Real-time meteorological correction and network adjustment model method based on measurement robot collaboration

Abstract

This invention relates to the field of dam deformation monitoring technology, and particularly to a real-time meteorological correction and network adjustment model method based on collaborative measurement robots. The method includes: collecting meteorological observation data using meteorological sensors and obtaining the distance between measurement points using several measurement robots; determining the number of measurement robots to be connected to the network based on actual measurement needs; updating the meteorological correction model based on the network adjustment data generated by the network adjustment model; if the operational stability of the network adjustment model does not meet requirements, determining whether to increase the time window length of the network adjustment model; determining whether the adaptability of the network adjustment model meets requirements based on the parameter update frequency of the meteorological correction model; determining whether to reduce the weight of the observations from the measurement robots; and determining whether to increase the minimum effective signal-to-noise ratio of the optimized observation data based on the response delay of the measurement robot sensors. This invention improves the operational stability of the network adjustment model.

Description

Technical Field

[0001] This invention relates to the field of dam deformation monitoring technology, and in particular to a real-time meteorological correction and network adjustment model method based on the collaboration of measurement robots. Background Technology

[0002] In existing technologies, surveying robots play a crucial role in fields such as engineering surveying and deformation monitoring. These are intelligent electronic total stations with automatic search, precise aiming, and reading capabilities. They calculate the coordinates of target points by acquiring the angle and distance relative to the station. However, single surveying robots have several limitations, such as being restricted by line-of-sight conditions and maximum target recognition distance. They can only be used for deformation monitoring of deformable bodies with good line-of-sight and small deformation areas, and their monitoring accuracy and speed are low. When multiple surveying robots are networked together, errors exist in the observations of each robot. Network adjustment models are needed to process these observations to obtain more accurate measurement results. Through real-time meteorological correction and network adjustment model methods, real-time network calculation of the dynamic positions of multiple surveying robot stations and monitoring points can be achieved, thereby obtaining high-precision real-time movement trajectories of monitoring points. Related technologies are applied to the automated measurement and control, data transmission, data processing, data management, and display of surveying robot systems for dams and slopes in water conservancy projects, providing high-precision monitoring results for surface deformation monitoring of dams and slopes in water conservancy projects.

[0003] Chinese Patent Publication No. CN116957431A discloses a method, apparatus, and device for constructing a meteorological and hydrological environmental data organization model, comprising: acquiring meteorological and hydrological environmental data to be processed; extracting spatial features, temporal features, and attribute features of the meteorological and hydrological environmental data; constructing a spatial grid coding expression model in the spatial dimension based on the spatial features and an Earth partitioning grid framework; constructing a temporal discrete coding expression model in the temporal dimension based on the temporal features; constructing a multi-level attribute coding expression model in the attribute dimension based on the attribute features; and integrating the spatial grid coding expression model, the temporal discrete coding expression model, and the multi-level attribute coding expression model to obtain a meteorological and hydrological environmental data organization model, wherein the meteorological and hydrological environmental data organization model is used to uniquely encode the meteorological and hydrological environmental data in the spatial dimension, the temporal dimension, and the attribute dimension.

[0004] It is evident that existing technologies have the following problems: if the granularity of the Earth meshing framework does not match the actual needs, excessively coarse meshing will result in the loss of key spatial details, while excessively fine meshing will increase the computational burden. Furthermore, the failure to consider the differences in coordinate references among multiple data sources can easily lead to coordinate transformation errors, resulting in the network adjustment model failing to meet operational stability requirements. Summary of the Invention

[0005] To address this, the present invention provides a real-time meteorological correction and network adjustment model method based on the collaboration of measurement robots. This method overcomes the problems in existing technologies where the granularity of the Earth mesh framework does not match actual needs, and being too coarse will result in the loss of key spatial details, while being too fine will increase the computational burden. Furthermore, the method does not consider the differences in coordinate references between multiple data sources, which can easily cause coordinate transformation errors and lead to the network adjustment model's operational stability not meeting requirements.

