Methods, monitoring terminals, and systems for monitoring landslides in short-distance, high-elevation-difference scenarios.
By calculating the double-difference tropospheric delay and constructing the Earth ellipsoid constraint equation in landslide monitoring, the problem of low RTK positioning accuracy in short-distance, large-elevation-difference scenarios is solved, achieving high-precision landslide monitoring and reliable early warning.
Patent Information
- Application Number
- CN202611335387.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-31
- Publication Date
- 2026-11-17
AI Technical Summary
In short-distance scenarios with large elevation differences, existing RTK positioning technology suffers from severe degradation of vertical positioning accuracy and low ambiguity fixation rate, resulting in insufficient reliability and accuracy of landslide monitoring. This is especially true during the rainy season when atmospheric conditions change drastically, making it difficult to achieve high-precision landslide monitoring.
By acquiring positioning data and meteorological data from base stations and rover stations, the double-difference tropospheric delay is calculated and the Earth ellipsoid constraint equation is constructed. Combined with the differential barometric altimetry model, the tropospheric delay is accurately corrected and the satellite's spatial geometry is improved, thereby enhancing the integer ambiguity fixation rate and vertical positioning accuracy.
High-precision three-dimensional landslide positioning was achieved in environments with large elevation differences, improving the reliability and accuracy of landslide monitoring, ensuring accurate monitoring and early warning of landslide displacement, and maintaining stable ambiguity and vertical positioning accuracy, especially during the rainy season.
Smart Images

Figure CN123111179A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of landslide monitoring, and in particular to methods, monitoring terminals, and systems for monitoring landslides in short-distance, high-elevation-difference scenarios. Background Technology
[0002] In existing technologies, hydropower stations employ Real-Time Kinematic (RTK) technology using the Global Navigation Satellite System for long-term deformation monitoring of landslides. A base station is established on the stable bedrock at the rear edge of the landslide, while multiple rover stations are set up in the middle and front edges of the landslide. The horizontal distance between the base station and each rover station ranges from 150m to 400m, and the relative elevation difference ranges from 120m to 220m. By acquiring raw GNSS observation data from the base station and rover stations in real time and performing double-difference relative positioning calculations, theoretically, the three-dimensional displacement of the flow points at the millimeter to centimeter level can be obtained.
[0003] However, the effective operation of conventional RTK positioning technology relies on the premise that the tropospheric delay error between the base station and the rover can be largely eliminated through double-difference processing. This requires the base station and the rover to be in similar atmospheric environments, meaning the relative elevation difference is typically no more than 100m. In this short-range, high-elevation-difference scenario, the elevation difference between the base station and the rover exceeds 200m. The exponential decay of tropospheric delay with altitude results in a difference of tens of centimeters in tropospheric delay between the two stations. This residual tropospheric delay difference cannot be eliminated through conventional double-difference processing, leading to significant systematic errors in the RTK positioning equations. This directly compromises the integer properties of integer ambiguity, causing ambiguity fixation failures or incorrect fixations. Specifically, during periods of clear weather and stable atmospheric conditions, the vertical positioning accuracy of the rover station is approximately 3-5 cm, which is barely sufficient to meet the requirements for landslide monitoring. However, during the rainy season with high humidity and drastic changes in atmospheric conditions, the vertical positioning accuracy deteriorates to 8-15 cm, with some periods showing even decimeter-level jumps in the calculation results. The integer ambiguity fixation rate drops sharply from over 85% during clear weather to below 60%. Since the main deformation direction of reservoir bank landslides is often vertical settlement, the severe degradation of vertical positioning accuracy directly leads to the rover system's inability to reliably identify minute deformation signals of the landslide body, resulting in multiple false alarms and missed alarms. This seriously affects the reliability of landslide early warning and the scientific nature of hydropower station safety management decisions.
[0004] To address the residual error problem of tropospheric delay in short-distance, high-elevation-difference scenarios, some improvement schemes have been proposed in existing technologies. For example, some schemes use standard atmospheric models to estimate the tropospheric delay at different elevations and then correct the double-difference observation equations. However, standard atmospheric models reflect the average state of the global or regional atmosphere and cannot accurately describe the real-time vertical distribution characteristics of air pressure, temperature, and humidity at the flow site. The correction accuracy is limited in high-elevation-difference scenarios, typically improving the tropospheric delay error by only 20%–30%, which still cannot meet the requirements of high-precision landslide monitoring. Another approach is to introduce external meteorological reanalysis data (such as ERA5) to optimize tropospheric model parameters, but this type of data has low spatiotemporal resolution (typically hourly or tens of kilometers), making it difficult to match the real-time and high-precision requirements of landslide monitoring.
[0005] Furthermore, in actual engineering projects for monitoring landslides in the field, there are also problems such as dispersed equipment deployment, power supply difficulties, and low integration of data acquisition and transmission. Especially when the flow points on the landslide body are distributed on slopes with dense vegetation and steep terrain, wired wiring is difficult and costly to construct. Using split-type equipment (GNSS antenna, receiver, meteorological sensor, communication module, battery, and solar panel are installed separately) is not only complicated to install, but the connecting cables between the components are exposed to the field environment for a long time, which is prone to failures such as insulation aging and poor contact, reducing the long-term operational reliability of the system.
[0006] Therefore, how to provide a high-precision GNSS-RTK positioning solution for short-distance landslide monitoring scenarios with large elevation differences, which can perform high-precision landslide monitoring based on meteorological measurement data, effectively compensate for residual errors of tropospheric delay due to large elevation differences, improve satellite spatial geometry, and adapt to long-term unattended operation in harsh field environments, is a technical problem that urgently needs to be solved in this field. Summary of the Invention
[0007] This application provides a method, monitoring terminal, and system for monitoring landslides in short-distance, high-elevation-difference scenarios. It is used to correct the double-difference tropospheric delay by using meteorological measurement data and construct the Earth ellipsoid constraint equation by combining differential barometric altimetry. This enables high-precision three-dimensional positioning of landslides in short-distance, high-elevation-difference environments, effectively improving vertical positioning accuracy and ambiguity fixation rate, thereby achieving high-precision landslide displacement flow.
[0008] Firstly, this application provides a method for monitoring landslides, applied at the monitoring end. The method includes: constructing a first double-difference observation equation based on acquired baseline positioning data and mobile positioning data, wherein the baseline positioning data includes the positioning information of a baseline station, and the mobile positioning data includes the positioning information of a rover station; determining the double-difference tropospheric delay based on acquired baseline meteorological data and mobile meteorological data, wherein the baseline meteorological data includes meteorological parameters of the baseline station, and the mobile meteorological data includes meteorological parameters of the rover station; correcting the troposphere based on the first double-difference observation equation and the double-difference tropospheric delay to obtain a second double-difference observation equation; determining the absolute elevation of the rover station using a differential barometric altimetry model based on the baseline meteorological data and the mobile meteorological data, and then... The absolute elevation of the rover station is used to construct the Earth ellipsoid constraint equation. The second double-difference observation equation and the Earth ellipsoid constraint equation are solved jointly to obtain the rover station's three-dimensional coordinate correction and integer ambiguity floating-point solution. The integer ambiguity floating-point solution is fixed to obtain the fixed integer ambiguity solution, and the rover station's three-dimensional coordinate correction is updated based on the fixed integer ambiguity solution. The three-dimensional coordinates of the rover station are corrected based on the updated rover station's three-dimensional coordinate correction to obtain the rover station's three-dimensional coordinate positioning data for consecutive epochs. Based on the rover station's three-dimensional coordinate positioning data for consecutive epochs, a displacement time series is generated for landslide monitoring.
[0009] In the above embodiments, the monitoring end acquires benchmark positioning data and benchmark meteorological data collected by the base station, as well as mobile positioning data and mobile meteorological data collected by the rover station. It calculates and compensates for the double-difference tropospheric delay, effectively eliminating the impact of tropospheric delay differences on positioning accuracy under conditions of large elevation differences. Simultaneously, it obtains the absolute elevation of the rover station through a differential barometric altimeter model and constructs an Earth ellipsoid constraint equation, essentially adding a pseudorange observation equation for a virtual satellite at the Earth's center. This provides additional observation constraints, improves the satellite's spatial geometry, and reduces the geometric accuracy attenuation factor, thereby effectively improving the integer ambiguity fixation rate and vertical positioning accuracy. Furthermore, by fixing the integer ambiguity floating-point solution and updating the rover station's three-dimensional coordinate corrections based on the fixed solution, the high-precision characteristics of the carrier phase observations are fully utilized, significantly improving the reliability of the positioning solution. Through the dual enhancement of tropospheric correction and ellipsoidal constraints, high-precision three-dimensional positioning is achieved in short-distance, large-elevation-difference landslide monitoring scenarios. Then, based on the high-precision positioning results, a displacement time series is generated, enabling accurate monitoring and early warning of landslide displacement.
[0010] In a second aspect, this application provides a monitoring terminal, comprising: one or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions; the one or more processors invoke the computer instructions to cause the monitoring terminal to perform the method as described in the first aspect and any possible implementation thereof.
[0011] Thirdly, this application provides a system for monitoring landslides, the system comprising: a monitoring terminal, at least two acquisition terminals, and a user terminal. One of the at least two acquisition terminals is located at a base station, used to collect base positioning data and base meteorological data and send them to the monitoring terminal. The base positioning data includes the location information of the base station, and the base meteorological data includes the meteorological parameters of the base station. The other of the at least two acquisition terminals is located at a rover station, used to collect rover positioning data and rover meteorological data and send them to the monitoring terminal. The rover positioning data includes the location information of the rover station, and the rover meteorological data includes the meteorological parameters of the rover station. The monitoring terminal is used to receive the base positioning data, the base meteorological data, the rover positioning data, and the rover meteorological data, and to execute the method described in the first aspect and any possible implementation thereof to monitor landslides. The monitoring terminal is also used to send landslide warning information to the user terminal when the displacement or displacement rate exceeds a preset displacement warning threshold; the user terminal is used to receive and display the landslide warning information. Attached Figure Description
[0012] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0013] Figure 1 These are schematic diagrams illustrating the methods for monitoring landslides provided in some embodiments; Figure 2 These are schematic diagrams of step S100 provided in some embodiments; Figure 3 These are schematic diagrams of step S200 provided in some embodiments; Figure 4 These are schematic diagrams of step S400 provided in some embodiments; Figure 5 These are schematic diagrams of step S500 provided in some embodiments; Figure 6These are schematic diagrams of step S530 provided in some embodiments; Figure 7 These are schematic diagrams of step S700 provided in some embodiments; Figure 8 This is a schematic diagram of the overall framework of a landslide monitoring system provided in some embodiments; Figure 9 These are cross-sectional views of the approximate reference ellipsoidal meridional plane where the rover is located, provided in some embodiments. Detailed Implementation
[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0015] In the reservoir area of a large hydropower station in the mountainous region of southwestern my country, multiple steep ancient landslides are distributed along the banks. The leading edge of one landslide is adjacent to the normal water level of the reservoir, and its stability directly affects the operational safety of the hydropower station, the flood control safety of downstream towns, and the navigation safety of the reservoir area. A high-speed landslide could potentially create a chain of sluice gate surges, causing significant casualties and property damage. However, rainfall and atmospheric conditions vary greatly from year to year and month to month, making a fixed landslide monitoring scheme unsuitable for practical needs. The hydropower station urgently needs a landslide monitoring system that can intelligently and accurately locate landslides based on local weather conditions and the displacement status of landslide points.
[0016] In existing technologies, landslide deformation monitoring can be achieved using conventional RTK positioning technology. However, conventional RTK suffers from severe degradation of vertical positioning accuracy in scenarios with large elevation differences. The following describes a scenario where landslide monitoring methods from related technologies are used.
[0017] During periods of clear weather and stable atmospheric conditions, the vertical positioning accuracy of the rover station is approximately 3-5 cm, which is barely sufficient to meet the requirements for landslide monitoring. However, during the rainy season with high humidity and drastic changes in atmospheric conditions, the vertical positioning accuracy deteriorates to 8-15 cm, with some periods showing even decimeter-level jumps in the calculation results. During a heavy rainfall event in July 2025, the rover system was unable to fix the integer ambiguity for eight consecutive hours, resulting in the failure to capture the accelerated deformation signal of the landslide body in time, nearly causing a serious safety accident.
[0018] The landslide monitoring method described in this application for short-distance, large-elevation-difference scenarios utilizes positioning data from a base station and a rover, along with meteorological data. It calculates the double-difference tropospheric delay and corrects for it, while simultaneously constructing Earth ellipsoidal constraint equations for joint solution. This effectively eliminates the impact of tropospheric delay differences under large-elevation-difference environments, improving vertical positioning accuracy and ambiguity fixation rate. The following describes a scenario where the landslide monitoring method for short-distance, large-elevation-difference scenarios described in this application is used.
