High-altitude large-altitude-difference IMU / DGNSS-assisted aerial survey aerial triangulation encryption method and system
By employing IMU/DGNSS-assisted differential positioning and signal optimization techniques, the problems of accuracy and reliability in aerial triangulation of aerial surveys in high-altitude and high-elevation environments were solved, resulting in high-precision mapping results.
Patent Information
- Application Number
- CN202511986891.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-26
- Publication Date
- 2026-03-31
AI Technical Summary
In high-altitude environments with large elevation differences, existing technologies struggle to effectively address multiple interferences such as atmospheric density fluctuations, refractive index anomalies, and terrain reflections, resulting in low accuracy and poor reliability of aerial triangulation.
The method employs IMU/DGNSS assistance, uses differential positioning technology to initially correct the signal path, combines atmospheric change characteristics and terrain reflection interference patterns to perform signal smoothing and adaptive adjustment, optimizes carrier signal stability, and ensures the accuracy of mapping coordinates through iterative verification.
It significantly improves the accuracy and reliability of surveying in complex terrain at high altitudes, ensuring high-precision aerial triangulation results within a small area and meeting the surveying needs of high-altitude environments.
Smart Images
Figure CN121763326A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of information processing technology, specifically to a method and system for aerial triangulation densification of aerial surveys with IMU / DGNSS assistance at high altitudes and large elevation differences. Background Technology
[0002] In the field of aerial surveying, aerial triangulation technology for high-altitude environments with significant elevation differences has crucial strategic value. It directly relates to the ability to accurately survey and acquire data in complex terrain areas, and is a vital technical support for land resource surveys, disaster early warning, and infrastructure construction. Especially in high-altitude regions, surveying tasks often face the challenges of extreme natural conditions, requiring extremely high reliability and accuracy of technical methods.
[0003] Existing methods often exhibit significant shortcomings when dealing with high-altitude environments with large elevation differences due to insufficient adaptability to complex environments. Many traditional techniques focus more on the mapping needs of flat or low-altitude areas, lacking the ability to dynamically respond to changes in the unique environment of high altitudes. Especially when faced with drastic terrain undulations and variable environmental interference, positioning accuracy and data reliability are difficult to guarantee. This often leads to data deviations or even data failures when conducting mapping tasks in areas such as high mountains and canyons.
[0004] Due to significant differences in altitude, the atmospheric density and refractive properties vary considerably at different altitudes. These differences directly interfere with the propagation path of positioning signals, causing unpredictable deviations during transmission. This signal interference leads to another critical issue: abnormal fluctuations in pseudorange and carrier signals within the positioning data. These fluctuations are particularly pronounced in areas with large elevation differences. For example, during an aerial survey mission in a high-altitude region, as the aircraft flew from a valley to a mountaintop, the positioning data frequently jumped due to the combined effects of atmospheric refraction and terrain reflection. This directly resulted in discrepancies between the survey results and the actual terrain, severely impacting subsequent data processing and applications. Summary of the Invention
[0005] The purpose of this invention is to provide a method and system for aerial triangulation densification in high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial surveying, effectively solving the technical problems of low accuracy and poor reliability of aerial triangulation densification in high-altitude, large-elevation-difference environments caused by multiple interferences such as atmospheric density fluctuations, refractive index anomalies, and terrain reflection. To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: This invention provides a method for aerial triangulation densification in high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial surveying, mainly including: Step 1: Collect data on atmospheric density fluctuation amplitude and refractive index anomaly changes in high-altitude environments with large elevation differences. Combine this with the propagation delay deviation of the real-time positioning signal and use differential positioning technology to make preliminary corrections to the signal path, thereby obtaining the initially adjusted pseudorange data and carrier signal data. Step 2: Based on the initially adjusted pseudorange data and carrier signal data, extract the temperature gradient influence factor and humidity distribution unevenness characteristics caused by atmospheric changes, and perform signal fluctuation smoothing processing on the intensity of air pressure change interference to obtain a smoothed signal sequence. Step 3: If there is residual fluctuation in the smoothed signal sequence that exceeds a preset threshold, it is determined to be turbulence interference caused by terrain reflection. The signal path curvature data is obtained, the specific pattern of interference is classified, and the classified interference category is obtained. Step 4: Based on the classified interference categories, obtain the carrier phase offset and signal amplitude attenuation characteristics under the corresponding elevation difference terrain, and adaptively adjust the multipath interference degree and terrain reflection delay time to obtain the corrected pseudorange data. Step 5: Based on the corrected pseudorange data, extract the frequency drift rate and signal-to-noise ratio reduction features from the carrier signal, and combine them with the carrier period jump frequency and signal receiving angle deviation data to optimize the carrier signal stability and obtain the optimized carrier signal data. Step 6: By combining the optimized carrier signal data with the corrected pseudorange data, the mapping coordinate deviation in the high-altitude environment is fused and corrected to obtain the corrected mapping coordinate result; Step 7: Based on the corrected mapping coordinates, identify potential data failure areas; automatically mark failure points by calculating the grid point signal coverage (coverage <75% is considered failure) and combining it with the residual fluctuation threshold (>0.05 meters). Using iterative verification technology, perform multiple rounds of comparison between the corrected coordinates and the high-precision control point database (matching success rate >90% and average deviation <0.01 meters is considered passing), obtaining verified high-precision, high-altitude, small-scale aerial triangulation results.
[0006] In addition, this invention provides an aerial survey triangulation system assisted by IMU / DGNSS for high-altitude, large-elevation-difference aerial surveys, mainly comprising: The data acquisition and preliminary correction module is used to acquire atmospheric density fluctuation amplitude and refractive index anomaly change data in high-altitude and large-elevation-difference environments. Combined with the propagation delay deviation of real-time positioning signals, differential positioning technology is used to perform preliminary correction of the signal path to obtain initially adjusted pseudorange data and carrier signal data. The signal smoothing module is used to extract the temperature gradient influence factor and humidity distribution unevenness caused by atmospheric changes based on the initially adjusted pseudorange data and carrier signal data, and to perform signal fluctuation smoothing processing on the intensity of air pressure change interference to obtain a smoothed signal sequence. The interference mode classification module is used to determine that if there is residual fluctuation in the smoothed signal sequence that exceeds a preset threshold, it is turbulence effect interference caused by terrain reflection, obtain signal path curvature data, classify the specific mode of interference, and obtain the classified interference category. The pseudorange correction module is used to obtain the carrier phase offset and signal amplitude attenuation characteristics under the corresponding elevation difference terrain based on the classified interference category, and to adaptively adjust the multipath interference degree and terrain reflection delay time to obtain the corrected pseudorange data. The carrier signal optimization module is used to extract the frequency drift rate and signal-to-noise ratio reduction characteristics in the carrier signal based on the corrected pseudorange data, and combine the carrier period jump frequency and signal receiving angle deviation data to optimize the carrier signal stability and obtain optimized carrier signal data. The coordinate fusion correction module is used to combine the optimized carrier signal data with the corrected pseudorange data to perform fusion correction on the mapping coordinate deviation in high-altitude environments, and obtain the corrected mapping coordinate results. The results verification module is used to identify potential data failure areas based on the corrected survey coordinates. It automatically marks failure points by calculating the grid point signal coverage (coverage <75% is considered failure) and combining this with the residual fluctuation threshold (>0.05 meters). Using iterative verification technology, the corrected coordinates are compared multiple times with the high-precision control point database (a matching success rate >90% and an average deviation <0.01 meters are considered passing), resulting in high-precision, high-altitude, small-scale aerial triangulation results that have passed verification.