[0006] To achieve the above objectives, this invention provides a method for real-time meteorological correction and network adjustment model based on collaborative measurement robots, comprising:

[0007] Meteorological observation data is collected by meteorological sensors and the distance between measurement points is obtained by several measurement robots. The number of measurement robots connected to the network is determined according to the actual measurement needs.

[0008] The meteorological observation data is preprocessed to output optimized observation data. A meteorological correction model is constructed based on the optimized observation data. Error correction is performed on the meteorological observation data to output corrected observation data. A network adjustment model is constructed based on the corrected observation data. The meteorological correction model is updated based on the network adjustment data generated by the network adjustment model.

[0009] Obtain the effective number and total number of network adjustment data respectively, and calculate the effectiveness rate of the network adjustment data;

[0010] The efficiency of network adjustment data is used to determine whether the operational stability of the network adjustment model meets the requirements.

[0011] If the operational stability of the network adjustment model does not meet the requirements, then determine whether it is necessary to increase the time window length of the network adjustment model.

[0012] If it is not necessary to increase the time window length of the network adjustment model, then the suitability of the network adjustment model is determined based on the parameter update frequency of the meteorological correction model.

[0013] If the adaptability of the network adjustment model does not meet the requirements, then determine whether it is necessary to reduce the weight of the observations of the measurement robot.

[0014] If it is not necessary to reduce the weight of the observations of the measurement robot, then determine whether it is necessary to increase the minimum effective signal-to-noise ratio of the optimized observation data based on the response delay of the measurement robot's sensors.

[0015] Furthermore, based on the effectiveness of the network adjustment data, determining whether the operational stability of the network adjustment model meets the requirements includes:

[0016] The effectiveness of the network adjustment data is compared with the preset second effectiveness.

[0017] If the effectiveness of the network adjustment data is greater than the preset second effectiveness, then the operational stability of the network adjustment model is determined to meet the requirements.

[0018] If the effectiveness of the network adjustment data is less than or equal to the preset second effectiveness, then the operational stability of the network adjustment model is determined to be unsatisfactory.

[0019] Further, determine whether the time window length of the network adjustment model needs to be increased, including:

[0020] The effectiveness of the network adjustment data is compared with the preset first effectiveness and the preset second effectiveness, respectively.

[0021] If the effectiveness of the network adjustment data is less than the preset first effectiveness, then it is determined that the time window length of the network adjustment model needs to be increased.

[0022] If the effectiveness of the network adjustment data is greater than or equal to the preset first effectiveness and less than or equal to the preset second effectiveness, then it is determined that it is not necessary to increase the time window length of the network adjustment model.

[0023] Furthermore, the effectiveness rate of the network adjustment data is the ratio of the effective number of network adjustment data to the total number.

[0024] Furthermore, the increase in the time window length of the network adjustment model is determined by the difference between the effectiveness of the network adjustment data and the preset first effectiveness.

[0025] Furthermore, the suitability of the parameter update frequency based on the meteorological correction function for the network adjustment model is determined, including:

[0026] The parameter update frequency of the meteorological correction model is compared with the preset second update frequency;

[0027] If the parameter update frequency of the meteorological correction model is greater than the preset second update frequency, then it is determined that the adaptability of the network adjustment model meets the requirements, and it is determined whether the time window length of the network adjustment model meets the requirements.

[0028] If the parameter update frequency of the meteorological correction model is less than or equal to the preset second update frequency, then the adaptability of the network adjustment model is determined to be unacceptable.

[0029] Further, determine whether it is necessary to reduce the weight of the robot's observations, including:

[0030] The parameter update frequency of the meteorological correction model is compared with the preset first update frequency and the preset second update frequency, respectively;

[0031] If the parameter update frequency of the meteorological correction model is greater than the preset first update frequency and less than or equal to the preset second update frequency, then it is determined that the weight of the observation value of the measurement robot does not meet the requirements, and the weight of the observation value of the measurement robot is reduced.

[0032] If the parameter update frequency of the meteorological correction model is less than or equal to the preset first update frequency, then it is determined that there is no need to reduce the weight of the observations of the measuring robot.