[0019] To address the aforementioned issues, the hydropower station adopted the landslide monitoring system described in this application. Please refer to... Figure 8 The landslide monitoring system provided in this application includes a monitoring terminal, at least two acquisition terminals, and a user terminal. The acquisition terminals include a first acquisition terminal (i.e., the base station) located at a reference station and a second acquisition terminal (i.e., the rover station) located at a rover station. In the landslide monitoring area, one base station is deployed at a stable reference point, and multiple rover stations are deployed at points to be moved. Each station is equipped with one acquisition terminal for collecting positioning observation data and meteorological measurement data. The monitoring terminal is deployed in a hydropower station's generator room or a cloud data center, communicating with each acquisition terminal via a 4G / 5G wireless network or wired fiber optic cable. The monitoring terminal receives reference positioning data, reference meteorological data, rover positioning data, and rover meteorological data sent by each acquisition terminal, executes the landslide monitoring positioning calculation method of this application, and obtains high-precision three-dimensional coordinate positioning data for each rover station. After completing the positioning calculation, the monitoring terminal generates a displacement time series based on the positioning results of consecutive epochs, and calculates the displacement amount and displacement rate. When the displacement amount or displacement rate exceeds a preset displacement warning threshold, the monitoring terminal automatically sends a landslide warning message to the user terminal. User terminals, including smartphones, tablets, personal computers, or dedicated mobile terminal devices, are used to receive and display landslide early warning information sent by the monitoring terminal. User terminals are also used to send query commands to the monitoring terminal to obtain real-time location data and historical displacement data.
[0020] The system is equipped with GNSS receivers and meteorological sensors at both the base station and the rover station to collect positioning and meteorological data in real time. During each rover cycle, the system determines the double-difference satellite distance based on the positioning data and the double-difference tropospheric delay based on the meteorological data, corrects the first double-difference observation equation, and constructs the Earth ellipsoid constraint equation using a differential barometric altimeter model. After joint solution, high-precision three-dimensional positioning results are obtained, and displacement time series are generated based on the positioning results from consecutive epochs. During the rainy season, the system maintains a stable ambiguity, with vertical positioning accuracy consistently better than 2.5 cm. It successfully captured a vertical settlement signal of approximately 3.2 mm from the landslide body and automatically sent a warning message when the displacement rate exceeded the warning threshold.
[0021] It is evident that the system and method for monitoring landslides in this application embodiment can achieve high-precision landslide displacement flow while effectively solving the problems of low vertical positioning accuracy and poor ambiguity fixation rate of conventional RTK in scenarios with large elevation differences, thereby realizing high-precision and high-reliability landslide displacement flow and early warning.
[0022] To facilitate understanding, the method provided in the embodiments of this application will be described in process below, taking into account the above scenarios. Please refer to... Figure 1 The method for monitoring landslides in short-distance, high-elevation-difference scenarios in this application includes steps S100 to S700.
[0023] Step S100: Construct the first double-difference observation equation based on the acquired baseline positioning data and rover positioning data. The baseline positioning data includes the positioning information of the base station, and the rover positioning data includes the positioning information of the rover station.
[0024] The monitoring system constructs the first double-difference observation equation based on the acquired baseline positioning data and mobile positioning data. The purpose of this step is to transform the raw observations into double-difference satellite distances that can be used for high-precision positioning calculations, providing a foundational observation equation for subsequent positioning calculations. Double-difference satellite distances are a core observation in RTK positioning calculations, and their accuracy directly affects the accuracy of subsequent positioning results.
[0025] In some embodiments, such as Figure 2 As shown, step S100 specifically includes steps S110 to S130.
[0026] Step S110: Obtain reference pseudorange observations and reference carrier phase observations from the reference positioning data, and obtain mobile pseudorange observations and mobile carrier phase observations from the mobile positioning data.
[0027] The monitoring unit extracts reference pseudorange and reference carrier phase observations from the received reference positioning data, and mobile pseudorange and mobile carrier phase observations from the mobile positioning data. The pseudorange observations reflect the approximate distance from the satellite to the receiver, while the carrier phase observations provide distance variations with millimeter-level accuracy. The original pseudorange and carrier phase observation equations between the GNSS satellite and the receiver can be expressed as: In equations (1) and (2), This represents the code pseudorange (in meters) between satellite s and receiver r. Let represent the observed carrier phase between satellite s and receiver r (unit: m). Let the spatial coordinates of satellite s at the instant of observation be . (Unit: m), the spatial coordinates of receiver r at the instant of observation are: (Unit: m), then the geometric distance between satellite s and receiver r is... , The speed of light in a vacuum (unit: m / s). Indicates receiver clock bias (unit: seconds). Satellite clock bias (unit: seconds). This represents the ionospheric refraction error (unit: m). The error is the tropospheric refraction error (unit: m). and Multipath delay error (unit: meters). This includes pseudorange measurement noise and other uncorrected errors (unit: m). Indicates the carrier wavelength (unit: m). Indicates the ambiguity over an integer period (unit: week). This includes carrier phase measurement noise and other uncorrected errors (unit: m).
[0028] Step S120: Construct a pseudorange single-difference observation equation based on the reference pseudorange observation value and the moving pseudorange observation value, and construct a carrier phase single-difference observation equation based on the reference carrier phase observation value and the moving carrier phase observation value.
[0029] The monitoring station performs inter-station single-difference processing between the base station and the rover. Inter-station single-difference refers to subtracting the observations of the same satellite between two stations (base station and rover), which can eliminate satellite-related errors, such as satellite clock bias. For pseudorange observations, the rover pseudorange observation is subtracted from the base station pseudorange observation to obtain the pseudorange single-difference observation equation; for carrier phase observations, the rover carrier phase observation is subtracted from the base station carrier phase observation to obtain the carrier phase single-difference observation equation. After inter-station single-difference processing, satellite clock bias is completely eliminated, and for short baselines, it also significantly reduces path propagation error terms such as ionospheric delay error, tropospheric delay error, and multipath effects.
[0030] Using simple difference operator Let i represent the difference between the observed values. Then, the pseudorange single-difference observation equation between the base station i and the observation station j can be expressed as: The carrier phase single-difference observation equation between reference station i and observation station j can be expressed as: In equations (3) and (4), the subscript i represents the base station, the subscript j represents the rover station, and the superscript s represents the satellite; These are the inter-station single-difference pseudorange observations (unit: m). Inter-station single-difference carrier phase observations (unit: m); The inter-station single-difference satellite distance (unit: m); Inter-station single-difference receiver clock bias (unit: seconds); Inter-station single-difference ionospheric delay (unit: m); Inter-station tropospheric delay (unit: m); and Inter-station single-difference multipath delay (unit: m); and Inter-station single-difference observation noise (unit: m); Inter-station single-difference integer ambiguity (unit: week).
[0031] Step S130: Perform inter-satellite double-difference processing on the pseudorange single-difference observation equation to obtain the first pseudorange double-difference observation equation, and perform inter-satellite double-difference processing on the carrier phase single-difference observation equation to obtain the first carrier phase double-difference observation equation.
[0032] The monitoring end then performs inter-satellite double-difference processing on the two different satellites. Inter-satellite double-difference refers to subtracting the single-difference observations from two different satellites again, which can eliminate receiver clock errors. After inter-station single-difference and inter-satellite double-difference processing, the first double-difference observation equation is obtained. For short baselines (0~10km), ionospheric delay errors and multipath effects are greatly reduced, but the double-difference tropospheric delay term in scenarios with large elevation differences cannot be ignored. The double-difference operator is used... The first pseudorange double-difference observation equation (5) and the first carrier phase double-difference observation equation (6) are obtained by expressing the double difference between the observations.
[0033] (5) (6) In equations (5) and (6), the superscripts p and q represent two different satellites in the same GNSS system (p is the reference satellite and q is the non-reference satellite), the subscript j represents the rover station, and the subscript i represents the base station; The pseudorange is a double-difference code (unit: m). These are double-difference carrier phase observations (unit: m). This indicates the distance between two different stations (unit: m). This indicates the double-difference tropospheric delay (unit: m). This represents the integer ambiguity of the double difference (unit: weeks). This represents the pseudorange observation noise of the double-difference code (unit: m). This represents the double-difference phase observation noise (unit: m). λ Indicates the carrier wavelength (unit: m).
[0034] Therefore, it can be seen that the monitoring end extracts the double-difference station star distance from the first double-difference observation equation. The double-difference station-to-satellite distance is a geometric distance calculated based on the satellite's spatial coordinates at the moment of observation and the receiver's approximate position. It is a key observation value in subsequent joint calculations.
[0035] Through the above steps S110 to S130, the monitoring end transforms the original reference pseudorange observations, reference carrier phase observations, mobile pseudorange observations, and mobile carrier phase observations into double-difference satellite distances and integer ambiguities to be determined, providing a mathematical model basis for subsequent high-precision positioning calculations.
[0036] Step S200: Determine the double-difference tropospheric delay based on the acquired baseline meteorological data and mobile meteorological data.
[0037] The monitoring station determines the double-difference tropospheric delay based on the acquired baseline meteorological data and mobile meteorological data. The baseline meteorological data includes meteorological parameters from the base station, and the mobile meteorological data includes meteorological parameters from the rover station. The purpose of this step is to accurately calculate the residual error of the tropospheric delay under conditions of large elevation differences using measured meteorological parameters, providing accurate correction values for subsequent tropospheric corrections.
[0038] In some embodiments, such as Figure 3 As shown, step S200 specifically includes steps S210 to S230.
[0039] Step S210: Obtain the reference air pressure, reference temperature and reference humidity from the reference meteorological data, and determine the oblique tropospheric delay of the reference station to each satellite based on the reference air pressure, reference temperature and reference humidity.
[0040] The monitoring station extracts reference air pressure, reference temperature, and reference humidity from the reference meteorological data. Based on these measured meteorological parameters, the zenith tropospheric dry delay and the reference station zenith tropospheric wet delay are calculated using a zenith tropospheric delay model. Then, based on the satellite elevation angle, the reference station zenith tropospheric dry delay and the reference station zenith tropospheric wet delay are projected onto the direction of each satellite signal propagation to obtain the oblique tropospheric delay of the reference station relative to each satellite.
[0041] In environments with significant elevation differences, the difference in tropospheric delay has a substantial impact on the accuracy of real-time / post-event RTK positioning. Improving the estimation accuracy of tropospheric delay helps to further improve positioning accuracy. Unlike ionospheric delay, tropospheric delay error is independent of the satellite signal frequency, meaning that the influence of tropospheric delay error cannot be eliminated by combining different frequencies. The Zenith Tropospheric Delay (ZTD) can reach 2.3m, increasing with decreasing elevation angle. When the elevation angle is less than 10°, the tropospheric delay can exceed 20m. The Zenith Tropospheric Delay (ZTD) consists of the Zenith Hydrostatic Delay (ZHD) and the Zenith Wet Delay (ZWD), namely: (7) In equation (7), ZHD is mainly composed of dry air (primarily dry air). , The tropospheric delay, accounting for approximately 80-90% of the total delay, is directly related to surface air pressure and can be accurately modeled using surface meteorological parameters (air pressure, temperature). The ZWD, primarily caused by atmospheric water vapor, accounts for approximately 10-20% of the total delay and is more difficult to estimate accurately using models. Assuming satellite 's', and the observation station 'r', the oblique tropospheric delay in the direction of satellite signal propagation... (Unit: m) is usually expressed as the product of the zenith dry and wet tropospheric delays and the mapping function, i.e.: (8) In equation (8), and Let E represent the dry delay mapping function and the wet delay mapping function (dimensionless), respectively, and E be the satellite elevation angle (unit: °). The zenith tropospheric dry delay at station r is given in meters. The zenith tropospheric wet delay (in meters) is the distance from station r to the station.
[0042] Tropospheric dry delay in this invention and moisture delay The calculations were performed using the Saastamoinen model based on measured meteorological parameters, and the calculation formula is as follows: (9) (10) (11) In equations (9) to (11), The latitude of the station (unit: rad) Because the station has a high altitude (m), This indicates the air pressure at the monitoring station (unit: mbar / hPa). Water vapor pressure (unit: mbar / hPa). Temperature (in K).
[0043] in, , and All of these are measured meteorological parameters from observation stations, obtained through meteorological equipment. Compared with the meteorological parameters obtained through standard atmospheric models in the conventional Saastamoinen model, they can obtain higher accuracy tropospheric delay.
[0044] The choice of mapping function directly affects the accuracy of tropospheric delay estimation. The global NMF (Niell Mapping Function) model proposed by Arthur Neill in 1996 is selected for calculation. It does not rely on measured values of surface meteorological parameters, but only uses annual day of year (DOY), station latitude, and station elevation as inputs. The specific formula is as follows: (12) (13) In equations (12) and (13), E is the satellite elevation angle (°). hs The altitude of the station is high (unit: m). a ht =2.53×10 -5 , b ht =5.94×10 -5 , c ht =1.14×10 -5 , a h 、b h , c h 、a w 、b w and c w These are the model coefficients, which can be obtained from a table and will not be elaborated further.