[0007] Compared with the prior art, the present invention has the following beneficial effects: This invention aims to solve the problem of interference from complex terrain and atmospheric environment on positioning signal propagation, especially the signal deviation problem caused by the coupled influence of multiple factors such as atmospheric density fluctuations, refractive index anomalies, terrain reflection turbulence effects, and multipath effects. This invention performs preliminary signal correction by collecting environmental data and combining it with differential positioning technology, extracting temperature gradient and humidity distribution characteristics to smooth signal fluctuations, classifying terrain reflection interference and adaptively adjusting pseudorange data, optimizing carrier signal stability, and finally fusing and correcting mapping coordinates. Iterative verification ensures accuracy and reliability. This invention comprehensively considers the multidimensional interference factors in high-altitude environments and constructs a full-process adaptive optimization mechanism from signal acquisition to coordinate correction, significantly improving the accuracy and reliability of small-scale aerial triangulation results, and providing technical support for accurate mapping in complex high-altitude terrain. Attached Figure Description
[0008] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.
[0009] Figure 1 This is the overall flowchart of the present invention.
[0010] Figure 2 This is a detailed flowchart of the signal fluctuation smoothing process of the present invention.
[0011] Figure 3 This is a detailed flowchart of the interference classification judgment of the present invention. Detailed Implementation
[0012] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the embodiments of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.
[0013] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.
[0014] Example 1: See Figures 1-3 This embodiment discloses a method for aerial triangulation densification of aerial surveys assisted by IMU / DGNSS at high altitudes with large elevation differences, specifically including: Step 1: Collect data on atmospheric density fluctuation amplitude and refractive index anomaly changes in high-altitude environments with large elevation differences. Combine this with the propagation delay deviation of the real-time positioning signal and use differential positioning technology to preliminarily correct the signal path, obtaining the initially adjusted pseudorange data and carrier signal data.
[0015] By collecting atmospheric density fluctuation data in high-altitude environments with large elevation differences and combining it with refractive index anomaly data, an initial environmental impact dataset is constructed, resulting in a preliminary set of environmental parameters. Based on this preliminary set, the correlation between atmospheric density fluctuation amplitude and refractive index anomalies is analyzed. Anomalies are filtered using a preset threshold to determine the filtered environmental impact dataset. For this filtered dataset, propagation delay deviation data of the positioning signal is integrated, and differential positioning technology is applied to correct the signal path, obtaining a corrected signal path dataset. Using this corrected dataset, the pseudorange data deviation is calculated. If the deviation exceeds a preset threshold, the pseudorange data undergoes secondary correction, resulting in an adjusted pseudorange dataset. Based on this adjusted pseudorange dataset, the phase offset of the carrier signal data is extracted, and Kalman filtering is used to smooth the phase offset, determining the smoothed carrier signal dataset. Finally, the smoothed carrier signal dataset, combined with the adjusted pseudorange dataset, is used to comprehensively analyze the propagation delay of the positioning signal to determine if residual deviations exist, resulting in the final corrected positioning signal dataset. By using the final positioning signal correction dataset, high-precision positioning result data is generated, thus completing the optimization processing of positioning signals in high-altitude and large-elevation-difference environments.
[0016] In practical implementation, when collecting data on atmospheric density fluctuation amplitude and refractive index anomaly changes in high-altitude environments with large elevation differences, multiple meteorological sensor nodes deployed at altitudes between 3000 and 5000 meters can collect atmospheric density data hourly. Assuming a node at an altitude of 4000 meters measures a density fluctuation amplitude of 0.002 kg / m³ and an anomaly in refractive index of 0.0001, time-frequency analysis of the collected fluctuation data is performed using a wavelet transform algorithm. The wavelet basis function is Daubechies4, and the scale parameter is set to 1-10. The continuous wavelet transform formula is then used. ; Where x(t) is the atmospheric density fluctuation signal (unit: kg / m³), and ψ(t) is the Daubechies 4 wavelet basis function. 'a' is the scale parameter (dimensionless), and 'b' is the translation parameter (unit: s). Extract the frequency components corresponding to scale a=5. This is used as the dominant frequency component. After extracting the dominant frequency component, the refractive index deviation is calculated by comparing it with the standard atmospheric model.
[0017] The main frequency components were extracted, and the dominant frequency of the density fluctuation was found to be 0.05 Hz, indicating the presence of periodic disturbances. These data were then compared with the standard atmospheric model, and the refractive index deviation was calculated to be 0.00005, which was used for subsequent signal correction.
[0018] Considering the propagation delay deviation of real-time positioning signals, assuming a satellite signal delay deviation of 2.5 nanoseconds in this environment, a time series analysis algorithm calculates the cumulative delay error along the signal propagation path to be 3.2 nanoseconds. This is then correlated with atmospheric refractive index deviation, revealing that the delay is primarily caused by the thin atmosphere at high altitudes. When using differential positioning technology for initial signal path correction, a double-difference algorithm between the ground reference station and the mobile receiver is employed. Assuming a pseudorange error of 0.1 meters for the reference station and 0.3 meters for the mobile station, the double difference is calculated to correct the pseudorange error to 0.05 meters. Simultaneously, phase smoothing is applied to the carrier signal data to reduce noise, resulting in an initial adjusted pseudorange of 23.40 meters (the initial reference value was 23.45 meters), with the carrier signal phase deviation reduced to 0.02 radians. This pseudorange data will serve as the reference input for subsequent steps. Through the above steps, a complete logical chain from data acquisition to signal correction is formed, ensuring improved positioning accuracy in high-altitude environments. At the same time, the corrected data can be combined with the terrain elevation difference model to analyze the further optimization space of the signal path in environments with large elevation differences. For example, in an area with an elevation difference of 1,000 meters, the signal path deviation can be further reduced by 0.01 meters.
[0019] Step 2: Based on the initially adjusted pseudorange data and carrier signal data, extract the temperature gradient influence factor and humidity distribution unevenness characteristics caused by atmospheric changes, and perform signal fluctuation smoothing processing on the intensity of air pressure change interference to obtain a smoothed signal sequence.