[0033] Furthermore, the reduction in the weight of the observations of the measuring robot is determined by the difference between the parameter update frequency of the meteorological correction model and the preset first update frequency.

[0034] Furthermore, based on the response delay of the robot's sensors, it is determined whether the minimum effective signal-to-noise ratio of the optimized observation data needs to be increased, including:

[0035] The response delay of the robot's sensors is compared with the preset delay.

[0036] If the response delay of the measurement robot's sensor is less than or equal to the preset delay, then it is determined that there is no need to increase the minimum effective signal-to-noise ratio of the optimized observation data, and it is determined whether the weight of the measurement robot's observations meets the requirements.

[0037] If the response delay of the measurement robot sensor is greater than the preset delay, then it is determined that the minimum effective signal-to-noise ratio of the optimized observation data needs to be increased, and the minimum effective signal-to-noise ratio of the optimized observation data is increased.

[0038] Furthermore, the increase in the minimum effective signal-to-noise ratio of the optimized observation data is determined by measuring the difference between the response delay of the robot sensor and the preset delay.

[0039] Compared with existing technologies, the beneficial effects of this invention are as follows: The method of this invention adjusts the time window length of the network adjustment model based on the effectiveness of the network adjustment data. Because the measuring robot has not been calibrated after long-term operation, its performance indicators such as ranging accuracy and angle measurement accuracy will change. This causes the actual observation error to exceed the preset range of the adjustment model. When the model processes these errors, it cannot accurately correct them, resulting in a significant increase in residuals, unit weight error exceeding the normal range, and an increase in the condition number of the normal equation matrix. Ultimately, this leads to a decrease in the stability of the network adjustment model, increased fluctuations in the solution results, and difficulty in meeting measurement accuracy requirements. By increasing the time window length of the network adjustment model, the model can integrate more historical observation data, enhance the statistical smoothing ability of the uncalibrated error of the measuring robot, and reduce the weight of single abnormal observations. The weight of the measuring robot's observations is adjusted according to the parameter update frequency of the meteorological correction model. Because when meteorological conditions change abruptly, the rigid update mechanism still uses fixed parameters or preset update strategies, and cannot adjust the meteorological correction coefficients and other parameters in the model in a timely manner. The accumulation of residuals leads to an inability to effectively eliminate errors in the adjustment calculations, resulting in a gradual increase in the deviation between observed values ​​and model predictions. Ultimately, this prevents the network adjustment model from accurately reflecting the actual measurement situation. By reducing the weight of the observations from the measurement robot, the interference of errors caused by untimely meteorological correction on the adjustment results can be effectively weakened, preventing the continuous accumulation of residuals and reducing the deviation between observed values ​​and model predictions. This allows the network adjustment model to remain as close as possible to the actual measurement situation as possible even in complex meteorological environments. The minimum effective signal-to-noise ratio of the optimized observation data is adjusted based on the response delay of the measurement robot's sensors. As the service life increases, the core components of the sensors experience physical performance degradation, leading to decreased measurement accuracy, data drift, and response delays. This makes it impossible to capture rapid changes in meteorological parameters in a timely manner, resulting in systematic errors in the meteorological correction of distance observations. By increasing the minimum effective signal-to-noise ratio of the optimized observation data, low-efficiency data can be filtered out, unreliable observations caused by aging can be eliminated, preventing contamination of the adjustment results, improving data quality, and suppressing error propagation.

[0040] Furthermore, the method of the present invention adjusts the time window length of the network adjustment model by setting a preset first efficiency and a preset second efficiency. Since the measurement robot has not been calibrated after long-term operation, its performance indicators such as ranging accuracy and angle measurement accuracy will change. This causes the actual observation error to exceed the preset range of the adjustment model. When the model processes these errors, it cannot accurately correct them, resulting in a significant increase in residuals, unit weight error exceeding the normal range, and an increase in the condition number of the normal equation matrix. Ultimately, this leads to a decrease in the stability of the network adjustment model, increased fluctuations in the solution results, and difficulty in meeting the measurement accuracy requirements. By increasing the time window length of the network adjustment model, the model can integrate more historical observation data, enhance the statistical smoothing ability of the measurement robot's uncalibrated error, reduce the weight of a single abnormal observation, and improve the operational stability of the network adjustment model.