[0045] The above calculations yielded the oblique tropospheric delay of the base station relative to each satellite.
[0046] Step S220: Obtain the flow pressure, flow temperature and flow humidity from the flow meteorological data, and determine the oblique tropospheric delay of the rover station to each satellite based on the flow pressure, flow temperature and flow humidity.
[0047] Similarly, the monitoring end extracts the flow pressure, flow temperature, and flow humidity from the flow meteorological data. Using the same method as in step S210, the rover's zenith tropospheric dry delay and rover's zenith tropospheric wet delay are determined based on the flow pressure, flow temperature, and flow humidity (also calculated using the Saastamoinen model). Then, based on the satellite elevation angle, the rover's zenith tropospheric dry delay and rover's zenith tropospheric wet delay are projected onto the direction of each satellite signal propagation using the Niell mapping function model to obtain the rover's oblique tropospheric delay relative to each satellite.
[0048] Step S230: Perform inter-station and inter-satellite differential processing on the oblique tropospheric delay of the base station and the oblique tropospheric delay of the rover to obtain the double-difference tropospheric delay.
[0049] The monitoring terminal performs inter-station and inter-satellite differential processing on the oblique tropospheric delay of the base station and the oblique tropospheric delay of the rover station to obtain the double-difference tropospheric delay. : (14) In equation (14), the superscripts p and q represent two different satellites in the same GNSS system, with p being the reference satellite, the subscript j representing the rover station, and the subscript i representing the base station; and These are the oblique tropospheric delays of reference station i relative to satellites p and q, respectively. and Let be the oblique tropospheric delays of rover j relative to satellites p and q, respectively. This double-difference tropospheric delay is the estimate of the residual error of the tropospheric delay.
[0050] Through steps S210 to S230, the monitoring station uses measured meteorological parameters instead of standard atmospheric model parameters, more accurately reflecting the tropospheric delay difference between the base station and the rover, providing precise correction values for subsequent tropospheric corrections. Compared to the scheme using standard atmospheric model parameters, this application can significantly improve the accuracy of tropospheric delay correction in scenarios with large elevation differences.
[0051] Step S300: Correct the troposphere based on the first double-difference observation equation and the double-difference tropospheric delay to obtain the second double-difference observation equation.
[0052] The monitoring end corrects the troposphere based on the double-difference star distance determined in step S100 and the double-difference tropospheric delay determined in step S200, obtaining double-difference observations. The purpose of this step is to eliminate the influence of residual tropospheric delay error on the RTK observation equation under conditions of large elevation differences and to restore the integer characteristics of integer ambiguity.
[0053] Specifically, the double-difference tropospheric delay The first double-difference observation equation is introduced, and the tropospheric delay term of the double-difference is compensated to obtain the second double-difference observation equation. Specifically, the second double-difference observation equation includes the second double-difference observation equation for double-difference code pseudorange (Equation (15)) and the second double-difference observation equation for double-difference carrier phase (Equation (16)). (15) (16) in, and These are the pseudorange of the double-difference code (unit: m) and the carrier phase observation (unit: m) after double-difference tropospheric delay correction.
[0054] After tropospheric correction, the double-difference tropospheric delay term is compensated for by the double-difference tropospheric delay, significantly reducing the residual error of the double-difference tropospheric delay and restoring the integer characteristic of integer ambiguity. This provides high-quality observations for high-precision positioning. This step is one of the key steps in achieving high-precision positioning, directly determining the reliability of subsequent ambiguity fixing and positioning calculation.
[0055] Step S400: Based on the baseline meteorological data and the mobile meteorological data, determine the absolute elevation of the rover station using the differential barometric altimetry model, and construct the Earth ellipsoid constraint equation based on the absolute elevation of the rover station.
[0056] The monitoring end constructs Earth ellipsoid constraint equations. The purpose of this step is to obtain the absolute elevation information of the rover station through the differential barometric altimetry model and introduce it as an additional observation constraint into the positioning solution, thereby increasing observation redundancy, improving satellite spatial geometry, and enhancing positioning accuracy in elevation direction.
[0057] In some embodiments, such as Figure 4 As shown, step S400 specifically includes steps S410 to S420.
[0058] Step S410: Obtain the reference air pressure from the reference meteorological data, obtain the reference station elevation from the reference positioning data, and obtain the mobile air pressure from the mobile meteorological data; based on the reference air pressure, mobile air pressure, reference station elevation, average temperature between the reference temperature and the mobile temperature, and average humidity between the reference humidity and the mobile humidity, determine the absolute elevation of the mobile station through the differential barometric altimetry model.
[0059] The monitoring unit extracts reference air pressure from reference meteorological data, reference station elevation information from reference positioning data, and mobile air pressure from mobile meteorological data. Then, the reference station elevation, reference air pressure, mobile air pressure, average temperature between reference temperature and mobile temperature, and average humidity between reference humidity and mobile humidity are substituted into the differential barometric altimetry model to calculate the absolute elevation of the rover station.
[0060] The differential barometric altimeter model is represented as follows: (17); In equation (17), h is the absolute elevation of the rover station (unit: m), h0 is the elevation of the base station (unit: m), P0 is the base pressure (unit: hPa), and P is the rover pressure (unit: hPa). The average temperature (in K) between the base station and the rover. t 0Temperature at the base station measurement point (unit: °C). t R Temperature measured at the rover station (unit: °C); The average humidity parameter is α1 = 1 / 273.15, the constant term is a = 0.00265, and φ is the latitude of the rover station (unit: rad).
[0061] This model utilizes the physical characteristic of high consistency in air pressure over short distances. By differentially correcting the rover's barometric altimetry results using the known elevation of the base station, it effectively eliminates the influence of spatiotemporal variations in the atmospheric physical environment on altimetry accuracy, obtaining absolute rover elevations with decimeter to sub-meter accuracy. The accuracy of the differential barometric altimetry model's calculations is evaluated using long-term static observation data, allowing for the determination of its error. σ h This provides prior parameters for weight determination in subsequent joint solutions.
[0062] Step S420: Based on the absolute elevation of the rover, construct the Earth ellipsoid constraint equation.
[0063] Based on the absolute elevation of the rover obtained in step S410, the monitoring end constructs the Earth ellipsoid constraint equation.
[0064] Specifically, such as Figure 9 As shown, the Earth is approximated as a standard normal ellipsoid. Using the rover's absolute elevation, another approximate ellipsoid with a height h above the reference ellipsoid (WGS-84) can be constructed. Based on the Earth reference ellipsoid parameters of the WGS-84 coordinate system, and using the rover's absolute elevation as the distance from the rover to the reference ellipsoid surface along the normal direction, a geometric constraint relationship between the rover's three-dimensional coordinates and its absolute elevation is established. The Earth ellipsoid constraint equation is expressed as: (18) In equation (18), (x, y, z) are the three-dimensional coordinates of the rover to be solved (unit: m), a and b are the semi-major axis (a = 6378137 m) and semi-minor axis (b = 6356752.3 m) of the WGS84 reference ellipsoid, respectively, and h is the absolute elevation of the rover (unit: m), that is, the distance from the rover along the normal of the Earth reference ellipsoid to the reference ellipsoid. Since h is much smaller than the semi-major / semi-minor axis of the Earth reference ellipsoid, an approximate reference ellipsoid with a semi-major axis of a+h and a semi-minor axis of b+h is used instead, which will not cause too much deviation.
[0065] This equation is equivalent to adding a virtual observation station at the Earth's center, providing a pseudorange observation equation for a virtual satellite. It increases the redundancy of the observation equation, effectively improves the satellite's spatial geometry, and reduces the geometric accuracy attenuation factor, with the most significant improvement in geometric strength in the elevation direction. The introduction of this constraint equation is a key means of improving vertical positioning accuracy.
[0066] Through steps S410 to S420 above, the monitoring end independently obtained the absolute elevation of the rover station based on the measured air pressure data and constructed the Earth ellipsoid constraint equation, providing additional observational constraints for subsequent joint calculations.
[0067] Step S500: Solve the second double-difference observation equation and the Earth ellipsoid constraint equation together to obtain the rover's three-dimensional coordinate correction and integer ambiguity floating-point solution.
[0068] The monitoring end performs a joint solution based on the second double-difference observation equation obtained in step S300 and the Earth ellipsoid constraint equation constructed in step S400 to obtain the three-dimensional coordinate corrections for the rover station. The purpose of this step is to optimally fuse GNSS observations with the Earth ellipsoid constraint equations and obtain high-precision three-dimensional coordinate corrections by solving the weighted least squares method.
[0069] In some embodiments, such as Figure 5 As shown, step S500 specifically includes steps S510 to S530: Step S510: At the approximate location of the rover station, linearize the second double-difference observation equation and the Earth ellipsoid constraint equation to obtain the linearized joint observation equation.
[0070] At the approximate location of the rover station ( X 0 ,Y 0 ,Z 0 At point ), the second double-difference observation equation and the Earth ellipsoid constraint equation are expanded using Taylor series, retaining only the first-order terms, to obtain the linearized joint observation equation: (19) In equation (19), For the residual vector, Let be the vector of parameters to be estimated. H The coefficient matrix, l It is a constant vector (unit: m).
[0071] Specifically, the residual vector (20) Specifically, the parameter vector to be estimated for (twenty one) In equation (21), dx, dy, dz The correction value for the three-dimensional coordinates of the rover station (unit: m). is the double-difference integer ambiguity parameter (unit: week), where the superscripts p and q represent two different satellites in the same GNSS system, with p being the reference satellite, and the superscript T represents matrix transpose.
[0072] Specifically, the coefficient matrix H is shown in (22): (twenty two) The coefficient matrix contains the cosine vector of the distance direction between the two differential satellites. , , (dimensionless), that is: , , These are the two-differential satellite distances. exist X 0 ,Y 0 ,Z 0 The direction cosines (dimensionless) in the three directions.
[0073] In equation (22), the last element 0 indicates that the ellipsoid constraint equation is independent of the integer ambiguity parameter; α , β , γ The expression is: (twenty three) (twenty four) (25) In equations (23) to (25), a and b These are the major and minor semi-axis of the WGS84 reference ellipsoid (unit: m). h (X0, Y0, Z0) represents the absolute elevation of the rover (in meters), and (X0, Y0, Z0) represents the approximate position of the rover (in meters).
[0074] In equation (19), the constant vector l for (26) Step S520: Determine the weights of GNSS observations based on the prior variance definition of double-difference observations, and determine the weights of the Earth ellipsoid constraint equations based on the prior accuracy of the differential barometric altimetry model, and construct a joint weight matrix.
[0075] Assuming all raw GNSS observations at the base station and rover are linearly independent or uncorrelated, then the double-difference observations...L The prior variance is defined as follows: (27) In equation (27), El Satellite elevation angle (unit: °); Factor The weighting factor represents the type of observation. For carrier phase observation ϕ, Factor=1; for pseudorange observation P, Factor=100. a and b are empirical coefficients, usually a=0.003m and b=0.003m.
[0076] According to the error propagation law, the GNSS inter-station single-difference cofactor matrix Q SD It can be represented as: (28) In equation (28), This represents the pseudo-distance between stations' single-difference cofactor matrix. Let represent the carrier phase inter-station single-difference cofactor matrix, and: (29) (30) In equations (29) and (30), s is the number of satellites. and The variance of pseudorange and carrier phase observation noise calculated by equation (27) is expressed in m². The variance is expressed as unit weight (unit: m²).
[0077] Cofactor matrix of double-difference observation equation Q DD It can be represented as: , (31) In equation (29), J The coefficient matrix of the single-difference to double-difference linear transformation. n and m These represent the number of double-difference pseudorange and the number of carrier observations, respectively. E It is a one-dimensional vector with elements equal to 1. I It is the identity matrix; J T for J The transpose of .
[0078] This yields the weight matrix of the GNSS double-difference observations. P GNSS : (32) For the Earth ellipsoid constraint equations, the weights are determined based on the prior accuracy of the differential barometric altimetry model: (33) In equation (33), The variance is expressed as unit weight (unit: m²). The elevation accuracy (mean error) of the rover obtained by the DBA technology in step S400 is calculated (unit: m²).
[0079] GNSS double-difference observation weight matrix P GNSS Weights of the Earth ellipsoid constraint equations P DBA Form a joint weight matrix P: (34) This adaptive weighting strategy achieves optimal fusion of different observation information, which can significantly improve positioning accuracy compared with the equal weight solution scheme.
[0080] Step S530: The weighted least squares method is used to solve the linearized joint observation equation to obtain the three-dimensional coordinate correction of the rover station and the integer ambiguity floating-point solution, and the covariance matrix of the parameter vector to be estimated is calculated.
[0081] The normal equation derived from equation (19) using the weighted least squares method is: (35) Solving equation (35) yields the parameters to be estimated. and its covariance matrix for: (36) (37) In equations (36) and (37), The estimated value is the vector of parameters to be estimated. Let m be the covariance matrix of the parameter vector to be estimated (unit: m²). Representation matrix The inverse matrix.