[0020] For pseudorange data and carrier signals, environmental parameters related to atmospheric changes are acquired, and data features of temperature gradient and humidity distribution are extracted to determine a preliminary set of environmental impacts. Using this preliminary set, the correlation between sudden changes in air pressure and interference intensity is analyzed. A pre-set threshold is used to filter abnormal interference data, resulting in a filtered interference dataset. Based on this filtered dataset and considering inhomogeneity characteristics, signal fluctuations are segmented for analysis, identifying signal segments with large fluctuation amplitudes and determining key signal intervals for processing. For these key intervals, smoothing techniques are applied to reduce the impact of interference intensity on the signal sequence, resulting in a smoothed set of signal segments. From this smoothed set, phase offset data of the carrier signal is extracted. If the phase offset exceeds a pre-set threshold, the signal sequence is locally adjusted, resulting in an adjusted signal dataset. Using this adjusted dataset and incorporating influence factors, the residual bias of the pseudorange data is analyzed to determine if any unprocessed interference exists, resulting in a final corrected signal dataset. Based on this final corrected signal dataset, an optimized positioning signal sequence is generated, completing the processing flow for the impact of atmospheric changes.
[0021] In practice, when processing the initially adjusted pseudorange data and carrier signal data in a high-altitude environment, the influence factors of temperature gradient caused by atmospheric changes and the characteristics of uneven humidity distribution can be extracted through multi-source data fusion methods, and signal fluctuation smoothing processing can be performed to address the interference intensity of sudden air pressure changes.
[0022] Based on the initial pseudorange data assuming a range of 23.8 meters and a carrier signal phase deviation of 0.03 radians, and using temperature distribution data provided by meteorological satellites combined with temperature change rate data measured by ground sensors at an altitude of 3500 meters, the temperature gradient influence factor is calculated using the following formula: ; in, This represents the change in temperature (unit: °C). This represents the change in altitude (in meters). For example, when the temperature drops by 0.6°C per 100 meters, This factor is used to correct for signal propagation delay.
[0023] Subsequently, this factor was correlated with pseudorange data, and the influence weight of temperature change on signal propagation was derived to be 0.002 meters. Meanwhile, considering the uneven humidity distribution, assuming a humidity fluctuation range of 3% to 8% at an altitude of 4000 meters, a spatial interpolation algorithm was used to model the humidity distribution. The analysis revealed a humidity unevenness characteristic value of 0.007, which was then mapped onto the carrier signal phase deviation, calculating a phase offset of 0.01 radians.
[0024] To address the interference intensity caused by sudden pressure changes, assuming a pressure change of 2.3 hPa at a certain time, a wavelet transform algorithm was used to decompose the signal sequence and extract high-frequency interference components. The dominant interference frequency was found to be 0.1 Hz. A low-pass filter was designed to smooth the signal fluctuations, with a cutoff frequency set at 0.08 Hz. The resulting smoothed signal sequence showed a reduction in pseudorange fluctuation range to 0.03 meters and a carrier signal phase noise reduction to 0.005 radians. To form a complete logical chain, the smoothed signal sequence was combined with wind speed distribution data from a high-altitude environment. Assuming a wind speed fluctuation of 5 m / s, a correlation analysis algorithm was used to evaluate the indirect impact of wind speed on signal propagation, yielding a weight value of 0.0015. This further provides data support for subsequent signal optimization. All of the above processes were implemented using automated algorithms and information technology to ensure the continuity and consistency of data processing.
[0025] Step 3: If there is residual fluctuation in the smoothed signal sequence that exceeds a preset threshold, it is determined to be turbulence interference caused by terrain reflection. The signal path curvature data is obtained, the specific mode of interference is classified, and the classified interference category is obtained.
[0026] For the smoothed signal sequence, the presence of turbulence interference caused by terrain reflection is determined by detecting whether the residual fluctuation exceeds a preset threshold, thus obtaining a preliminary interference assessment result. Based on the preliminary interference assessment result, the curvature data of the signal path is acquired, and the degree of influence of terrain reflection on the signal path is analyzed to determine the key data set for path curvature. For the key data set for path curvature, a classification processing technique is used to divide the interference patterns caused by turbulence effects, resulting in classified interference pattern categories. By combining the classified interference pattern categories with the signal path curvature data, the correlation between each interference pattern and turbulence effects is analyzed to determine a specific list of interference types. Based on the specific list of interference types, for each interference type, the corresponding residual fluctuation segments in the signal sequence are extracted to obtain a fluctuation data set matching the interference type. For the fluctuation data set, a preset smoothing filter tool is applied to locally process the segments in the signal sequence affected by turbulence effects, resulting in an optimized set of signal segments.
[0027] In practical implementation, when analyzing the smoothed signal sequence in complex terrain environments, assuming a preset residual fluctuation threshold of 0.05 meters, if a residual fluctuation of 0.07 meters is detected in a certain segment of the signal sequence, an automated algorithm determines it to be interference from turbulence caused by terrain reflection. Subsequently, a high-precision terrain database is invoked, combined with a signal propagation path model, to obtain data on the curvature of the signal path. Assuming a curvature angle of 0.12 radians for a certain path segment, a geometric analysis algorithm calculates the path deviation to be 0.008 meters. Comparison with historical data shows that this deviation matches the terrain reflection characteristics with 85% accuracy.
[0028] Interference was classified using a pattern recognition algorithm. Based on curvature and residual fluctuation characteristics, the interference patterns were divided into three categories: mild turbulence, moderate turbulence, and severe turbulence. Assuming the current path curvature and fluctuation value corresponded to the moderate turbulence category, the classification results showed a 78% match rate, and corresponding interference category labels were generated. To form a complete logical chain, the classification results were correlated with terrain slope data. Assuming the slope change rate in a certain area was 2.5 degrees per 100 meters, a slope impact analysis algorithm was used to calculate the weighting value of slope on the turbulence effect as 0.003, further providing a reference for subsequent signal correction. All of the above processes were implemented through automated algorithms, with seamless integration of the data processing module, ensuring the accuracy and consistency of the classification results.
[0029] Step 4: Based on the classified interference categories, obtain the carrier phase offset and signal amplitude attenuation characteristics under the corresponding elevation difference terrain, and adaptively adjust the multipath interference degree and terrain reflection delay time to obtain the corrected pseudorange data.