[0041] Furthermore, the method of the present invention adjusts the weight of the observations of the measurement robot by setting a preset first update frequency and a preset second update frequency. Since the rigid update mechanism still uses fixed parameters or preset update strategies when meteorological conditions change abruptly, it cannot adjust the meteorological correction coefficients and other parameters in the model in a timely manner. As a result, the adjustment calculation cannot effectively eliminate errors, the residuals continue to accumulate, and the deviation between the observed values ​​and the model predictions gradually increases. Ultimately, the network adjustment model cannot accurately reflect the actual measurement situation. By reducing the weight of the observations of the measurement robot, the interference of errors caused by untimely meteorological correction on the adjustment results can be effectively weakened, the continuous accumulation of residuals can be avoided, and the deviation between the observed values ​​and the model predictions can be reduced. This allows the network adjustment model to still be as close as possible to the actual measurement situation in complex meteorological environments, further improving the operational stability of the network adjustment model.

[0042] Furthermore, the method described in this invention adjusts the minimum effective signal-to-noise ratio of the optimized observation data by setting a preset delay duration. As the service life of the sensor increases, the core components experience physical performance degradation, leading to decreased measurement accuracy, data drift, and response delay. This results in the inability to capture rapid changes in meteorological parameters in a timely manner, which in turn causes systematic errors in the meteorological correction of distance observations. By increasing the minimum effective signal-to-noise ratio of the optimized observation data, low-efficiency data can be filtered out, unreliable observations caused by aging can be eliminated, and contamination of the adjustment results can be avoided, thereby improving data quality, suppressing error propagation, and further enhancing the operational stability of the network adjustment model. Attached Figure Description

[0043] Figure 1 This is an overall flowchart of the real-time meteorological correction and network adjustment model method based on measurement robot collaboration in an embodiment of the present invention;

[0044] Figure 2 This is a flowchart illustrating the process of determining whether to increase the residual truncation threshold of air pressure observation data using a real-time meteorological correction and network adjustment model method based on measurement robot collaboration in an embodiment of the present invention.

[0045] Figure 3 This is a flowchart illustrating the process of determining whether to reduce the sampling interval of air pressure observation data using a real-time meteorological correction and network adjustment model method based on measurement robot collaboration, as described in this embodiment of the invention.

[0046] Figure 4 This is a flowchart illustrating the process of determining whether to increase the minimum effective signal-to-noise ratio of the optimized observation data using a real-time meteorological correction and network adjustment model method based on measurement robot collaboration, as described in this embodiment of the invention. Detailed Implementation

[0047] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0048] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0049] Please see Figure 1 , Figure 2 , Figure 3 as well as Figure 4 The diagrams shown are, respectively, the overall flowchart of the real-time meteorological correction and network adjustment model method based on measurement robot collaboration according to an embodiment of the present invention, the logical flowchart of the process of determining whether to increase the residual truncation threshold of the air pressure observation data, the logical flowchart of the process of determining whether to decrease the sampling interval of the air pressure observation data, and the logical flowchart of the process of determining whether to increase the minimum effective signal-to-noise ratio of the optimized observation data. The present invention provides a real-time meteorological correction and network adjustment model method based on measurement robot collaboration, comprising:

[0050] Step S1: Collect meteorological observation data through meteorological sensors and use several measuring robots to obtain the distance of the measuring points. Determine the number of measuring robots to be connected to the network according to the actual measurement needs.

[0051] Step S2: Preprocess the meteorological observation data to output optimized observation data, construct a meteorological correction model based on the optimized observation data, correct the errors in the meteorological observation data to output corrected observation data, construct a network adjustment model based on the corrected observation data, and update the meteorological correction model based on the network adjustment data generated by the network adjustment model.

[0052] Step S3: Obtain the effective number and total number of network adjustment data respectively, and calculate the effectiveness rate of the network adjustment data;

[0053] Step S4: Determine whether the operational stability of the network adjustment model meets the requirements based on the efficiency of the network adjustment data.