[0082] in Includes rover 3D coordinate correction data dx , dy , dz (Unit: m) and double-difference integer ambiguity floating-point solution (unit: cycle).
[0083] In some embodiments, such as Figure 6As shown, step S530 may further include: constructing a normal equation based on the linearized joint observation equation (S531); solving the normal equation to obtain an estimated value of the parameter vector to be estimated (S532); determining whether the three-dimensional coordinate correction of the rover obtained in a certain iteration is less than a preset convergence threshold (S533); if it is less than the preset convergence threshold, then the iteration is determined to be converged, and the current three-dimensional coordinate correction of the rover and the integer ambiguity floating-point solution are output (S534); if it is not less than the preset convergence threshold, then the approximate position of the rover is updated based on the three-dimensional coordinate correction of the rover obtained in this iteration, and the linearization expansion step is returned to be executed until the convergence condition is met (S535).
[0084] In this embodiment, the preset convergence threshold is the iteration termination condition, set according to the positioning accuracy requirements of landslide monitoring. The preset convergence threshold is the modulus threshold of the rover's three-dimensional coordinate correction, which can be determined by considering GNSS receiver observation noise and the desired positioning accuracy of the monitoring scenario. For example, in a short-distance, large-elevation-difference landslide monitoring scenario, the convergence threshold can be set to the millimeter level, such as 1-3 mm. In actual engineering deployment, those skilled in the art can adapt the preset convergence threshold according to the monitoring level and equipment accuracy indicators. When the modulus of the iteratively calculated three-dimensional coordinate correction is less than the preset convergence threshold, the coordinate correction is considered sufficiently small, the iteration has converged, and the iterative calculation stops.
[0085] By using the above iterative solution method, the influence of linearization error on the solution accuracy can be eliminated, and the accuracy of the positioning results and the stability of the solution process can be guaranteed by setting a convergence threshold (such as 1 mm).
[0086] This step achieves optimal fusion of GNSS double-difference observations and Earth ellipsoid constraint equations, leveraging the high precision of GNSS observations in the horizontal direction and utilizing the enhancement effect of ellipsoid constraints in the elevation direction, effectively improving three-dimensional positioning accuracy, especially enhancing positioning reliability in the elevation direction.
[0087] Step S600: Fix the integer ambiguity floating-point solution to obtain the integer ambiguity fixed solution, and update the rover's three-dimensional coordinate corrections based on the integer ambiguity fixed solution.
[0088] The monitoring terminal fixes the integer ambiguity floating-point solution obtained in step S500 to obtain the fixed integer ambiguity solution, and updates the rover's 3D coordinate corrections based on this fixed solution. The purpose of this step is to utilize the high precision of the fixed ambiguity solution to further improve the reliability of the 3D coordinate corrections.
[0089] The parameter estimates are obtained from equations (36) and (37). and its variance-covariance matrix Rewritten as follows: (38) In equation (38), This represents the rover coordinate correction parameter. This represents the floating-point solution of the double-difference ambiguity parameters, where the superscripts p and q represent two different satellites in the same GNSS system, and p is the reference satellite; This represents the covariance matrix of the rover coordinate correction parameters. This represents the floating-point solution covariance matrix of the double-difference ambiguity parameters. and Let represent the covariance matrix of the two.
[0090] Then, the ambiguity floating-point solution and its covariance matrix As input parameters to the least squares ambiguity decorrelation adjustment (LAMBDA) algorithm, search for integer alternative solutions for ambiguity. The ambiguity fixing is then checked based on the Ratio value, with a minimum Ratio value of 3. If fixing is successful, the ambiguity fixing solution is... Substitute back to the fixed solution of rover coordinate correction parameters and corresponding covariance matrix : (39) (40) In equations (39) and (40), Representing a matrix Inverse matrix.
[0091] Otherwise, output the floating-point solution directly.
[0092] Through the above steps, the high precision of carrier phase observations is fully utilized, significantly improving the reliability of the positioning solution.
[0093] Step S700: Correct the three-dimensional coordinates of the rover based on the updated three-dimensional coordinate corrections to obtain the three-dimensional coordinate positioning data of the rover for consecutive epochs.
[0094] The monitoring terminal corrects the rover's 3D coordinates based on the updated 3D coordinate correction values from step S600. The approximate position of the rover is then corrected according to these correction values, resulting in the rover's 3D coordinate positioning data. (41) In equation (41), (x,y,z) are the corrected three-dimensional coordinates of the rover (unit: m), (X0,Y0,Z0) are the approximate positions of the rover (unit: m), and (dx,dy,dz) are the corrections to the three-dimensional coordinates of the rover (unit: m).
[0095] The horizontal accuracy of the positioning result is better than 2 cm, and the vertical accuracy is better than 3 cm. The above calculation process (steps S100 to S700) is repeated for consecutive epochs to obtain the three-dimensional coordinate positioning data sequence of the rover for consecutive epochs. This step applies the calculated corrections to the actual coordinates, completing the conversion from observations to high-precision positioning results.
[0096] Through steps S100 to S700, the monitoring terminal accurately compensates for the residual error of double-difference tropospheric delay in environments with large elevation differences using baseline meteorological data and mobile meteorological data, solving the problem of low ambiguity fixation rate of conventional RTK in scenarios with large elevation differences. Simultaneously, it independently obtains the absolute elevation of the rover station using a differential barometric altimeter model and constructs the Earth ellipsoid constraint equation, solving the problems of poor satellite spatial geometry and low vertical positioning accuracy in scenarios with large elevation differences. The synergistic effect of these two methods restores the high-precision positioning capability of conventional RTK methods that previously failed in scenarios with large elevation differences. When applied to landslide monitoring scenarios with short distances and large elevation differences (relative elevation difference between the baseline station and the rover station > 100m), this technical solution can improve the vertical positioning accuracy from 5-8cm in conventional RTK to better than 2cm, and the integer ambiguity fixation rate from 60%-65% to over 95%.
[0097] Step S800: Based on the three-dimensional coordinate positioning data of the rover station in continuous epochs, generate a displacement time series for landslide monitoring.
[0098] The monitoring terminal generates displacement time series based on continuous epochs of three-dimensional coordinate positioning data from the rover station for landslide monitoring. The purpose of this step is to transform high-precision positioning results into displacement monitoring information that can be directly used for landslide disaster early warning, forming a complete monitoring closed loop from data acquisition to disaster early warning.
[0099] In some embodiments, such as Figure 7 As shown, step S800 may specifically include steps S810 to S830: Step S810: Using the three-dimensional coordinate positioning data at the time of the first observation as the initial coordinates, calculate the three-dimensional displacement of each epoch relative to the initial coordinates to form a displacement time series.
[0100] Using the three-dimensional coordinate positioning data at the time of the first observation Using the initial coordinates, calculate the three-dimensional displacement of each epoch relative to the initial coordinates: (42) In equation (42), Let be the three-dimensional displacement (in meters) at epoch t. This provides the three-dimensional coordinate positioning data (unit: m) for the t-th epoch. These are the initial coordinates (unit: m).
[0101] Horizontal displacement is (Unit: m), the three-dimensional displacement is (Unit: m). The displacements of each epoch are arranged in chronological order to form a displacement time series.
[0102] Step S820: Calculate the displacement rate based on the displacement time series.
[0103] The displacement rate, including horizontal and vertical rates (unit: mm / d), is calculated using least-squares linear fitting of the most recent N epochs (e.g., N=10, corresponding to a 10-minute window). The displacement rate reflects the speed of landslide deformation and is an important indicator for determining whether a landslide has entered an accelerated deformation stage.
[0104] Step S830: When the displacement or displacement rate exceeds the preset displacement warning threshold, send landslide warning information to the user terminal.
[0105] The system compares the displacement amount or displacement rate with a preset displacement warning threshold. When the displacement amount or displacement rate exceeds the preset threshold, a landslide warning message is sent to the user terminal, enabling real-time early warning of landslide disasters. This warning message includes key information such as the monitoring point number, current displacement amount, displacement rate, warning level, and timestamp.
[0106] In some embodiments, the preset displacement warning threshold includes a displacement amount threshold and a displacement rate threshold, which are determined comprehensively based on the landslide geological survey report, slope grade, historical deformation monitoring data, and the safety control level of the hydropower station reservoir area. The displacement amount threshold is the upper limit of the allowed cumulative three-dimensional displacement of the rover relative to its initial coordinates; the displacement rate threshold is the upper limit of the allowed displacement change per unit time. Different threshold parameters can be configured for landslides of different risk levels; for example, for high-risk reservoir bank landslides, a cumulative displacement threshold of 5-20 mm and a displacement rate threshold of 0.5-2 mm / d can be set. Those skilled in the art can manually configure and modify the preset displacement warning threshold based on the on-site geological survey results and the landslide risk level. The threshold can be stored in the monitoring terminal's memory, supporting updates and adjustments based on actual on-site conditions.
[0107] Through the above steps S810 to S830, the monitoring terminal forms a complete monitoring closed loop from data acquisition and high-precision calculation to disaster early warning, realizing automated real-time early warning and providing decision support for disaster prevention and mitigation.
[0108] The above embodiments provide a specific landslide displacement monitoring process. In practical applications, factors such as the geographical location of monitoring points and historical displacement patterns can be considered to optimize the monitoring cycle and early warning thresholds, thus developing a more practical landslide monitoring scheme. The technical solution of this application will be described in more detail below from the perspective of the data flow of the method steps.
[0109] In step S100, the monitoring end constructs a first double-difference observation equation based on the acquired reference positioning data and rover positioning data. The reference positioning data includes the positioning information of the reference station (reference pseudorange observations and reference carrier phase observations), and the rover positioning data includes the positioning information of the rover station (rover pseudorange observations and rover carrier phase observations). The spatial coordinates of the satellite at the instant of observation are calculated based on ephemeris data, and the geometric distance between the satellite and the receiver is calculated using the receiver's approximate coordinates. Inter-station single-difference processing is performed between the reference station and the rover station: a pseudorange single-difference observation equation is constructed based on the reference pseudorange observations and the rover pseudorange observations, and a carrier phase single-difference observation equation is constructed based on the reference carrier phase observations and the rover carrier phase observations. Inter-station single-difference eliminates satellite-related errors (such as satellite clock bias). Then, inter-satellite double-difference processing is performed on the two different satellites to eliminate receiver clock bias, resulting in the first double-difference observation equation. The double-difference satellite distance is determined based on the first double-difference observation equation. The output of this step is the double-difference satellite distance, as well as the double-difference integer ambiguity to be calculated and the double-difference tropospheric delay term to be corrected.
[0110] In step S200, the monitoring end determines the double-difference tropospheric delay based on the acquired baseline meteorological data and mobile meteorological data. The baseline meteorological data includes the meteorological parameters of the baseline station (baseline air pressure, baseline temperature, and baseline humidity), and the mobile meteorological data includes the meteorological parameters of the rover station (mobile air pressure, mobile temperature, and mobile humidity). Water vapor pressure is calculated from the measured relative humidity and temperature. Based on the baseline air pressure, baseline temperature, and baseline humidity, the Saastamoinen model is used to calculate the baseline station's zenith tropospheric dry delay and baseline station's zenith tropospheric wet delay. Based on the satellite elevation angle, the Niell mapping function model is used to project the zenith delay onto the propagation direction of each satellite signal to obtain the oblique tropospheric delay of the baseline station relative to each satellite. Similarly, based on the mobile air pressure, mobile temperature, and mobile humidity, the rover station's zenith tropospheric dry delay and rover station's zenith tropospheric wet delay are calculated and projected to obtain the rover station's oblique tropospheric delay relative to each satellite. Inter-station and inter-satellite differential processing is performed on the oblique tropospheric delay of the base station and the oblique tropospheric delay of the rover to obtain the double-difference tropospheric delay. The output of this step is the double-difference tropospheric delay.
[0111] In step S300, the monitoring end corrects the troposphere based on the first double-difference observation equation and the double-difference tropospheric delay to obtain the second double-difference observation equation. The double-difference tropospheric delay is then introduced into the first double-difference observation equation to compensate for the tropospheric delay term, resulting in tropospheric-corrected double-difference observation values. The output of this step is the tropospheric-corrected double-difference code pseudorange and carrier phase observation values.
[0112] In step S400, the monitoring end constructs the Earth ellipsoid constraint equation. Reference pressure is obtained from reference meteorological data, reference station elevation from reference positioning data, and flow pressure from flow meteorological data. The average temperature between the reference temperature and the flow temperature, and the average humidity between the reference humidity and the flow humidity are calculated, and substituted into the differential barometric altimetry model to calculate the rover station's absolute elevation. Then, based on the Earth reference ellipsoid parameters of the WGS84 coordinate system, using the rover station's absolute elevation as the distance from the rover station along the reference ellipsoid normal direction to the reference ellipsoid surface, a geometric constraint relationship between the rover station's three-dimensional coordinates and its absolute elevation is constructed as the Earth ellipsoid constraint equation. The output of this step is the rover station's absolute elevation and its prior mean square error, as well as the Earth ellipsoid constraint equation and its linearization coefficients.