[0030] Based on the classified interference categories, carrier phase offset data and signal amplitude attenuation parameters under terrain conditions with varying elevations are obtained. Preliminary data grouping is performed for different types of interference, resulting in a stratified dataset. Based on the stratified dataset, the impact of terrain elevation on carrier phase offset data is analyzed. A pre-defined filtering tool is used to smooth the offset data, resulting in a processed phase offset dataset. For the processed phase offset dataset, combined with the signal amplitude attenuation parameters, the intensity of multipath interference is detected. If the interference intensity exceeds a preset threshold, local correction is performed on the signal amplitude data, resulting in a corrected amplitude data set. Using the corrected amplitude data set, the delay characteristics caused by terrain reflection are analyzed. Time series analysis is used to quantify the delay, resulting in a quantized delay data set. Based on the quantized delay data set and the distribution of multipath interference intensity, an adaptive correction model based on the Least Mean Square (LMS) algorithm is constructed, with the following formula: ; in, For the weight vector, For step size parameters, Let be the error signal, where The ideal carrier phase offset, which varies with time, is calculated in real time using reference station data or a high-altitude atmospheric propagation model. The measured carrier phase offset (unit: radians) The ideal carrier phase offset is calculated using reference station data or a high-altitude atmospheric propagation model; in this example, it is set as... radian; For input vectors The signal amplitude attenuation value (dB) Multipath interference is defined as the power ratio (dimensionless) of the multipath signal to the direct signal, and the model output is a correction factor. (Unit: meters), used to correct pseudorange data: ; in, This represents the pseudorange.
[0031] The initial value of the weight vector is set to The model output is the adjusted composite data set; The adjusted composite data set is obtained. For this adjusted composite data set, relevant information on elevation differences and terrain reflections is integrated to perform final correction processing on the pseudorange data, resulting in optimized pseudorange data results. Using the optimized pseudorange data results, combined with the correlation information between interference category and delay duration, a data mapping relationship under each category is generated, determining the final signal correction dataset.
[0032] In practical implementation, in complex terrain environments, based on the classified interference categories, such as classified as moderate terrain reflection interference, the elevation difference terrain database is automatically called to extract the elevation difference data of the corresponding area. Assuming that the elevation difference of a certain area is 15.2 meters, the carrier phase offset is calculated to be 0.034 radians through the terrain impact assessment algorithm. At the same time, combined with the signal propagation model, the signal amplitude attenuation value is analyzed to be 2.7 dB, and the feature matching degree reaches 82%.
[0033] Regarding the multipath interference level, assuming the detected multipath interference index is 0.06, an adaptive filtering algorithm is used to dynamically adjust the filtering parameters, reducing the interference level to 0.02. During the adjustment process, historical interference data is compared to confirm that the signal stability after adjustment is improved by about 75%.
[0034] Regarding the terrain reflection delay, assuming an initial delay of 3.5 milliseconds, the system calls a delay compensation algorithm, which, combined with the elevation difference terrain features and the reflection path model, calculates a compensation coefficient of 0.015, correcting the delay to 1.8 milliseconds. The delay error analysis before and after correction shows a consistency of 88%.
[0035] Integrating the aforementioned phase shift, amplitude attenuation, and delay correction data, a pseudorange reconstruction algorithm is used. Taking the pseudorange data of 23.40 meters output from step 1 as input, and corrected by the LMS model, the corrected pseudorange data is 23.18 meters, reducing the error range to 0.27 meters. To ensure logical integrity, the corrected pseudorange data is correlated with the terrain elevation change trend. Assuming the elevation change rate increases by 1.8 meters every 50 meters, a trend matching algorithm yields a correlation coefficient of 0.92 between the pseudorange correction value and the elevation change, further validating the rationality of the corrected data. The entire process is executed by an automated module, with seamless data flow achieved through internal system interfaces.
[0036] Step 5: Based on the corrected pseudorange data, extract the frequency drift rate and signal-to-noise ratio drop characteristics in the carrier signal, and combine them with the carrier period jump frequency and signal receiving angle deviation data to optimize the carrier signal stability and obtain the optimized carrier signal data.
[0037] By extracting frequency drift and signal-to-noise ratio (SNR) drop information from the corrected pseudorange data, the frequency drift data is smoothed using a pre-defined filtering tool, resulting in a smoothed frequency drift dataset. Based on this smoothed dataset and the SNR information, the drop characteristics in the carrier signal are analyzed. Time-series processing is used to quantize these drop characteristics, resulting in a quantized drop data set. For this quantized drop data set, the periodic transitions and transition frequencies in the carrier signal are obtained and compared against a pre-defined threshold. If the transition frequency exceeds the threshold, the periodic transition data is locally corrected, resulting in a corrected transition data set. Based on this corrected transition data set and the receiving angle and angle deviation information in the carrier signal, the interference of angle deviation on signal stability is analyzed. If the angle deviation exceeds a pre-defined range, the receiving angle data is calibrated, resulting in a calibrated angle data set. Finally, using the calibrated angle data set, combined with the frequency drift dataset and the drop data set, a comprehensive stability assessment model is constructed. A pre-trained Support Vector Machine (SVM) model is used to classify and evaluate the stability of carrier signals. The SVM model is trained on historical high-altitude mapping datasets. The training data comes from publicly available high-altitude GNSS signal datasets (containing historical observation features such as carrier phase, signal-to-noise ratio, and Doppler shift under different altitude gradients and seasonal variations), or is pre-collected and calibrated by the implementer in typical high-altitude areas such as the Qinghai-Tibet Plateau and the Himalayas according to the method described herein. The training set contains 5000 samples, and the accuracy on the test set reaches 92%. The trained model parameters are stored internally in the system's signal processing module and can be directly accessed in practical applications. The model parameters are set as follows: penalty factor C = 1.0, RBF kernel parameter γ = 0.1, and optimized using 5-fold cross-validation.
[0038] SVM uses radial basis function (RBF) kernels. ,in .
[0039] Feature vector ; in, Frequency drift rate (Hz / s); Signal-to-noise ratio decrease (dB); Carrier period transition frequency (times / second); Signal reception angle deviation (radians); All features are Z-score normalized before input.
[0040] The training data comes from historical high-altitude mapping datasets, with category labels including "stable" (+1) and "unstable" (-1). Based on the classification results: if it is "unstable", Kalman filtering is activated to smooth the carrier phase; if it is "stable", the current signal is preserved.
[0041] Based on the stability assessment results after classification, and combined with the corrected transition data set and calibrated angle data set, the carrier signal is comprehensively optimized and adjusted to obtain optimized signal sequence data. According to the optimized signal sequence data, a corresponding signal processing log file is generated. The log file records the processing of frequency drift, signal-to-noise ratio decrease, and angle deviation, thus determining the final optimized signal sequence record.
[0042] The stability of the carrier signal is assessed by classifying it using a Support Vector Machine (SVM) algorithm, employing a Radial Basis Function (RBF) kernel. ; in, The kernel parameter is set to 0.1. The feature vector includes the frequency drift rate, signal-to-noise ratio decrease, carrier period hopping frequency, and signal reception angle deviation. The training dataset is based on annotations. ,in For feature vectors, As category labels, solve the optimization problem. Obtain the classification model, with the classification label being 'stable' or 'unstable'.