[0054] Step S5: If the operational stability of the network adjustment model does not meet the requirements, determine whether it is necessary to increase the time window length of the network adjustment model.

[0055] Step S6: If it is not necessary to increase the time window length of the network adjustment model, then determine whether the adaptability of the network adjustment model meets the requirements based on the parameter update frequency of the meteorological correction model.

[0056] Step S7: If the adaptability of the network adjustment model does not meet the requirements, determine whether it is necessary to reduce the weight of the observations of the measurement robot.

[0057] Step S8: If it is not necessary to reduce the weight of the observations of the measurement robot, then determine whether it is necessary to increase the minimum effective signal-to-noise ratio of the optimized observation data based on the response delay of the measurement robot's sensors.

[0058] Specifically, meteorological observation data includes air pressure data, temperature data, and humidity data.

[0059] Specifically, preprocessing includes data cleaning, data correction, and normalization.

[0060] Specifically, the optimized observation data includes cleaned temperature data, corrected air pressure data, and normalized humidity data.

[0061] Specifically, meteorological correction models include the 6S model, the MODTRAN model, and the ATCOR model.

[0062] Specifically, the correction includes temperature error correction, air pressure error correction, and humidity error correction. It is a process of systematically correcting errors in the original meteorological observation data and the observation data obtained by the measurement robot by constructing a meteorological correction model.

[0063] Specifically, the corrected observation data includes corrected temperature data, corrected air pressure data, and corrected humidity data.

[0064] Specifically, network adjustment models include the strip method network adjustment, the independent model method network adjustment, and the bundle method network adjustment.

[0065] Specifically, the time window length of the network adjustment model is the time span of the data used by the system within the continuous time range in the model calculation.

[0066] Specifically, the weights of the robot's observations are used to characterize the degree of influence of the observations on the final measurement results.

[0067] Specifically, the minimum effective signal-to-noise ratio of optimized observation data is the minimum acceptable ratio of the effective signal strength to the noise signal strength of the preprocessed optimized observation data.

[0068] Specifically, residuals are the difference between observed values ​​and predicted values ​​calculated by the model.

[0069] In implementation, the method of this invention adjusts the time window length of the network adjustment model based on the validity of the network adjustment data. Because the measuring robot has not been calibrated after long-term operation, its performance indicators such as ranging accuracy and angle measurement accuracy will change. This causes the actual observation error to exceed the preset range of the adjustment model. When processing these errors, the model cannot accurately correct them, resulting in a significant increase in residuals, unit weight error exceeding the normal range, and an increase in the condition number of the normal equation matrix. Ultimately, this leads to a decrease in the stability of the network adjustment model, increased fluctuations in the solution results, and difficulty in meeting measurement accuracy requirements. By increasing the time window length of the network adjustment model, the model can integrate more historical observation data, enhance the statistical smoothing ability of the uncalibrated error of the measuring robot, and reduce the weight of single abnormal observations. The sampling interval of the air pressure observation data is adjusted according to the parameter update frequency of the meteorological correction model. However, when meteorological conditions change abruptly, the rigid update mechanism still uses fixed parameters or preset update strategies, failing to adjust parameters such as the meteorological correction coefficients in the model in a timely manner, leading to adjustments in the network adjustment model. The calculation cannot effectively eliminate errors, and the residuals continue to accumulate. The deviation between the observed values ​​and the model predictions gradually increases, eventually making the network adjustment model unable to accurately reflect the actual measurement situation. By reducing the weight of the observation values ​​of the measurement robot, the interference of errors caused by untimely meteorological correction on the adjustment results can be effectively weakened, the continuous accumulation of residuals can be avoided, and the deviation between the observed values ​​and the model predictions can be reduced. This allows the network adjustment model to still be as close as possible to the actual measurement situation in complex meteorological environments. The minimum effective signal-to-noise ratio of the optimized observation data is adjusted according to the response delay of the measurement robot's sensors. As the service life increases, the core components of the sensors experience physical performance degradation, leading to decreased measurement accuracy, data drift, and response delay. This makes it impossible to capture rapid changes in meteorological parameters in a timely manner, resulting in systematic errors in the meteorological correction of distance observations. By increasing the minimum effective signal-to-noise ratio of the optimized observation data, low-efficiency data can be filtered out, unreliable observations caused by aging can be eliminated, the contamination of the adjustment results can be avoided, data quality can be improved, and error propagation can be suppressed.