[0113] In step S500, the monitoring end performs a joint solution based on the double-difference observations and the Earth ellipsoid constraint equations to obtain the rover's three-dimensional coordinate corrections and integer ambiguity floating-point solutions. At the approximate location of the rover... At this point, the two observation equations are linearized and expanded to obtain the linearized joint observation equation. The weights of GNSS observations are determined based on the prior variance of the double-difference observations, and the inter-station single-difference cofactor matrix is calculated. Depend on and Composition, obtained from the double difference cofactor matrix. P GNSS Determining the weights of the ellipsoidal constraint equations based on the prior accuracy of differential barometric altimetry. Build a joint power matrix Based on the linearized joint observation equations, a normal equation is constructed. The weighted least squares method is used to solve for the parameter vector to be estimated and its covariance matrix, including the rover's 3D coordinate corrections and double-difference integer ambiguity parameters. It is determined whether the rover's 3D coordinate corrections obtained in this iteration are less than a preset convergence threshold. If so, the iteration converges, and the current rover's 3D coordinate corrections and integer ambiguity floating-point solutions are output. If not, the approximate position is updated, and the iteration continues. The output of this step is the rover's 3D coordinate corrections and integer ambiguity floating-point solutions.
[0114] In step S600, the monitoring terminal from and Extracting rover coordinate correction parameters And double-difference ambiguity floating-point solution parameters ,Right now , ,in This includes the double-difference ambiguity floating-point solution with reference star p. The ambiguity floating-point solution... and its covariance matrix As input parameters to the LAMBDA algorithm, search for fuzzy integer alternative solutions. The ambiguity fixing is then checked based on the Ratio value, with a minimum Ratio value of 3. If fixing is successful, the ambiguity fixing solution is... Substitute back to the fixed solution of rover coordinate correction parameters : Otherwise, output the floating-point solution directly. The output of this step is the updated 3D coordinate correction for the rover.
[0115] In step S700, the monitoring terminal corrects the rover's 3D coordinates based on the updated rover 3D coordinate correction values, obtaining the rover's 3D coordinate positioning data for consecutive epochs. The approximate position of the rover is then corrected based on the rover's 3D coordinate correction values, yielding the rover's 3D coordinate positioning data. This process is repeated for consecutive epochs to obtain the rover's 3D coordinate positioning data sequence for consecutive epochs. The output of this step is the rover's 3D coordinate positioning data and the 3D coordinate positioning data sequence for consecutive epochs.
[0116] In step S800, the monitoring terminal generates a displacement time series based on the three-dimensional coordinate positioning data of the rover station over consecutive epochs for landslide monitoring. Using the three-dimensional coordinate positioning data at the time of the first observation as the initial coordinates, the three-dimensional displacement relative to the initial coordinates is calculated for each epoch, forming the displacement time series. The displacement rate is then calculated. The displacement amount or displacement rate is compared with a preset displacement warning threshold; if the threshold is exceeded, a landslide warning message is sent. The output of this step is the displacement time series and the landslide warning message.
[0117] The monitoring terminal protected in the second aspect of this application executes computer program instructions stored in the memory through a processor to realize the data processing and positioning calculation functions described in the various method embodiments of this application. This monitoring terminal is deployed in a hydropower station engine room or cloud data center, and communicates with each acquisition terminal via a 4G / 5G wireless network or wired fiber optic cable. It receives reference positioning data, reference meteorological data, mobile positioning data, and mobile meteorological data sent by each acquisition terminal, executes the landslide monitoring and positioning calculation method of this application, obtains high-precision three-dimensional coordinate positioning data of each mobile station, and generates a displacement time series based on the positioning results of continuous epochs, calculating the displacement amount and displacement rate. When the displacement amount or displacement rate exceeds a preset displacement warning threshold, the monitoring terminal automatically sends landslide warning information to the user terminal.
[0118] The system protected in the third aspect of this application integrates a monitoring end, at least two acquisition ends, and a user end into a complete mobile system. The acquisition ends are responsible for collecting and uploading baseline and flow data; the monitoring ends are responsible for high-precision positioning calculations and landslide displacement flow judgment; and the user end is responsible for receiving and displaying early warning information. Working together, these three components achieve high-precision, high-reliability, and fully automated landslide displacement flow and early warning in short-distance, large-elevation-difference landslide monitoring scenarios. The acquisition ends integrate a GNSS antenna, meteorological sensor components, equipment protective box, and solar panels onto a single vertical pole, enabling rapid deployment with "pole-to-use" capability. This avoids the complex wiring and installation work of separate equipment, reduces construction and maintenance costs, and improves the long-term operational reliability of the system in harsh field environments.
[0119] The acquisition terminal in this application embodiment is described below from the perspective of hardware structure. Please refer to... Figure 7 This is a schematic diagram of the acquisition end in an embodiment of this application.
[0120] This application also provides a data acquisition terminal. This acquisition terminal is set at a mobile location and serves as either a first acquisition terminal (base station) set at a base station or a second acquisition terminal (rover station) set at a mobile station. It is used to simultaneously acquire GNSS satellite signals and near-surface atmospheric meteorological data in a long-term unattended field environment and transmit the acquired data to a monitoring terminal.
[0121] The data acquisition unit comprises the following core components: main mast, GNSS antenna, meteorological sensor assembly, equipment protective case, and solar panel. These components are electrically connected via feeder lines, data lines, and power lines, forming a complete signal link and power supply path. The following provides a detailed description of the structure, positional relationships, connection relationships, operational relationships, and dimensional relationships of each component.
[0122] 1. Main shaft The main pole serves as the foundation support for the data acquisition unit, vertically positioned on the ground at the mobile location. It is constructed from hot-dip galvanized seamless steel pipe with a diameter of 76mm, a wall thickness of 4mm, and a total length of 3.5m. A flange base is welded to the bottom of the main pole. The base is made of Q235 steel plate, 10mm thick, with a planar dimension of 300mm × 300mm, and 20mm diameter bolt holes at the four corners. The main pole is secured to a cast-in-place concrete foundation using four M16 × 200 anchor bolts. The concrete foundation has a planar dimension of 500mm × 500mm, a depth of 600mm, and uses C30 concrete. The top of the foundation is approximately 50mm above ground level to prevent water from soaking the base.
[0123] The top of the main pole is welded with a GNSS antenna mounting flange. The flange is made of Q235 steel plate, 8mm thick, with a planar dimension of 150mm × 150mm, and a 30mm diameter feed line through-hole is located in the center. Multiple clamp mounting positions are spaced along the height of the main pole for fixing the meteorological sensor components, equipment protective boxes, and solar panel mounting brackets. The height spacing between each mounting position is determined based on functional requirements and structural mechanics optimization.
[0124] The main pole serves as a structural support for the GNSS antenna, weather sensor components, equipment protective box, and solar panels, ensuring the rational vertical layout and relative stability of each component. Simultaneously, the hollow interior of the main pole acts as a conduit for feeder and data cables, enabling concealed cable installation and preventing cables from being exposed to the external environment.
[0125] The fit between the main body and each component is as follows: the mounting brackets of each component are tightly connected to the pipe wall of the main body via U-shaped clamps. Rubber gaskets are placed between the clamps and the pipe wall to increase friction and prevent scratches on the pipe wall. The clamping torque is controlled between 40 N·m and 60 N·m to ensure reliable fixation of each component on the main body, while avoiding excessive clamping that could cause pipe wall deformation.
[0126] 2. GNSS antenna The GNSS antenna, the signal receiving element at the acquisition end, is fixedly mounted on the top of the main pole and is used to receive L-band microwave signals transmitted by GNSS satellites. The GNSS antenna is a dual-frequency, four-system, full-band antenna, capable of simultaneously receiving satellite signals in the GPS L1 (1575.42MHz) and L2 (1227.60MHz), BeiDou B1 (1561.098MHz) and B2 (1207.14MHz), GLONASS G1 (1602.5625MHz) and G2 (1246.4375MHz), and Galileo E1 (1575.42MHz) and E5a (1176.45MHz) bands.
[0127] The internal structure of a GNSS antenna includes an antenna element, a low-noise amplifier (LNA), a bandpass filter, and an impedance matching network. The antenna element is a two-layer microstrip patch structure; the upper patch receives L1 band signals, and the lower patch receives L2 band signals. The two patches are separated by a low-loss dielectric substrate. The LNA amplifies the received weak satellite signals (approximately -130dBm to -110dBm) by about 30dB. The bandpass filter filters out out-of-band interference signals, and the impedance matching network matches the antenna output impedance to 50Ω to match the feed line impedance.
[0128] The GNSS antenna is fixed to the flange at the top of the main pole via an antenna mounting bracket. The mounting bracket is made of aluminum alloy, is disc-shaped, 120mm in diameter, and 5mm thick. The GNSS antenna is secured to the bracket with four M4 screws, with the bottom of the antenna flush against the upper surface of the bracket. The mounting bracket is then fixed to the flange at the top of the main pole with four M8 screws. After installation, the phase center of the GNSS antenna is located on the central axis of the main pole, at a height of 65mm ± 5mm from the top surface of the main pole.
[0129] The GNSS antenna is electrically connected to the GNSS receiver mainboard inside the equipment protective box via a coaxial feeder (i.e., RF cable). The coaxial feeder is an RG-58 type coaxial cable with a characteristic impedance of 50Ω, a length of 3.2m, and SMA type male connectors at both ends. The coaxial feeder extends from the bottom of the GNSS antenna, runs down along the hollow channel inside the main pole to the equipment protective box, and connects to the RF input port of the GNSS receiver mainboard after passing through the protective box. A cable fixing clip is installed every 0.5m inside the main pole to prevent the feeder from shaking within the pole.
[0130] The working process of a GNSS antenna is as follows: the antenna element senses the microwave signal transmitted by the GNSS satellite, generating a weak induced electromotive force. This signal is amplified by a low-noise amplifier, filtered by a filter, and then transmitted to the GNSS receiver mainboard via a coaxial feed line. In the entire signal transmission link, the role of the GNSS antenna is to convert the electromagnetic wave signal propagating in space into an electrical signal and perform preliminary processing, providing high-quality radio frequency input for the receiver's subsequent signal acquisition, tracking, and demodulation.
[0131] The GNSS antenna is installed at the very top of the main pole, above all other components (weather sensor assembly, solar panel, etc.) to ensure an unobstructed 360° field of view for satellite signal reception. Geometric calculations ensure that the vertical distance between the phase center of the GNSS antenna and the upper edge of the protective housing of the weather sensor assembly is no less than 0.5m, and the vertical distance between the phase center and the upper edge of the solar panel is no less than 1.0m. This distance guarantees that the weather sensor assembly and solar panel will not obstruct the GNSS antenna's reception of satellite signals at elevation angles greater than 5°.
[0132] 3. Meteorological sensor components The meteorological sensor assembly, serving as the environmental sensing element at the data acquisition end, is fixedly mounted in the middle of the main pole. It is used to collect real-time measured meteorological parameters of the near-surface atmosphere at the station location. The meteorological sensor assembly is positioned 1.5 meters above the ground, a height within the near-surface atmospheric mixing layer, which represents the atmospheric conditions at the station location while protecting it from direct influences from ground heat radiation and humidity.
[0133] The meteorological sensor assembly includes a barometric pressure sensor, a temperature sensor, and a humidity sensor, all integrated onto a single PCB circuit board and encapsulated within a louvered protective enclosure. The enclosure is made of ABS engineering plastic, cylindrical in shape, with a diameter of 120mm and a height of 150mm. Annular louvered ventilation openings are located on the walls, with the openings tilted downwards at a 45° angle. This ensures adequate ventilation between the enclosure and the outside atmosphere while preventing rain, snow, and sand from directly entering the enclosure.
[0134] The barometric pressure sensor is a MEMS piezoresistive barometric pressure sensor chip, model BMP390, with a measurement range of 300 hPa to 1100 hPa, a measurement accuracy of ±0.3 hPa, a resolution of ±0.01 hPa, and a response time of less than 10 ms. The sensing element of the barometric pressure sensor is a silicon-based piezoresistive Wheatstone bridge. When the external air pressure changes, the silicon diaphragm deforms, causing an imbalance in the bridge resistance and outputting a differential voltage signal, which is converted into a digital barometric pressure value by an ADC.
[0135] The temperature sensor is a platinum resistance temperature sensor chip, model PT1000, with a measurement range of -40℃ to +85℃, a measurement accuracy of ±0.2℃, and a resolution of ±0.01℃. The sensing element of the temperature sensor is a platinum metal thin-film resistor, whose resistance value changes linearly with temperature, and is converted into a digital temperature value by an ADC.
