellites to obtain an observation residual sequence and a receiver clock error sequence.

3. The flow station differential positioning method of claim 2, wherein, The basic data includes the speed of light, ionospheric delay and tropospheric delay; the satellite information includes satellite clock error and satellite coordinates of each satellite; the reference station information includes reference station coordinates; the pseudo-range residual calculation according to the basic data, the satellite information, the reference station information and the pseudo-range observation value of the plurality of satellites to obtain an observation residual sequence and a receiver clock error sequence comprises the following steps: obtaining the satellite-geodetic distance of each satellite according to the satellite coordinates and the reference station coordinates; calculating the ionospheric delay and the tropospheric delay according to a preset ionospheric model and a preset tropospheric model respectively to obtain ionospheric delay correction values and tropospheric delay correction values; calculating the satellite-geodetic distance and the pseudo-range observation value of each satellite according to the ionospheric delay correction value, the tropospheric delay correction value, the satellite clock error and the speed of light to obtain the observation residual of each satellite, and obtaining an observation residual sequence according to the observation residuals of the plurality of satellites; obtaining the corresponding receiver clock error according to the observation residual of each satellite, and obtaining a receiver clock error sequence according to the receiver clock errors of the plurality of satellites.

4. The flow station differential positioning method of claim 1, wherein, The residual calculation on the observation residual sequence according to the median of the receiver clock error sequence to obtain a residual sequence comprises the following steps: confirming the median of the receiver clock error sequence as a receiver clock error reference value; The observation residual in the observation residual sequence is calculated according to the receiver clock difference reference value one by one, so as to obtain the target residual of each satellite, and a residual sequence is obtained according to the target residuals of the plurality of satellites.

5. The flow station differential positioning method of claim 4, wherein, The observation residual in the observation residual sequence and the receiver clock difference in the receiver clock difference sequence are eliminated according to the mean and standard deviation of the residual sequence, so as to obtain a target observation residual sequence and a target clock difference sequence, and the method comprises the following steps: The mean and standard deviation of the residual sequence are calculated; The out-of-limit residual of the target residual in the residual sequence that satisfies the preset confidence condition is confirmed according to the mean and the standard deviation; The target residual with the largest difference value from the mean in the out-of-limit residual of the residual sequence is eliminated, so as to obtain a first residual sequence, and the observation residual in the observation residual sequence and the corresponding receiver clock difference in the receiver clock difference sequence are eliminated according to the first residual sequence, so as to obtain a target observation residual sequence and a target clock difference sequence.

6. The flow station differential positioning method of claim 5, wherein, After the observation residual in the observation residual sequence and the receiver clock difference in the receiver clock difference sequence are eliminated according to the mean and standard deviation of the residual sequence, so as to obtain a target observation residual sequence and a target clock difference sequence, the method further comprises the following steps: After the observation residual and the receiver clock difference are eliminated, the corresponding satellite observation value in the first navigation signal is eliminated according to the eliminated observation residual, so as to obtain an effective satellite observation value; The receiver clock difference reference value is updated according to the mean of the target clock difference sequence, so as to obtain a target receiver clock difference reference value; The observation residual in the target observation residual sequence is calculated one by one according to the target receiver clock difference reference value, so as to obtain a target residual sequence; The observation residual in the target observation residual sequence and the receiver clock difference in the target clock difference sequence are eliminated according to the mean and standard deviation of the target residual sequence, so as to obtain a second residual sequence, a cyclic observation residual sequence and a cyclic clock difference sequence, and the effective satellite observation value is updated according to the eliminated observation residual, so as to obtain a cyclic effective satellite observation value; When all the residuals in the second residual sequence do not satisfy the preset confidence condition or the number of the cyclic effective satellite observation values is less than a preset minimum satellite number, the cyclic observation residual sequence and the cyclic clock difference sequence are determined as a target observation residual sequence and a target clock difference sequence.

7. The flow station differential positioning method of claim 1, wherein, The comprehensive correction number is obtained according to the target observation residual sequence and the target clock difference sequence, and the method comprises the following steps: An optimal receiver clock difference reference value is determined according to the target clock difference sequence; The comprehensive correction number of each satellite is calculated according to the optimal receiver clock difference reference value and the observation residual in the target observation residual sequence.

8. The flow station differential positioning method of claim 7, wherein, The positioning information of the rover station is obtained according to the comprehensive correction number and the second navigation signal, and the method comprises the following steps: The comprehensive correction number of each satellite and the pseudo-range observation value in the second navigation signal are calculated based on an observation equation, so as to obtain a calculation result; the calculation result comprises the relationship between the geometric distance between the satellite and the rover station and the receiver clock difference of the rover station; The position and clock difference of the rover station are obtained by calculating the results of all satellites based on a least square method; and the positioning information comprises the position and clock difference.

9. The flow station differential positioning method of claim 8, wherein, The observation equation is: wherein is the pseudorange observation of the rover station to the satellite in the second navigation signal; is the speed of light; is the geometric distance between the satellite and the rover station; and denote the receiver clock bias of the rover station and the clock bias of the satellite , respectively; and are the model corrections for ionosphere and troposphere delays at the rover station, respectively; is the combined correction number for the satellite provided by the reference station; is the observation noise and other unmodeled errors.

10. A flow station differential positioning apparatus, characterized by, Comprise: The signal receiving module is configured to acquire first navigation signals of multiple satellites received by the reference station and second navigation signals of the multiple satellites received by the rover station; The pseudo-range calculation module is configured to perform pseudo-range observation calculation according to the first navigation signals to obtain an observation residual sequence and a receiver clock difference sequence; The gross error elimination module is configured to perform residual calculation on the observation residual sequence according to a median of the receiver clock difference sequence to obtain a residual sequence, and perform gross error elimination on observation residuals in the observation residual sequence and receiver clock differences in the receiver clock difference sequence according to a mean and a standard deviation of the residual sequence to obtain a target observation residual sequence and a target clock difference sequence; The information positioning module is configured to obtain a comprehensive correction number according to the target observation residual sequence and the target clock difference sequence, and obtain positioning information of the rover station according to the comprehensive correction number and the second navigation signals.

Images