[0070] Specifically, determining whether the operational stability of the network adjustment model meets the requirements based on the effectiveness of the network adjustment data includes:

[0071] The effectiveness of the network adjustment data is compared with the preset second effectiveness.

[0072] If the effectiveness of the network adjustment data is greater than the preset second effectiveness, then the operational stability of the network adjustment model is determined to meet the requirements.

[0073] If the effectiveness of the network adjustment data is less than or equal to the preset second effectiveness, then the operational stability of the network adjustment model is determined to be unsatisfactory.

[0074] One reason why the network adjustment data might not meet the requirements could be that the network adjustment model's operational stability is insufficient, or that the time window length of the network adjustment model is inappropriate. The next step is to determine which specific cause it is, which is essentially the process of determining whether the time window length of the network adjustment model needs to be increased.

[0075] Specifically, determining whether the time window length of the network adjustment model needs to be increased includes:

[0076] The effectiveness of the network adjustment data is compared with the preset first effectiveness and the preset second effectiveness, respectively.

[0077] If the effectiveness of the network adjustment data is less than the preset first effectiveness, then it is determined that the time window length of the network adjustment model needs to be increased.

[0078] If the effectiveness of the network adjustment data is greater than or equal to the preset first effectiveness and less than or equal to the preset second effectiveness, then it is determined that it is not necessary to increase the time window length of the network adjustment model.

[0079] It is understandable that the three intervals defined by the preset first efficiency and the preset second efficiency correspond to three different scenarios:

[0080] The first interval is where the effectiveness of the network adjustment data is less than the preset first effectiveness. The corresponding situation is as follows: Due to the lack of calibration after long-term operation of the measurement robot, its performance indicators such as ranging accuracy and angle measurement accuracy will change. This causes the actual observation error to exceed the preset range of the adjustment model. When the model processes these errors, it cannot accurately correct them, resulting in a significant increase in residuals, the unit weight error exceeding the normal range, and an increase in the condition number of the normal equation matrix. Ultimately, this leads to a decrease in the stability of the network adjustment model, increased fluctuations in the solution results, and difficulty in meeting the measurement accuracy requirements. At this time, it is necessary to adjust the time window length of the network adjustment model.

[0081] The second interval is where the effectiveness of the network adjustment data is greater than the first preset effectiveness rate but less than the second preset effectiveness rate. The corresponding situation is as follows: When meteorological conditions change abruptly, the rigid update mechanism still uses fixed parameters or preset update strategies, and cannot adjust the meteorological correction coefficients and other parameters in the model in a timely manner. As a result, the adjustment calculation cannot effectively eliminate errors, the residuals continue to accumulate, and the deviation between the observed values ​​and the model prediction values ​​gradually increases. Ultimately, the network adjustment model cannot accurately reflect the actual measurement situation. At this time, it is necessary to further determine whether the adaptability of the network adjustment model meets the requirements.

[0082] The third interval is when the effectiveness of the network adjustment data is greater than the preset second effectiveness, which corresponds to the situation where the adaptability of the network adjustment model meets the requirements and no adjustment is needed.

[0083] Understandably, the preset efficiency rate can be set according to actual working conditions, aiming to ensure the accuracy and practicality of the test results. Optionally, the preset efficiency rate is initially set based on the evaluation results, then theoretically derived using an error propagation model to verify whether the preset value meets the accuracy requirements after adjustment. Finally, a real-time feedback adjustment mechanism is established, statistically analyzing the residual distribution after adjustment calculation, and dynamically adjusting the preset efficiency rate according to the proportion of out-of-limit residuals until the adjustment result reaches the expected accuracy. For example, the preset first efficiency rate is generally selected within the range of [90%, 94%], and the preset second efficiency rate is generally selected within the range of [95%, 99%].