[0136] The humidity sensor is a capacitive humidity sensing chip, model SHT31, with a measurement range of 0% to 100%RH, a measurement accuracy of ±2%RH, and a resolution of ±0.1%RH. The sensing element of the humidity sensor is a polymer capacitor, whose dielectric constant changes with the ambient humidity, causing a change in capacitance value, which is converted into a digital humidity value by an ADC.
[0137] The protective enclosure for the weather sensor assembly is secured to the side of the main pole using stainless steel clamps and an L-shaped bracket. The L-shaped bracket is made of 304 stainless steel, 2mm thick, with a horizontal arm length of 150mm and a vertical arm length of 200mm. The protective enclosure is fixed to the horizontal arm of the L-shaped bracket with four M4 screws, with the bottom of the enclosure flush against the upper surface of the horizontal arm. After installation, the horizontal distance between the center of the protective enclosure and the central axis of the main pole is 180mm, ensuring that the protective enclosure extends beyond the surface of the main pole and provides good ventilation.
[0138] The weather sensor assembly is electrically connected to the GNSS receiver motherboard inside the equipment's protective enclosure via a multi-core shielded data cable. The multi-core shielded data cable is a 6-core AWG26 shielded cable, 2.0m in length. One end of the cable has an aviation plug that mates with the connector on the weather sensor assembly's PCB board, while the other end has pins that connect to the I2C or UART interface on the GNSS receiver motherboard. The data cable is laid along the outer wall of the main pole, secured to the pole with a cable tie every 0.3m. The data cable is protected by a PVC corrugated tubing.
[0139] The meteorological sensor assembly operates as follows: a barometric pressure sensor senses external atmospheric pressure, a temperature sensor senses external atmospheric temperature, and a humidity sensor senses external atmospheric relative humidity. These three sensors transmit the collected pressure, temperature, and humidity data as digital signals to the GNSS receiver motherboard via an I2C or UART digital interface. During signal transmission, the meteorological sensor assembly converts the physical atmospheric parameters (pressure, temperature, and humidity) at the station location into digitized electrical signals, providing measured meteorological data for tropospheric delay calculations and differential barometric altimetry at the monitoring end.
[0140] The barometric pressure sensor is installed 1.5m above the ground. This height avoids the area with the largest temperature and humidity gradients in the near-surface layer (0-0.5m) and also avoids areas below 1.0m that may be affected by surface water and vegetation transpiration, ensuring the representativeness of meteorological parameter measurements. The temperature and humidity sensors are housed in the same protective enclosure as the barometric pressure sensor, and all three measure the atmospheric conditions at the same location, ensuring spatial consistency of meteorological parameters.
[0141] 4. Equipment protective box The equipment protective box, the core electrical integrated component of the acquisition end, is fixedly installed at the lower part of the main pole. It houses the GNSS receiver motherboard, data transmission module, battery, and optional solar charging controller and rain gauge. The protective box is positioned 0.5m above the ground, a height that facilitates routine inspections and troubleshooting by maintenance personnel while preventing ground water from seeping into the box.
[0142] The equipment enclosure is a waterproof and dustproof sealed box made of 304 stainless steel with a thickness of 1.5mm. Its dimensions are 400mm (width) × 300mm (height) × 200mm (depth). The enclosure door is equipped with a waterproof sealing strip, and the door lock is a stainless steel waterproof latch, achieving an IP65 protection rating. The bottom of the enclosure has four waterproof cable connectors for introducing coaxial feeders, meteorological sensor data cables, solar panel photovoltaic cables, and rain gauge data cables.
[0143] The equipment protective box is fixed to the side of the main pole using stainless steel clamps and a mounting backplate. The mounting backplate is made of 304 stainless steel, 2mm thick, and is fixed to the back of the equipment protective box with four M8 screws. The mounting backplate is then tightly secured to the main pole using two U-shaped clamps, with a clamping torque of 40N·m to 60N·m. After installation, the horizontal distance between the rear wall of the equipment protective box and the central axis of the main pole is 120mm.
[0144] The equipment protective enclosure contains a mounting base plate, made of 3mm thick epoxy resin insulating board, which is fixed to the rear wall of the enclosure with screws. The GNSS receiver mainboard, data transmission module, solar charge controller, and rain gauge data acquisition board are each fixed to the mounting base plate with nylon studs. The battery is placed at the bottom of the enclosure, with a 5mm thick rubber shock-absorbing pad between it and the mounting base plate.
[0145] The internal layout of the equipment protective box is as follows: The battery, weighing approximately 25kg, is located at the center of the bottom of the box to maintain the stability of the entire pole-mounted equipment. The GNSS receiver mainboard is located at the upper left of the box, 30mm from the top wall and 20mm from the left side wall. The data transmission module is located at the upper right of the box, 30mm from the top wall and 20mm from the right side wall. The solar charging controller is located below the GNSS receiver mainboard, 20mm from the mainboard mounting studs. Air gaps of at least 10mm are maintained between each PCB board to ensure heat dissipation and ventilation. The layout of the mounting base, PCB boards, and battery has been optimized so that the overall center of gravity of all components inside the protective box is located near the geometric center of the box, preventing eccentric moments under wind loads.
[0146] 4.1 GNSS Receiver Mainboard The GNSS receiver motherboard is the core component for signal processing and control at the acquisition end. It is used to process the radio frequency signals transmitted from the GNSS antenna, generate raw GNSS observation data, and receive meteorological data transmitted from the meteorological sensor components. It performs time synchronization and packet processing on the two types of data.
[0147] The GNSS receiver motherboard is a PCB board integrating a GNSS baseband signal processing chip, an ARM processor, memory, and power management circuitry. It measures 100mm × 80mm and is 1.6mm thick. The baseband signal processing chip is a multi-system, multi-frequency GNSS baseband processor capable of simultaneously acquiring, tracking, and demodulating L1 / L2 dual-frequency signals from four systems: GPS, BeiDou, GLONASS, and Galileo. The ARM processor is a Cortex-M7 core with a 400MHz clock speed, used to run signal processing firmware, data packetization protocols, and communication control programs. Memory includes 512MB of DDR3 RAM and 8GB of eMMC flash memory for storing firmware programs and data cache.
[0148] The RF input port of the GNSS receiver motherboard is electrically connected to the GNSS antenna via a coaxial feed line to receive the RF signal transmitted from the GNSS antenna. The baseband signal processing chip down-converts the input RF signal, converting the L1 and L2 band RF signals into intermediate frequency signals, which are then sampled by an ADC and sent to the digital signal processing unit. The digital signal processing unit completes satellite signal acquisition, carrier tracking, and code tracking, outputting raw observation data such as pseudorange observations (in meters), carrier phase observations (in cycles), Doppler shift (in Hz), and signal-to-noise ratio (in dB-Hz).
[0149] The I2C interface of the GNSS receiver motherboard is electrically connected to the meteorological sensor components via a multi-core shielded data cable to receive digital signals of air pressure, temperature, and humidity. The ARM processor reads the meteorological data according to a preset sampling sequence (synchronized with the sampling of GNSS observations) and associates and packages the meteorological data with the GNSS observations using the same timestamp.
[0150] The UART interface of the GNSS receiver motherboard is electrically connected to the data transmission module via a ribbon cable, and sends the time-synchronized data packets (containing the original GNSS observation values and measured meteorological parameters) to the data transmission module according to the preset data frame format.
[0151] The power input port of the GNSS receiver motherboard is electrically connected to the battery via a power cable. The power management circuit on the motherboard converts the voltage to the operating voltage (+3.3V, +1.8V, +1.2V) required by each chip, thus powering the entire motherboard. The power management circuit includes overvoltage protection, reverse connection protection, and surge suppression circuitry to protect the motherboard from damage in case of abnormal power supply.
[0152] 4.2 Data Transmission Module The data transmission module is the communication element of the acquisition end, used to send the data packets output by the GNSS receiver motherboard to the monitoring end via wireless or wired means.
[0153] The data transmission module is a 4G / 5G wireless communication module, model EC200A, supporting LTE Cat.4 with a downlink speed of 150Mbps and an uplink speed of 50Mbps. The module features a SIM card slot (supporting nano-SIM cards), an antenna interface (for external 4G antennas), and a UART data interface. The data transmission module is electrically connected to the GNSS receiver motherboard via the UART serial interface, receiving data frames sent by the GNSS receiver motherboard. The data transmission module encapsulates the received data frames according to MQTT or TCP / IP protocols and transmits them via the 4G / 5G wireless network to the monitoring terminal's fixed IP address and port number.
[0154] The power input port of the data transmission module is electrically connected to the battery via a power cord, operating at 12V DC with a typical operating current of 200mA. In areas without 4G / 5G signals, the data transmission module can be replaced with a BeiDou short message communication module or a fiber optic communication module.
[0155] 4.3 Storage batteries The battery is an energy storage element at the acquisition end, used to store the electrical energy generated by the solar panel and to provide a stable DC power supply for power-consuming modules such as the GNSS receiver motherboard, data transmission module, and meteorological sensor components.
[0156] The battery is a lithium iron phosphate battery pack with a rated voltage of 12.8V, a rated capacity of 200Ah, and an energy of 2560Wh. The battery consists of 8 individual lithium iron phosphate batteries (3.2V / 200Ah) connected in series, and has a built-in BMS battery management system to monitor the voltage, current, and temperature of each individual battery, and to achieve equal charging, overcharge protection, over-discharge protection, overcurrent protection, and temperature protection.
[0157] The battery measures 260mm × 168mm × 210mm and weighs approximately 26kg. It is placed at the bottom center inside the equipment protective enclosure, with a 5mm thick rubber shock-absorbing pad between the battery and the enclosure bottom to cushion vibrations and impacts during transportation and installation. A gap of at least 30mm is maintained between the battery and the enclosure side walls to facilitate heat dissipation and airflow.
[0158] The battery is electrically connected to the positive and negative output terminals of the solar charge controller via two 16mm² copper core flexible cables, forming a charging circuit. The battery is also electrically connected to the positive and negative power input terminals of the GNSS receiver mainboard and the data transmission module via two 14mm² copper core flexible cables, forming a discharging circuit. A 20A fuse is connected in series on the positive cable for short-circuit protection.
[0159] 4.4 Solar Charge Controller The solar charge controller is a charging management element at the data acquisition end. It is electrically connected between the solar panel and the battery and is used to manage the charging process of the battery by the solar panel.
[0160] The solar charge controller is an MPPT type controller with a rated input power of 300W, a rated input voltage of 18V~24V DC, a rated output voltage of 12V DC, and a rated charging current of 20A. The controller integrates an MPPT algorithm, which dynamically adjusts the equivalent load impedance by monitoring the output voltage and current of the solar panel in real time, ensuring the solar panel always operates at its maximum power point (MPP) and improving charging efficiency.
[0161] The input terminal of the solar charge controller is electrically connected to the junction box of the solar panel via two 4mm² photovoltaic cables, and the output terminal is electrically connected to the positive and negative terminals of the battery via two 16mm² copper core flexible cables. The communication interface of the solar charge controller is electrically connected to the UART interface of the GNSS receiver motherboard via a ribbon cable, sending charging status information (charging current, charging voltage, battery level, etc.) to the GNSS receiver motherboard, which is then uploaded to the monitoring terminal by the data transmission module.
[0162] 4.5 Rain gauge Rain gauges are optional environmental sensing elements at the data acquisition end, used to collect rainfall data at the station location, providing data support for landslide meteorological factor analysis.
[0163] The rain gauge is a tipping bucket rain gauge with a resolution of 0.2 mm, a measuring range of 0~240 mm / h, and an error of ±2%. The rain gauge's inlet diameter is 200 mm, and it is installed on the top of the equipment protective box or separately on the side of the main pole at a height of 0.8 m above the ground. Rainwater is guided out through a PVC drain pipe. The rain gauge's signal output is a reed switch pulse signal, outputting one pulse for each tipping of the bucket (corresponding to 0.2 mm of rainfall). This pulse is electrically connected to the digital input port of the GNSS receiver's mainboard via a signal cable. The GNSS receiver's mainboard counts the pulse signals, converts them into rainfall data, and uploads them along with other data.
[0164] 5. Solar panels The solar panel is the power generation element at the data collection end. It is fixedly installed on the side of the main pole and is used to convert solar radiation energy into DC power, which is then used to charge the battery through the solar charge controller.
[0165] The solar panel is a monocrystalline silicon photovoltaic panel with a rated power of 120W, an open-circuit voltage of 21.6V, a short-circuit current of 7.3A, a maximum power point voltage of 17.8V, and a maximum power point current of 6.7A. The panel's dimensions are 1200mm × 540mm × 30mm, and it weighs approximately 9kg. An aluminum alloy frame and a junction box are located on the back of the panel, with positive and negative terminals leading out from the junction box.