[0043] In complex signal environments, based on the corrected pseudorange data, using the 23.18 meters output from step 4 as input, the frequency drift rate and signal-to-noise ratio (SNR) drop in the carrier signal are analyzed through a signal feature extraction algorithm. The frequency drift rate is calculated to be 0.021 Hz / s (based on typical Doppler frequency shift values at high altitudes), while the SNR drop is 1.5 dB. During the feature extraction process, a historical signal database is called for comparison to confirm that the fluctuation range of the extracted values is within an acceptable threshold, with a consistency of 85%.
[0044] Based on the carrier period transition frequency data, assuming the detected transition frequency is 0.013 times per second, the influence of the transition on signal continuity is evaluated through a periodic stability analysis model, and the influence coefficient is found to be 0.008. The carrier tracking parameters are then automatically adjusted to reduce transition interference.
[0045] Introducing signal reception angle deviation data, assuming a deviation of 0.045 radians, an angle correction algorithm, combined with a receiving antenna array model, calculates a correction factor of 0.012, reducing the deviation to 0.018 radians. Signal consistency analysis before and after correction shows a matching degree of 90%. Finally, the system integrates frequency drift rate, noise ratio reduction, frequency jump, and angle deviation correction data, and generates optimized carrier signal data through a carrier signal optimization algorithm. Assuming the signal stability index was 0.65 before optimization, it improves to 0.88 after optimization, indicating a signal quality improvement of approximately 35%. To ensure logical rigor, the optimization results are correlated with the signal environment's changing trends. Assuming an environmental noise change rate of 0.03 dB per hour, an environmental adaptability matching model yields a correlation coefficient of 0.89 between the optimized sequence and environmental changes, verifying the reliability of the optimized data. The entire process is completed by an automated signal processing module, with data transmitted uninterruptedly via an internal interface.
[0046] Step 6: By combining the optimized carrier signal data with the corrected pseudorange data, the mapping coordinate deviation in the high-altitude environment is fused and corrected to obtain the corrected mapping coordinate result.
[0047] By combining optimized carrier signal data with corrected pseudorange data, preliminary data alignment is performed using a preset signal matching tool to obtain an aligned joint dataset of signal and pseudorange. For this joint dataset, interference information from high-altitude environmental factors on signal propagation is acquired. If the interference exceeds a preset threshold, environmental compensation is applied to the joint dataset to determine the compensated joint data set. Based on the compensated joint data set, deviation information in the mapping coordinates is extracted, and a support vector machine algorithm is used to classify the deviation data, resulting in a classified deviation data set. Using this classified deviation data set, a fusion method is employed to correct the deviation data, obtaining the corrected deviation correction value and determining the corrected coordinate offset dataset. Based on the corrected coordinate offset dataset, point-by-point correction is performed on the mapping coordinates to obtain a corrected mapping coordinate result set. Finally, using the corrected mapping coordinate result set and optimized carrier signal data, the coordinate accuracy is verified a second time. If the verification result does not meet a preset standard, the coordinate result set is fine-tuned locally to obtain the final coordinate accuracy dataset.
[0048] In high-altitude environments, the optimized carrier signal data is first acquired, assuming a time synchronization error of 0.002 seconds. Combined with the corrected pseudorange data, and using the 23.18 meters output in step 4 as input, the data is initially integrated through a data fusion algorithm. The initial coordinate deviation is calculated to be 0.034 meters. During the fusion process, the historical mapping database in high-altitude environments is called for comparison to confirm that the deviation value is within the expected range and the matching degree reaches 88%.
[0049] To address the unique influences of air pressure and temperature in high-altitude environments, an environmental parameter correction model is introduced. Assuming an air pressure of 850 hPa and a temperature of -5.2 degrees Celsius, an environmental impact analysis algorithm is used to calculate the refraction error of the environment on signal propagation as 0.015 meters, and a corresponding correction coefficient of 0.007 is generated. The pseudorange data is then automatically adjusted to reduce environmental interference.
[0050] By utilizing multi-source data fusion technology, carrier signal data and pseudorange data are deeply integrated, and the Kalman filter algorithm is used to dynamically estimate the coordinate deviation. The state equation and observation equation are as follows: ; ; in, For state vectors, This includes three-dimensional position (m), three-dimensional velocity (m / s), and three-dimensional coordinate deviation (m). Here is the state transition matrix. For the observation matrix, and For process noise and observation noise (the covariance matrix is set as a diagonal matrix), coordinate bias is estimated through prediction and update steps.
[0051] Where: State transition matrix and observation matrix Defined as: ; ; in, The second is the filtering time interval; Process noise and observation noise The covariance matrix is based on Allan analysis of variance.
[0052] Specifically, by performing Allan variance analysis on the IMU / DGNSS integrated navigation system under static reference station and high-dynamic aerial survey flight scenarios, the intensity characteristics of major noise sources such as angle random walk, velocity random walk, and accelerometer bias instability were identified. The values of the aforementioned covariance matrix were then obtained through statistical calculations. This method is a conventional technique in high-precision integrated navigation filtering in this field.
[0053] The specific value is: ; ; in, Corresponding state vector The noise variance of each component.
[0054] Assuming the filtered deviation convergence value is 0.012 meters, and analyzing the stability of the coordinate data, the fluctuation range is found to be within 0.005 meters, which meets the requirements of high-precision surveying.
[0055] To further improve the accuracy of the correction, a terrain elevation model was introduced. Assuming that the average elevation of the high-altitude area is 4500 meters, the terrain error compensation algorithm was used to calculate that the additional deviation caused by the elevation was 0.009 meters, and a compensation value of 0.004 meters was automatically generated, reducing the final coordinate deviation to 0.008 meters.
[0056] The reliability of the corrected survey coordinates was verified by comparing them with regional survey benchmark data. Assuming a benchmark deviation threshold of 0.01 meters, the verification results showed a 92% agreement between the corrected coordinates and the benchmark data, confirming the reliability of the correction results. The entire process was completed by an automated survey processing module, with data seamlessly integrated through internal interfaces to ensure processing efficiency and accuracy.
[0057] Step 7: Based on the corrected mapping coordinates, identify potential data failure areas; automatically mark failure points by calculating the grid point signal coverage (coverage <75% is considered failure) and combining it with the residual fluctuation threshold (>0.05 meters). Using iterative verification technology, perform multiple rounds of comparison between the corrected coordinates and the high-precision control point database (matching success rate >90% and average deviation <0.01 meters is considered passing), obtaining verified high-precision, high-altitude, small-scale aerial triangulation results.