[0084] Preferably, the first efficiency is 92% in the preferred embodiment, and the second efficiency is 97% in the preferred embodiment.

[0085] In practice, the method of the present invention determines the operational stability of the network adjustment model by setting a preset first efficiency and a preset second efficiency, thereby reducing the impact of inaccurate determination of the operational stability of the network adjustment model on the decrease in the accuracy of real-time meteorological correction, and further improving the operational stability of the network adjustment model.

[0086] Specifically, the effectiveness rate of the network adjustment data is the ratio of the effective number of network adjustment data to the total number of data.

[0087] Specifically, the increase in the time window length of the network adjustment model is determined by the difference between the effectiveness of the network adjustment data and the preset first effectiveness.

[0088] Specifically, when the difference between the effectiveness of the network adjustment data and the preset first effectiveness is within 2%, the time window length of the network adjustment model is increased to 1.2 times the original. When the difference between the effectiveness of the network adjustment data and the preset first effectiveness exceeds 2%, in addition to increasing it to 1.2 times the original, the time window length of the network adjustment model is increased by 1 minute for every 1% exceeding the original. For example, if the difference between the effectiveness of the network adjustment data and the preset first effectiveness is 4%, and the current time window length of the network adjustment model is 5 minutes, the increased time window length of the network adjustment model is 5 × 1.2 + 2 × 1 = 8 minutes.

[0089] In practice, the method of this invention adjusts the time window length of the network adjustment model by setting a preset first efficiency and a preset second efficiency. Since the measuring robot has not been calibrated after long-term operation, its performance indicators such as ranging accuracy and angle measurement accuracy will change. This causes the actual observation error to exceed the preset range of the adjustment model. When processing these errors, the model cannot accurately correct them, resulting in a significant increase in residuals, unit weight error exceeding the normal range, and a larger condition number in the normal equation matrix. Ultimately, this leads to a decrease in the stability of the network adjustment model, increased fluctuations in the solution results, and difficulty in meeting measurement accuracy requirements. By increasing the time window length of the network adjustment model, the model can integrate more historical observation data, enhance the statistical smoothing ability of the uncalibrated error of the measuring robot, reduce the weight of single abnormal observations, and improve the operational stability of the network adjustment model.

[0090] Specifically, the increase in the time window length of the network adjustment model is determined by the difference between the effectiveness of the network adjustment data and the preset first effectiveness.

[0091] Specifically, the suitability of the parameter update frequency based on the meteorological correction function for the network adjustment model is determined, including:

[0092] The parameter update frequency of the meteorological correction model is compared with the preset second update frequency;

[0093] If the parameter update frequency of the meteorological correction model is greater than the preset second update frequency, then it is determined that the adaptability of the network adjustment model meets the requirements, and it is determined whether the time window length of the network adjustment model meets the requirements.

[0094] If the parameter update frequency of the meteorological correction model is less than or equal to the preset second update frequency, then the adaptability of the network adjustment model is determined to be unacceptable.

[0095] Specifically, when the parameter update frequency of the meteorological correction model is greater than the preset second update frequency, it is determined that the adaptability of the network adjustment model meets the requirements. However, if it has been previously determined that the operational stability of the network adjustment model does not meet the requirements, then it is necessary to further determine whether the time window length of the network adjustment model meets the requirements.

[0096] In implementation, the time window length of the actual network adjustment model is compared with the predetermined residual truncation threshold to determine whether the time window length of the network adjustment model meets the requirements. If the time window length of the actual network adjustment model is less than or equal to the predetermined residual truncation threshold, the time window length of the network adjustment model is determined to be unacceptable.

[0097] If the time window length of the network adjustment model does not meet the requirements, the time window length of the network adjustment model is increased; if the time window length of the network adjustment model meets the requirements, the effective number and total number of network adjustment data are re-acquired, and the operational stability of the network adjustment model is reassessed.