[0166] The solar panels are fixed to the side of the main pole using aluminum alloy mounting brackets. The mounting brackets are made of 304 stainless steel and include a pair of triangular support frames and a horizontal beam. The vertical sides of the triangular support frames are fixed to the main pole using two U-shaped clamps, while the horizontal sides support the lower edge of the photovoltaic panels. The upper edge of the photovoltaic panels is fixed to the top beam of the mounting bracket using clamping blocks. The tilt angle between the photovoltaic panels and the horizontal plane is determined based on the latitude of the measuring station. In the Northern Hemisphere, it is set to the latitude angle plus 10° to 15°. In this embodiment, the station latitude is 28.5°, and the tilt angle is set to 40°. The photovoltaic panels are oriented due south (in the Northern Hemisphere) to obtain the maximum annual solar radiation reception.
[0167] The junction box of the solar panel is electrically connected to the input terminal of the solar charge controller inside the equipment protective box via two 4mm² photovoltaic cables. The photovoltaic cables are 2.5m long, laid along the outer wall of the main pole, protected by a PVC corrugated pipe, and connected to the input terminal of the charge controller after passing through the equipment protective box.
[0168] After installation, the lower edge of the solar panel is 1.2m above the ground, and the upper edge is 2.1m above the ground. The vertical distance between the upper edge of the solar panel and the bottom of the GNSS antenna is 1.4m, ensuring that the solar panel will not obstruct the GNSS antenna's satellite signal reception at elevation angles above 5°. The solar panel is installed on the south side of the main pole (in the Northern Hemisphere), and the GNSS antenna's signal reception direction is towards the zenith; the spatial relationship between the two is non-interfering.
[0169] 6. The operational relationships and signal flow between components The operational relationships and signal flows between the various components at the acquisition end are as follows: Signal flow direction one (GNSS signal link): Microwave signal transmitted by GNSS satellite → GNSS antenna receives and converts into electrical signal → Coaxial feeder transmission → GNSS receiver motherboard RF input → Baseband processing generates pseudorange and carrier phase observation data.
[0170] Signal flow direction two (meteorological signal link): Atmospheric pressure, temperature, and humidity at the station location → sensing by barometric pressure sensor, temperature sensor, and humidity sensor → transmission via multi-core shielded data cable → GNSS receiver motherboard I2C interface → generation of digital data for barometric pressure, temperature, and humidity.
[0171] Signal flow direction three (data output link): The GNSS receiver motherboard packages the above two types of data according to the timestamp → UART interface → data transmission module → 4G / 5G wireless network → monitoring end.
[0172] Signal flow direction four (power supply link): Solar panel photovoltaic power generation → Photovoltaic cable → Solar charge controller → Battery charging → Battery via power line → Various power modules (GNSS receiver motherboard, data transmission module, weather sensor components).
[0173] Energy flow direction: solar radiation energy → solar panel (light → electricity) → solar charge controller (voltage conversion) → storage battery (chemical energy storage) → various power modules (electrical energy → various forms of energy required for equipment operation: radio frequency signal processing, digital computing, wireless transmission, etc.).
[0174] 7. Mathematical Relationships and Optimization of Structural Dimensions The mounting height and spacing of each component at the acquisition end have been optimized based on structural mechanics and electromagnetic compatibility to meet the following dimensional constraints: Constraint Relationship 1 (Overturning Stability): The total height of the H-rod is ≤4.0m, the total center of gravity height of the equipment H-center of gravity is ≤1.8m, and the foundation depth D-foundation is ≥0.5m, ensuring that the equipment will not overturn under the worst wind load conditions (wind speed 25m / s).
[0175] Constraint 2 (Signal Reception Field of View): The vertical distance ΔH between the phase center of the GNSS antenna and the edge of the solar panel is ≥0.8m, ensuring no obstruction within a range of 5° above the antenna cutoff height angle.
[0176] Constraint Relationship 3 (Representativeness of Meteorological Parameters): The installation height Hmeteorological of the meteorological sensor assembly is [1.2m, 2.0m]. This height range corresponds to the near-surface atmospheric constant flux layer, ensuring the representativeness of air pressure, temperature, and humidity measurements.
[0177] Constraint Relationship 4 (Equipment Protection and Maintenance): The installation height H of the equipment protection box is ∈ [0.3m, 0.8m]. This height allows maintenance personnel to conduct daily inspections and troubleshooting without having to climb to a height. At the same time, the bottom of the box is higher than the designed ground water level.
[0178] Constraint Relationship 5 (Structural Mechanics): The slenderness ratio of the main column λ = L / r ≤ 150 (where L is the length of the main column and r is the radius of gyration), ensuring that the main column does not become unstable and bend under wind load and its own weight.
[0179] The aforementioned dimensional constraints collectively ensure the structural safety, signal reception quality, meteorological measurement representativeness, and maintenance convenience of the acquisition terminal in the field environment.
[0180] Application Example 1: Application of Landslide Monitoring Scenarios at a Large Hydropower Station A large hydropower station is located in the southwestern mountainous region, with several steep slopes along its reservoir bank. The HP-07 landslide is located approximately 3.2 km upstream of the dam. The leading edge of the landslide is adjacent to the normal water level of the reservoir, while the trailing edge is about 220 m higher than the leading edge, with a horizontal distance of about 280 m. The elevation difference between the base station and the rover station is approximately 220 m, representing a typical short-distance landslide monitoring scenario with a large elevation difference. The stability of this landslide is directly related to the operational safety of the hydropower station and the flood control safety of downstream towns.
[0181] The data acquisition system of this application was deployed. One base station was set up on the stable bedrock at the rear edge of the landslide, and five rover stations were set up in the middle and front edges of the landslide. The data acquisition parameters of each station are as follows: the main pole is 3.5m high, 76mm in diameter, and 4mm thick, made of hot-dip galvanized steel pipe; the GNSS antenna is a dual-frequency, four-system, full-band antenna; the meteorological sensor assembly is installed 1.5m above the ground, including a barometric pressure sensor, a temperature sensor, and a humidity sensor; the equipment protective box is installed 0.5m above the ground, containing a GNSS receiver motherboard, a 4G data transmission module, a 200Ah lithium iron phosphate battery, and an MPPT solar charging controller; the solar panels are two 120W monocrystalline silicon photovoltaic panels, installed at a 40° angle towards due south. After all equipment is installed and debugged, the base station data and rover data are transmitted to the monitoring station in real time via a 4G network.
[0182] The monitoring terminal is deployed in the hydropower station's machine room, configured with dual Intel Xeon Gold 6248R processors, 256GB of memory, a 4TB RAID10 storage array, and runs the Ubuntu Server operating system and the landslide monitoring and location calculation software of this application.
[0183] Continuous observation data from July 15 to August 15, 2026, was selected for calculation verification. This period coincided with the local rainy season, during which there were seven thunderstorms and heavy rainfall events, resulting in drastic changes in atmospheric conditions and posing a severe test to the stability of the positioning algorithm.
[0184] The monitoring terminal performs calculations using the method described in this application. The key parameters are set as follows: satellite cutoff elevation angle 10°, GNSS observation sampling rate 1Hz, meteorological data sampling rate 0.2Hz, tropospheric delay model using Saastamoinen model + NMF mapping function, differential barometric altimeter humidity correction coefficient α of 0.012, weighted least squares convergence threshold of 1mm, and elevation angle weighting using sin²(E) model.
[0185] The solution results are shown in Table 1: Table 1. Comparison of positioning accuracy between the method in this application and conventional RTK methods. As shown in Table 1, after adopting the method of this application, the average integer ambiguity fixation rate of the five rover stations (rover stations) reached 96.8% throughout the entire observation period, with a minimum of 92.3% and a maximum of 99.1%; while the average integer ambiguity fixation rate of the conventional RTK method was only 61.7%. The method of this application improved the ambiguity fixation rate by approximately 35 percentage points. The average vertical positioning error was 1.7 cm, while that of the conventional RTK method was 7.6 cm, and the method of this application improved the vertical positioning accuracy by approximately 77%. In particular, during a heavy rainfall event on July 22, the conventional RTK method failed to fix the integer ambiguity for 8 consecutive hours; while the method of this application maintained a stable ambiguity fixation state throughout, successfully capturing a vertical settlement signal of approximately 3.2 mm from the landslide body during that period.
[0186] Application Example 2: Accuracy Evaluation and Weighting Strategy Implementation of Differential Barometric Altimeter Model To verify the accuracy of the differential barometric altimetry model in the scenario described in this application and to provide prior parameters for its weighting, a 30-day accuracy assessment test was conducted on the HP-07 landslide.
[0187] The experimental setup is as follows: the base station (base station) has an elevation of 1050.000m (known), and the rover station (rover station) has an elevation of approximately 1270.000m (to be evaluated). The horizontal distance between the two stations is approximately 260m, and the elevation difference is approximately 220m. Each station is equipped with a Vaisala PTU310 high-precision barometric pressure / temperature / humidity sensor (barometric pressure accuracy ±0.1hPa, temperature accuracy ±0.1℃, humidity accuracy ±1%RH). Barometric pressure, temperature, and humidity data are collected at 8:00, 12:00, 16:00, and 20:00 daily. The rover station elevation is calculated using a differential barometric altimeter model, and a baseline value for the rover station elevation is obtained using high-precision static GNSS calculation (24-hour static observation, precise ephemeris calculation).
[0188] A total of 120 sets of valid comparison data were obtained over 30 days. The experimental results show that the mean elevation error (calculated value minus the reference value) calculated by the differential barometric altimeter model is -0.12m, the standard deviation σ_h is 0.35m, the maximum positive error is 0.68m, and the maximum negative error is -0.75m. The error distribution basically conforms to a normal distribution (after Shapiro-Wilk normality test, the p-value is greater than 0.05).
[0189] Based on the above prior accuracy assessment results, in the joint solution, the weight P_ellipse of the Earth ellipsoid constraint equation is determined according to the following formula: P_ellipse = σ0² / σ_h² = 1 / (0.35)² ≈ 8.16.
[0190] For GNSS double-difference observations, the elevation angle weighting model of equation (23) is adopted. Taking the satellite elevation angle El=30° as an example, its weight correlation factor is sin²(30°) = 0.25; taking El=60° as an example, its weight correlation factor is sin²(60°) = 0.75.
[0191] To verify the effectiveness of the above weighting strategy, three solution schemes were compared: Scheme A is an equal-weight solution (all observations and constraint equations have a weight of 1), Scheme B is GNSS elevation angle weighting only (ellipsoid constraint weight is 1), and Scheme C is the adaptive weighting strategy of this application (GNSS elevation angle weighting + ellipsoid constraint accuracy weighting). The results of the solution comparison for the same 10-day observation data are as follows: Table 2 Comparison of positioning accuracy for different weighting strategies As shown in Table 2, compared to Scheme A (equal weighting), the adaptive weighting strategy (Scheme C) of this application improves the vertical positioning accuracy by approximately 50% (from 3.2cm to 1.6cm), and by approximately 33% (from 2.4cm to 1.6cm) compared to Scheme B (weighting only elevation angle). This embodiment verifies that the differential barometric altimeter model can stably provide sub-meter level elevation accuracy (σ_h approximately 0.3~0.4m) in short-distance, large-elevation-difference scenarios, and that adaptive weighting based on this accuracy prior information can further improve the vertical positioning accuracy of the joint solution.
[0192] Application Example 3: Reliability Verification of Long-Term Field Operation of Data Acquisition Terminal The five data acquisition units on the HP-07 landslide have been operating continuously for over 18 months since their installation was completed in March 2025. During this period, they have withstood multiple extreme weather conditions, including strong winds (maximum wind speed 22 m / s), torrential rain (maximum daily rainfall 156 mm), high temperatures (maximum temperature 38℃), and low temperatures (minimum temperature -6℃), fully verifying the long-term operational reliability of the equipment in harsh field environments.
[0193] Operational data statistics are as follows: The average online rate (effective data reception time / total operating time) reached 97.6%. Offline time due to communication network failures accounted for 65% of the total offline time, offline time due to equipment failures accounted for only 10%, and offline time due to insufficient power supply accounted for 25% (mainly occurring during periods of continuous rainy weather exceeding 10 days). No water ingress, condensation, or dust accumulation was observed inside the equipment's protective enclosure, indicating excellent sealing performance. Dust accumulation on the surface of the solar panels caused a decrease in power generation efficiency of approximately 15%, but this is still within the design margin. After 18 months of charge-discharge cycles (one charge-discharge cycle per day, totaling approximately 540 cycles), the battery capacity decayed by approximately 8%, still sufficient to meet the power supply needs for 5 consecutive days of rainy weather.
[0194] Equipment Failure Record: During 18 months of operation, the 5 devices experienced a total of 3 failures: the data transmission module of one device was damaged by lightning (and has been replaced); the humidity sensor of one device drifted out of tolerance (and has been calibrated); and the solar charge controller of one device was damaged by overvoltage (and has been replaced). The mean time between failures (MTBF) was approximately 30 months, meeting the reliability requirements for long-term unattended mobile equipment in the field.