[0058] For the corrected mapping coordinate data, environmental interference information in high-altitude areas is acquired. If the interference information exceeds a preset threshold, the coordinate data is initially screened to obtain a filtered coordinate dataset. Based on the filtered coordinate dataset, potential data failure ranges are identified, and data points within these ranges are marked using a preset comparison tool to determine the marked failed data subset. For the marked failed data subset, information on associated surrounding coordinate points is acquired, and reliability analysis is performed on the subset through point-by-point comparison to obtain a data group after reliability analysis. Based on the data group after reliability analysis, a support vector machine algorithm is used to classify the data points within each group, determining a set of data accuracy levels after classification. For the set of data accuracy levels after classification, data points that do not meet the preset accuracy standard are identified. If the proportion of data points that do not meet the standard exceeds a preset ratio, local interpolation is performed on these data points to obtain an interpolated coordinate data group. Based on the interpolated coordinate data group, a layer-by-layer verification tool is used to perform multiple rounds of reliability checks on the coordinate points within the data group to determine the final high-precision coordinate dataset. For the final high-precision coordinate dataset, combined with the aerial triangulation encryption requirements for a small area, the coordinate points are formatted using a data mapping tool to obtain a high-precision result set that meets the aerial triangulation encryption standard.
[0059] In high-altitude surveying scenarios, a series of automated processes are carried out based on the corrected surveying coordinate results. First, a data anomaly detection algorithm is used to identify potential data failure areas. Assuming that a total of 12 grid points with a signal coverage of less than 75% are detected in a small area, these points are automatically marked as potential failure areas, and their coordinate distribution density is recorded as 3.2 points per square kilometer. Then, a spatial interpolation algorithm is used to perform data integrity analysis on these areas, and the interpolation prediction error is calculated to be 0.018 meters. It is confirmed that the impact weight of failure areas on the overall accuracy is 6.3%.
[0060] Iterative verification technology is used to check the reliability of coordinate accuracy. A multi-round comparison algorithm is invoked to match the corrected coordinates with historical control point data in the high-altitude area. Assuming a total of 50 control points and a matching success rate of 90%, the calculated average deviation is 0.007 meters. A reliability score is generated, assuming a score of 85, exceeding the preset threshold of 80, automatically determining that the coordinate accuracy meets the requirements. To further achieve high-precision, high-altitude, small-scale aerial triangulation results, a gridded encryption algorithm is introduced. The area is divided into tiny grids with sides of 50 meters. Data augmentation processing is performed on each grid point. Assuming the original point cloud density for each grid point is 8.5 points per square meter, after removing noise points using a point cloud filtering algorithm, the density is reduced to 7.2 points per square meter. Combined with triangulation network construction technology, the encrypted results are generated, with the average side length of the triangulation network calculated to be 2.1 meters, meeting the high-precision requirements for a small area. The entire process is executed automatically through internal system modules. Data flow relies on a high-performance computing platform to ensure a logical closed loop from failure area identification to encrypted result generation. During this process, the storage business of surveying and mapping results is also linked, automatically storing the encrypted results in the cloud database and generating backup index numbers such as "HB-2023-001" for easy subsequent retrieval and traceability.
[0061] This invention addresses the fundamental physical constraints faced by positioning signals propagating in high-altitude environments with significant elevation differences. It systematically constructs a comprehensive correction mechanism, achieving significant beneficial effects. The inaccuracies of traditional methods stem from their failure to fundamentally address physical layer deviations caused by changes in the signal propagation medium and path. This invention directly tackles the core issues of signal propagation—atmospheric density fluctuations and refractive index anomalies—by employing differential positioning technology to directly correct the initial signal path at the physical level, effectively rectifying signal delays arising from changes in the propagation path.
[0062] For the signal waveform itself, intrinsic physical parameters affecting the signal phase, such as temperature gradient and humidity distribution, are extracted for signal smoothing, effectively suppressing the impact of environmental parameter disturbances on signal stability. When residual fluctuations caused by terrain reflection turbulence are detected, they are classified according to the principle that different interference modes correspond to different physical causes, and adaptive adjustments are made to the effects of geometric and temporal dimensions such as multipath effects and terrain reflection delay, achieving accurate correction of pseudorange data.
[0063] By fusing optimized carrier signal data with corrected pseudorange data, the scheme achieves fusion correction of mapping coordinate deviations at the data level, and employs iterative verification technology to ensure the reliability of the results. The entire process constitutes a closed-loop optimization system from physical interference identification to signal parameter correction, and then to geometric coordinate generation. The full-process correction mechanism strictly follows the physical laws of signal propagation and error generation mechanisms. By peeling away and correcting errors at their physical root causes layer by layer, it fundamentally ensures the accuracy and reliability of small-scale aerial triangulation results in high-altitude, high-elevation-difference environments, providing effective technical support for accurate mapping in complex terrain areas.
[0064] Example 2: This embodiment discloses an aerial survey triangulation system assisted by IMU / DGNSS for high-altitude, large-elevation-difference aerial surveys, including: The data acquisition and preliminary correction module is used to acquire atmospheric density fluctuation amplitude and refractive index anomaly change data in high-altitude and large-elevation-difference environments. Combined with the propagation delay deviation of real-time positioning signals, differential positioning technology is used to perform preliminary correction of the signal path to obtain initially adjusted pseudorange data and carrier signal data. The signal smoothing module is used to extract the temperature gradient influence factor and humidity distribution unevenness caused by atmospheric changes based on the initially adjusted pseudorange data and carrier signal data, and to perform signal fluctuation smoothing processing on the intensity of air pressure change interference to obtain a smoothed signal sequence. The interference mode classification module is used to determine that if there is residual fluctuation in the smoothed signal sequence that exceeds a preset threshold, it is turbulence effect interference caused by terrain reflection, obtain signal path curvature data, classify the specific mode of interference, and obtain the classified interference category. The pseudorange correction module is used to obtain the carrier phase offset and signal amplitude attenuation characteristics under the corresponding elevation difference terrain based on the classified interference category, and to adaptively adjust the multipath interference degree and terrain reflection delay time to obtain the corrected pseudorange data. The carrier signal optimization module is used to extract the frequency drift rate and signal-to-noise ratio reduction characteristics in the carrier signal based on the corrected pseudorange data, and combine the carrier period jump frequency and signal receiving angle deviation data to optimize the carrier signal stability and obtain optimized carrier signal data. The coordinate fusion correction module is used to combine the optimized carrier signal data with the corrected pseudorange data to perform fusion correction on the mapping coordinate deviation in high-altitude environments, and obtain the corrected mapping coordinate results. The results verification module is used to identify potential data failure areas based on the corrected survey coordinates. It automatically marks failure points by calculating the grid point signal coverage (coverage <75% is considered failure) and combining this with the residual fluctuation threshold (>0.05 meters). Using iterative verification technology, the corrected coordinates are compared multiple times with the high-precision control point database (a matching success rate >90% and an average deviation <0.01 meters are considered passing), resulting in high-precision, high-altitude, small-scale aerial triangulation results that have passed verification.
[0065] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.