[0195] Regarding the ease of installation at the data acquisition end, the deployment time for a single station has been reduced from 2-3 days for split-type equipment to 2-4 hours, significantly reducing on-site construction difficulty and labor costs. All connecting cables between components are concealed inside the main pole or run through conduits along the pole surface, avoiding the problem of exposed and aging cables.
[0196] This embodiment verifies the long-term operational reliability and maintenance-free capability of the data acquisition terminal in harsh field environments, which can meet the requirements of landslide monitoring projects for continuous operation time and reliability of equipment.
[0197] Example 4: Comparative Test of Landslide Monitoring Accuracy under Different Elevation Differences To further verify the applicability and accuracy improvement effect of the proposed method under different elevation differences, five rover stations with different relative elevation differences were selected on the HP-07 landslide. The elevation differences were 50m (control point), 120m, 160m, 200m, and 220m. For each rover station, both the conventional RTK method and the proposed method were used for continuous calculations over a period of 10 days, and the vertical positioning accuracy of the two methods was compared.
[0198] The experimental results are shown in Table 3: Table 3. Comparison of vertical positioning accuracy between the two methods under different elevation differences. As shown in Table 3, when the relative elevation difference is within 100m (50m control point), the conventional RTK method can maintain a relatively good vertical positioning accuracy (1.8cm), while the improvement of the proposed method is limited (16.7%). However, when the relative elevation difference exceeds 100m, the vertical positioning accuracy of the conventional RTK method drops sharply, from 4.2cm at an elevation difference of 120m to 8.5cm at an elevation difference of 220m; while the vertical positioning accuracy of the proposed method remains stable (1.5~1.9cm), with the accuracy improvement increasing from 54.8% to 80.0%. These experimental results fully demonstrate that the proposed method has a significant accuracy improvement effect in short-distance, large elevation difference (>100m) scenarios, and the improvement increases with the increase of the elevation difference.
[0199] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. By employing a method that determines the double-difference satellite distance based on baseline positioning data and mobile positioning data, determines the double-difference tropospheric delay based on baseline meteorological data and mobile meteorological data, corrects the troposphere based on the double-difference satellite distance and double-difference tropospheric delay to obtain double-difference observations, and constructs Earth ellipsoid constraint equations for joint solution, it can effectively compensate for the residual error of tropospheric delay in environments with large elevation differences. At the same time, it increases observation redundancy and improves the satellite's spatial geometry, effectively solving the problems of low vertical positioning accuracy and poor ambiguity fixation rate of conventional RTK in scenarios with large elevation differences. Thus, it achieves the goal of improving the vertical positioning accuracy from 8.5cm to 1.7cm and the integer ambiguity fixation rate from 60%~65% to over 95%.
[0200] 2. By equipping both the base station and the rover with GNSS receivers and meteorological sensors to collect positioning observation data and meteorological measurement data in real time, it is possible to obtain real-time meteorological parameters such as air pressure, temperature and humidity at the station. This can replace the parameters of the standard atmospheric model, effectively solving the problem that the standard atmospheric model cannot accurately describe the real-time atmospheric state of the rover site in scenarios with large elevation differences, and thus achieving accurate compensation for tropospheric delay residual error.
[0201] 3. By adopting a method that uses a differential barometric altimetry model to determine the absolute elevation of the rover station based on the reference pressure, the flowing pressure, and the elevation information of the reference station, and then constructing the Earth ellipsoid constraint equation based on this absolute elevation, it is equivalent to adding a pseudorange observation equation for a virtual satellite at the Earth's center. This effectively solves the problems of poor satellite spatial geometry and weak elevation direction observation intensity in scenarios with large elevation differences, thereby achieving a significant improvement in elevation direction positioning accuracy.
[0202] 4. High system reliability: The data acquisition unit integrates the GNSS receiver, weather sensor, data transmission module, and solar power supply module onto a single pole, avoiding the complex wiring and connections of separate devices and reducing potential faults such as cable aging and poor contact. The average online rate of the equipment reaches over 97.6%, and the average mean time between failures (MTBF) exceeds 30 months.
[0203] 5. Low deployment cost: The data acquisition terminal enables rapid deployment with "pole-to-use", reducing the deployment time of a single station from 2-3 days for split-type equipment to 2-4 hours, which greatly reduces on-site construction costs.
[0204] 6. Strong real-time performance: Each monitoring station transmits data in real time through 4G / 5G networks, and the monitoring end calculates and outputs the positioning results in real time, with a total latency of less than 2 seconds.
[0205] 7. Wide applicability: It is suitable for various short-distance landslide monitoring scenarios with large elevation differences, including high and steep slopes of reservoir banks, deep-cut road slopes, and high and steep mine spoil heaps. When the relative elevation difference exceeds 100m, the accuracy improvement effect becomes more significant as the elevation difference increases.
[0206] In summary, this application effectively solves the technical problems of low vertical positioning accuracy and poor ambiguity fixation rate of conventional RTK methods in short-distance large-elevation-difference landslide monitoring by correcting the synergistic effect of tropospheric delay and differential barometric altimetry ellipsoid constraints through the correction of benchmark meteorological data and mobile meteorological data. It realizes high-precision landslide displacement flow based on meteorological data, and at the same time, it realizes low-cost, high-reliability long-term unattended mobile flow in the field through the acquisition end, which has good technical effects and engineering application prospects.
[0207] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.
Claims
1. A method for monitoring landslides, characterized in that, Applied to the monitoring end, the method includes: The first double-difference observation equation is constructed based on the acquired baseline positioning data and mobile positioning data. The baseline positioning data contains the positioning information of the base station, and the mobile positioning data contains the positioning information of the rover station. The double-difference tropospheric delay is determined based on the acquired baseline meteorological data and mobile meteorological data, wherein the baseline meteorological data includes meteorological parameters of the baseline station and the mobile meteorological data includes meteorological parameters of the rover station. Based on the first double-difference observation equation and the double-difference tropospheric delay correction of the troposphere, the second double-difference observation equation is obtained. Based on the baseline meteorological data and the mobile meteorological data, the absolute elevation of the rover station is determined by the differential barometric altimetry model, and the Earth ellipsoid constraint equation is constructed based on the absolute elevation of the rover station. By jointly solving the second double-difference observation equation and the Earth ellipsoid constraint equation, the three-dimensional coordinate corrections and integer ambiguity floating-point solutions of the rover station are obtained. The integer ambiguity floating-point solution is fixed to obtain the integer ambiguity fixed solution, and the rover's three-dimensional coordinate correction is updated according to the integer ambiguity fixed solution. The three-dimensional coordinates of the rover are corrected based on the updated three-dimensional coordinate correction values to obtain the three-dimensional coordinate positioning data of the rover for consecutive epochs. Based on the three-dimensional coordinate positioning data of the rover station in continuous epochs, a displacement time series is generated for landslide monitoring.
2. The method according to claim 1, characterized in that, The first double-difference observation equation is constructed based on the acquired baseline positioning data and mobile positioning data, including: Obtain reference pseudorange observations and reference carrier phase observations from the reference positioning data, and obtain mobile pseudorange observations and mobile carrier phase observations from the mobile positioning data; A pseudorange single-difference observation equation is constructed based on the reference pseudorange observation value and the mobile pseudorange observation value; a carrier phase single-difference observation equation is constructed based on the reference carrier phase observation value and the mobile carrier phase observation value. The pseudorange single-difference observation equation is processed by inter-satellite double-difference to obtain the first pseudorange double-difference observation equation, and the carrier phase single-difference observation equation is processed by inter-satellite double-difference to obtain the first carrier phase double-difference observation equation.
3. The method according to claim 1, characterized in that, The double-difference tropospheric delay is determined based on the acquired baseline meteorological data and mobile meteorological data, including: Obtain reference air pressure, reference temperature, and reference humidity from reference meteorological data; based on reference air pressure, reference temperature, and reference humidity, determine the oblique tropospheric delay of the reference station to each satellite; The flow pressure, flow temperature, and flow humidity are obtained from the flow meteorological data, and the oblique tropospheric delay of the rover station to each satellite is determined based on the flow pressure, flow temperature, and flow humidity. The oblique tropospheric delay of the base station and the oblique tropospheric delay of the rover station are processed by inter-station differential and inter-satellite differential respectively to obtain the double-difference tropospheric delay.
4. The method according to claim 3, characterized in that, Based on the baseline meteorological data and the mobile meteorological data, the absolute elevation of the rover station is determined using a differential barometric altimetry model, and an Earth ellipsoid constraint equation is constructed based on the absolute elevation of the rover station, including: The baseline air pressure is obtained from the baseline meteorological data, the baseline station elevation is obtained from the baseline positioning data, and the mobile air pressure is obtained from the mobile meteorological data. Based on the baseline air pressure, mobile air pressure, baseline station elevation, average temperature between the baseline temperature and the mobile temperature, and average humidity between the baseline humidity and the mobile humidity, the absolute elevation of the mobile station is determined by the differential barometric altimetry model. Based on the absolute elevation of the rover, an Earth ellipsoid constraint equation is constructed.
5. The method according to claim 4, characterized in that, By jointly solving the second double-difference observation equation and the Earth ellipsoid constraint equation, the three-dimensional coordinate corrections and integer ambiguity floating-point solutions for the rover station are obtained, including: At the approximate location of the rover station, the second double-difference observation equation and the Earth ellipsoid constraint equation are linearized and expanded to obtain the linearized joint observation equation; The weights of GNSS observations are determined based on the prior variance definition of double-difference observations, and the weights of the Earth ellipsoid constraint equations are determined based on the prior accuracy of the differential barometric altimetry model, thus constructing a joint weight matrix. The weighted least squares method is used to solve the linearized joint observation equations to obtain the three-dimensional coordinate corrections and integer ambiguity floating-point solutions of the rover station, and the covariance matrix of the parameter vector to be estimated is calculated.
6. The method according to claim 1, characterized in that, The weighted least squares method is used to solve the linearized joint observation equations, obtaining the rover's three-dimensional coordinate corrections and integer ambiguity floating-point solutions. The covariance matrix of the parameter vector to be estimated is then calculated, including: Construct the normal equations based on the linearized joint observation equations; Solve the normal equations to obtain the estimated values of the parameter vector to be estimated; Determine whether the 3D coordinate correction of the rover obtained in a certain iteration is less than the preset convergence threshold; If the value is less than the preset convergence threshold, it is determined that the iteration has converged, and the current three-dimensional coordinate correction of the rover and the integer ambiguity floating-point solution are output. If the value is not less than the preset convergence threshold, the approximate position of the rover is updated based on the three-dimensional coordinate correction value of the rover obtained in this iteration, and the linearization expansion step is returned to be executed until the convergence condition is met.
7. The method according to claim 6, characterized in that, Based on the corrected double-difference observations and the Earth ellipsoid constraint equations, a joint solution is performed to obtain the rover's three-dimensional coordinate corrections and integer ambiguity floating-point solutions, including: At the approximate location of the rover station, the corrected double-difference observations and the Earth ellipsoid constraint equations are linearized and expanded to obtain the linearized joint observation equations. The weights of the GNSS double-difference observations are determined based on the elevation angles of each satellite, and the weights of the Earth ellipsoid constraint equations are determined based on the prior accuracy of the differential barometric altimetry model, thus constructing a joint weight matrix. The linearized joint observation equations are solved using the weighted least squares method to obtain the three-dimensional coordinate corrections and integer ambiguity floating-point solutions for the rover.
8. The method according to claim 1, characterized in that, Based on continuous epochs of three-dimensional coordinate positioning data from the rover station, a displacement time series is generated for landslide monitoring, including: Using the three-dimensional coordinate positioning data at the time of the first observation as the initial coordinates, the three-dimensional displacement of each epoch relative to the initial coordinates is calculated to form a displacement time series; Calculate the displacement rate based on the displacement time series; When the displacement or displacement rate exceeds the preset displacement warning threshold, a landslide warning message is sent to the user terminal.
9. A monitoring terminal, characterized in that, include: One or more processors; A memory coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the monitoring terminal to perform the method as described in any one of claims 1 to 8.
10. A system for monitoring landslides, characterized in that, The system includes: At least two acquisition terminals are provided. One of the acquisition terminals is located at a base station and is used to collect base positioning data and base meteorological data and send them to the monitoring terminal. The base positioning data includes the positioning information of the base station, and the base meteorological data includes the meteorological parameters of the base station. The other of the acquisition terminals is located at a rover station and is used to collect rover positioning data and rover meteorological data and send them to the monitoring terminal. The rover positioning data includes the positioning information of the rover station, and the rover meteorological data includes the meteorological parameters of the rover station. The monitoring terminal is used to receive the reference positioning data, the reference meteorological data, the mobile positioning data, and the mobile meteorological data, and to execute the method as described in any one of claims 1 to 8 to conduct landslide displacement flow; the user terminal is used to receive and display landslide warning information, wherein the landslide warning information is information sent by the monitoring terminal to the user terminal when the displacement amount or displacement rate exceeds a preset displacement warning threshold.