[0066] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for aerial triangulation densification in high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial surveying, characterized in that... The method includes the following steps: Step 1: Collect data on atmospheric density fluctuation amplitude and refractive index anomaly changes in high-altitude environments with large elevation differences. Combine this with the propagation delay deviation of the real-time positioning signal and use differential positioning technology to make preliminary corrections to the signal path, thereby obtaining the initially adjusted pseudorange data and carrier signal data. Step 2: Based on the initially adjusted pseudorange data and carrier signal data, extract the temperature gradient influence factor and humidity distribution unevenness characteristics caused by atmospheric changes, and perform signal fluctuation smoothing processing on the intensity of air pressure change interference to obtain a smoothed signal sequence. Step 3: If there is residual fluctuation in the smoothed signal sequence that exceeds a preset threshold, it is determined to be turbulence interference caused by terrain reflection. The signal path curvature data is obtained, the specific pattern of interference is classified, and the classified interference category is obtained. Step 4: Based on the classified interference categories, obtain the carrier phase offset and signal amplitude attenuation characteristics under the corresponding elevation difference terrain, and adaptively adjust the multipath interference degree and terrain reflection delay time to obtain the corrected pseudorange data. Step 5: Based on the corrected pseudorange data, extract the frequency drift rate and signal-to-noise ratio reduction features from the carrier signal, and combine them with the carrier period jump frequency and signal receiving angle deviation data to optimize the carrier signal stability and obtain the optimized carrier signal data. Step 6: By combining the optimized carrier signal data with the corrected pseudorange data, the mapping coordinate deviation in the high-altitude environment is fused and corrected to obtain the corrected mapping coordinate result; Step 7: Identify potential data failure areas based on the corrected mapping coordinates; By calculating the grid point signal coverage and automatically marking failure points in conjunction with the residual fluctuation threshold, and using iterative verification technology, the corrected coordinates are compared with the high-precision control point database in multiple rounds to obtain high-precision, high-altitude, small-scale aerial triangulation results that have passed verification.
2. The aerial triangulation densification method for high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial surveying according to claim 1, characterized in that, Step 1 includes: By collecting atmospheric density fluctuation data in high-altitude environments with large elevation differences, and combining it with refractive index anomaly change data, an initial environmental impact dataset is constructed, and a preliminary set of environmental parameters is obtained. Based on the preliminary set of environmental parameters, the correlation between atmospheric density fluctuation amplitude and abnormal refractive index changes was analyzed. Anomaly data was filtered using a preset threshold to determine the filtered environmental impact dataset. For the filtered environmental impact dataset, the propagation delay deviation data of the positioning signal is fused, and differential positioning technology is applied to correct the signal path to obtain the corrected signal path dataset. The deviation value of the pseudorange data is calculated using the corrected signal path dataset. If the deviation value exceeds the preset threshold, the pseudorange data is corrected a second time to obtain the adjusted pseudorange dataset. Based on the adjusted pseudorange dataset, the phase offset of the carrier signal data is extracted, and the phase offset is smoothed using Kalman filtering technology to determine the smoothed carrier signal dataset. The smoothed carrier signal dataset is obtained, and combined with the adjusted pseudorange dataset, the propagation delay of the positioning signal is comprehensively analyzed to determine whether there is any residual bias, and the final positioning signal correction dataset is obtained. By using the final positioning signal correction dataset, high-precision positioning result data is generated, thus completing the optimization processing of positioning signals in high-altitude and large-elevation-difference environments.
3. The aerial triangulation densification method for high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial surveying according to claim 1, characterized in that, Step 2 includes: Based on pseudorange data and carrier signals, environmental parameters related to atmospheric changes are obtained, data features of temperature gradient and humidity distribution are extracted, and a preliminary set of environmental impacts is determined. By using a preliminary environmental impact dataset, the correlation between sudden changes in air pressure and the intensity of disturbance is analyzed. Anomaly disturbance data is filtered using a preset threshold to obtain the filtered disturbance dataset. Based on the filtered interference dataset and combined with the non-uniformity characteristics, the signal fluctuations are segmented and analyzed to obtain the signal segments with larger fluctuation amplitudes and determine the signal intervals that need to be processed. For the key signal ranges, smoothing techniques are applied to reduce the impact of interference intensity on the signal sequence, resulting in a smoothed set of signal segments. From the smoothed signal segment set, the phase offset data of the carrier signal is extracted. If the phase offset exceeds the preset threshold, the signal sequence is locally adjusted to obtain the adjusted signal dataset. By analyzing the residual bias of the pseudorange data using the adjusted signal dataset and influencing factors, we can determine whether there is any unprocessed interference and obtain the final signal correction set. Based on the final signal correction set, an optimized positioning signal sequence is generated to complete the processing flow for the impact of atmospheric changes.
4. The aerial triangulation densification method for high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial surveying according to claim 1, characterized in that, Step 3 includes: For the smoothed signal sequence, by detecting whether the residual fluctuation exceeds a preset threshold, it is determined whether there is turbulence interference caused by terrain reflection, and a preliminary interference judgment result is obtained; Based on the preliminary interference assessment results, the curvature data of the signal path is obtained, the degree of influence of terrain reflection on the signal path is analyzed, and the key data set of path curvature is determined. For the key dataset of path curvature, a classification processing technique is used to divide the interference patterns caused by turbulence effects and obtain the classified interference pattern categories. By classifying the interference modes and combining the signal path curvature data, the correlation between each interference mode and turbulence effect is analyzed to determine a specific list of interference types. Based on the specific list of interference types, for each type of interference, the corresponding residual fluctuation segment in the signal sequence is extracted to obtain a fluctuation data set that matches the interference type. For fluctuating data sets, a preset smoothing filter tool is applied to locally process the segments of the signal sequence affected by turbulence, resulting in an optimized set of signal segments.
5. The aerial triangulation densification method for high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial surveying according to claim 1, characterized in that, Step 4 includes: By classifying the interference categories, carrier phase offset data and signal amplitude attenuation parameters are obtained in terrain environments with elevation differences. Preliminary data grouping is performed for different types of interference categories to obtain a hierarchical dataset. Based on the layered dataset, the impact of elevation differences on carrier phase offset data is analyzed, and the offset data is smoothed using a preset filtering tool to obtain the processed phase offset dataset. For the processed phase offset dataset, the magnitude of multipath interference intensity is detected by combining the signal amplitude attenuation parameter. If the interference intensity exceeds the preset threshold, the signal amplitude data is locally corrected to obtain the corrected amplitude data set. By analyzing the delay duration characteristics caused by terrain reflection through the corrected amplitude data set, the delay duration is quantified using time series analysis methods to obtain the quantized delay data set. Based on the quantized delay data set and the distribution of multipath interference intensity, an adaptive correction model based on the Least Mean Square (LMS) algorithm is constructed, with the following formula: ； in, For the weight vector, For step size parameters, For error signals, The measured carrier phase offset (unit: radians) The ideal carrier phase offset is calculated using reference station data or a high-altitude atmospheric propagation model; in this example, it is set as... radian; For input vectors The signal amplitude attenuation value (dB) Multipath interference is defined as the power ratio (dimensionless) of the multipath signal to the direct signal, and the model output is a correction factor. (Unit: meters), used to correct pseudorange data: ； in, This represents the pseudorange; the initial values of the weight vector are set to... The model output is the adjusted composite data set; For the adjusted integrated data set, relevant information on elevation difference and terrain reflection is integrated, and the pseudorange data is finally corrected to obtain the optimized pseudorange data results. By combining the optimized pseudorange data with the correlation information of interference category and delay duration, a data mapping relationship under the classification is generated to determine the final signal correction dataset.
6. The aerial triangulation densification method for high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial surveying according to claim 1, characterized in that, Step 5 includes: By using the corrected pseudorange data, information related to frequency drift and signal-to-noise ratio decrease in the carrier signal is extracted. The frequency drift data is then smoothed using a preset filtering tool to obtain a smoothed frequency drift dataset. Based on the smoothed frequency drift dataset and combined with the noise ratio information, the drop value characteristics in the carrier signal are analyzed. The drop value characteristics are quantized using time series processing methods to obtain the quantized drop value data set. For the quantized drop value data set, the periodic transition and transition frequency information in the carrier signal are obtained and compared with a preset threshold. If the transition frequency exceeds the preset threshold, the periodic transition data is locally corrected to obtain the corrected transition data set. Based on the corrected transition data set, combined with the receiving angle and angle deviation information in the carrier signal, the degree of interference of the angle deviation on signal stability is analyzed. If the angle deviation exceeds the preset range, the receiving angle data is calibrated to obtain the calibrated angle data set. By combining the calibrated angle data set with the frequency drift dataset and the drop value data set, a comprehensive stability assessment model is constructed. The support vector machine algorithm is used to classify and assess the stability of the carrier signal, and the classified stability assessment results are obtained. Based on the stability assessment results after classification, the carrier signal is optimized and adjusted as a whole by combining the corrected jump data set and the calibrated angle data set to obtain the optimized signal sequence data. Based on the optimized signal sequence data, a corresponding signal processing log file is generated. The log file records the processing of frequency drift, signal-to-noise ratio decrease, and angle deviation, thus determining the final optimized signal sequence record.
7. The aerial triangulation densification method for high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial surveying according to claim 1, characterized in that, Step 6 includes: By combining the optimized carrier signal data with the corrected pseudorange data, a preliminary data alignment process is performed using a preset signal matching tool to obtain the aligned signal and pseudorange joint dataset. For the aligned signal and pseudorange joint dataset, obtain the interference information of environmental factors in high-altitude areas on signal propagation. If the interference information exceeds the preset threshold range, perform environmental compensation adjustment on the joint dataset to determine the compensated joint data set. Based on the compensated joint data set, the deviation information in the mapping coordinates is extracted, and the deviation data is classified using the support vector machine algorithm to obtain the classified deviation data set. By using the classified deviation data set and combining the fusion method to perform correction calculations on the deviation data, the corrected deviation correction value is obtained, and the corrected coordinate offset dataset is determined. Based on the corrected coordinate offset dataset, point-by-point correction processing is performed on the survey coordinates to obtain the corrected survey coordinate result set. By combining the calibrated coordinate result set with the optimized carrier signal data, the coordinate accuracy is verified a second time. If the verification result does not meet the preset standard, the coordinate result set is finely adjusted locally to obtain the final coordinate accuracy dataset.
8. The aerial triangulation densification method for high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial surveying according to claim 1, characterized in that, Step 7 includes: For the corrected mapping coordinate data, environmental interference information in the high-altitude area is obtained. If the interference information exceeds the preset threshold range, the coordinate data is initially filtered to obtain the filtered coordinate dataset. Based on the filtered coordinate dataset, identify potential data failure ranges, use a pre-set comparison tool to mark data points within the failure range, and determine the marked failure data subset. For the labeled subset of failed data, obtain the surrounding coordinate point information associated with it, and perform reliability analysis on the subset of failed data by comparing it point by point to obtain the data group after reliability analysis; Based on the data grouping after reliability analysis, the support vector machine algorithm is used to classify the data points within the group and determine the set of data accuracy levels after classification. For the set of data precision levels after classification, the data points that do not meet the preset precision standard are obtained. If the proportion of data points that do not meet the standard exceeds the preset proportion, local interpolation is performed on the data points to obtain the interpolated coordinate data set. Based on the interpolated coordinate data set, the coordinate points in the data set are checked multiple times using a layer-by-layer verification tool to determine the final high-precision coordinate dataset. For the final high-precision coordinate dataset, combined with the aerial triangulation encryption requirements for a small area, the coordinate points are formatted using a data mapping tool to obtain a high-precision result set that meets the aerial triangulation encryption standard.
9. A high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial triangulation system, used to execute the high-altitude, large-elevation-difference IMU / DGNSS-assisted aerial triangulation method according to any one of claims 1-8, characterized in that, include: The data acquisition and preliminary correction module is used to acquire atmospheric density fluctuation amplitude and refractive index anomaly change data in high-altitude and large-elevation-difference environments. Combined with the propagation delay deviation of real-time positioning signals, differential positioning technology is used to perform preliminary correction of the signal path to obtain initially adjusted pseudorange data and carrier signal data. The signal smoothing module is used to extract the temperature gradient influence factor and humidity distribution unevenness caused by atmospheric changes based on the initially adjusted pseudorange data and carrier signal data, and to perform signal fluctuation smoothing processing on the intensity of air pressure change interference to obtain a smoothed signal sequence. The interference mode classification module is used to determine that if there is residual fluctuation in the smoothed signal sequence that exceeds a preset threshold, it is turbulence effect interference caused by terrain reflection, obtain signal path curvature data, classify the specific mode of interference, and obtain the classified interference category. The pseudorange correction module is used to obtain the carrier phase offset and signal amplitude attenuation characteristics under the corresponding elevation difference terrain based on the classified interference category, and to adaptively adjust the multipath interference degree and terrain reflection delay time to obtain the corrected pseudorange data. The carrier signal optimization module is used to extract the frequency drift rate and signal-to-noise ratio reduction characteristics in the carrier signal based on the corrected pseudorange data, and combine the carrier period jump frequency and signal receiving angle deviation data to optimize the carrier signal stability and obtain optimized carrier signal data. The coordinate fusion correction module is used to combine the optimized carrier signal data with the corrected pseudorange data to perform fusion correction on the mapping coordinate deviation in high-altitude environments, and obtain the corrected mapping coordinate results. The results verification module is used to identify potential data failure areas based on the corrected mapping coordinate results; By calculating the grid point signal coverage and automatically marking failure points in conjunction with the residual fluctuation threshold, and using iterative verification technology, the corrected coordinates are compared with the high-precision control point database in multiple rounds to obtain high-precision, high-altitude, small-scale aerial triangulation results that have passed verification.