[0098] When the parameter update frequency of the meteorological correction model is less than or equal to the preset second update frequency, the reason for the network adjustment model's operational stability not meeting requirements can be determined as the network adjustment model's adaptability not meeting requirements. The reason for the network adjustment model's adaptability not meeting requirements may be that the weights of the measurement robot's observations are not appropriate, or that the accuracy of real-time meteorological correction is not appropriate. The next step is to determine which specific reason it is; this process is also the process of determining whether the weights of the measurement robot's observations need to be reduced.

[0099] It is understandable that the three intervals divided by the preset first update frequency and the preset second update frequency correspond to three different scenarios:

[0100] The first interval is when the parameter update frequency of the meteorological correction model is less than or equal to the preset first update frequency. The corresponding situation is: as the years of use increase, the physical performance of the core components of the sensor deteriorates, resulting in decreased measurement accuracy, data drift and response delay. It is unable to capture rapid changes in meteorological parameters in time, which leads to systematic errors in the meteorological correction of distance observations. At this time, it is necessary to further determine whether the accuracy of real-time meteorological correction meets the requirements.

[0101] The second interval is when the parameter update frequency of the meteorological correction model is greater than the preset first update frequency and less than or equal to the preset second pressure. The corresponding situation is as follows: When meteorological conditions change abruptly, the rigid update mechanism still uses fixed parameters or preset update strategies and cannot adjust the meteorological correction coefficients and other parameters in the model in a timely manner. As a result, the adjustment calculation cannot effectively eliminate errors, the residuals continue to accumulate, and the deviation between the observed values ​​and the model prediction values ​​gradually increases. In the end, the network adjustment model cannot accurately reflect the actual measurement situation. At this time, it is necessary to adjust the weight of the observation values ​​of the measurement robot.

[0102] The third interval is when the parameter update frequency of the meteorological correction model is greater than the preset second update frequency. The corresponding situation is: the adaptability of the network adjustment model is determined to meet the requirements, and no adjustment is required.

[0103] Understandably, the preset update frequency can be set according to actual working conditions, aiming to ensure the accuracy and practicality of the test results. Optionally, the preset update frequency is verified through simulation, testing the impact of different update frequencies on correction accuracy under typical working conditions, determining the critical effective frequency, and finally establishing a dynamic feedback mechanism to trigger forced parameter updates using real-time residual analysis. For example, the preset first update frequency is generally selected in the range of [1 time / minute, 3 times / minute], and the preset second update frequency is generally selected in the range of [4 times / minute, 6 times / minute].

[0104] Preferably, the first preset update frequency is 2 times / minute in a preferred embodiment, and the second preset update frequency is 5 times / minute in a preferred embodiment.

[0105] In practice, the method of the present invention determines the adaptability of the network adjustment model by setting a preset first update frequency and a preset second update frequency, thereby reducing the impact of the decrease in the operational stability of the network adjustment model due to inaccurate determination of the adaptability of the network adjustment model, and further improving the operational stability of the network adjustment model.

[0106] Specifically, the reduction in the weight of the observations of the measuring robot is determined by the difference between the parameter update frequency of the meteorological correction model and the preset first update frequency.

[0107] Specifically, when the difference between the parameter update frequency of the meteorological correction model and the preset first update frequency is within 1 time / minute, the weight of the measurement robot's observations is reduced to 0.8 times the original value. When the difference between the parameter update frequency of the meteorological correction model and the preset first update frequency exceeds 1 time / minute, in addition to reducing it to 0.8 times the original value, for every 2 times / minute the difference exceeds 1 time / minute, the weight of the measurement robot's observations is reduced by 0.05. For example, if the difference between the parameter update frequency of the meteorological correction model and the preset first update frequency is 3 times / minute, and the current weight of the measurement robot's observations is 0.5, the reduced weight of the measurement robot's observations will be 0.5 × 0.8 - 1 × 0.05 = 0.35.

[0108] In practice, the method of this invention adjusts the sampling interval of air pressure observation data by setting a preset first update frequency and a preset second update frequency. When meteorological conditions change abruptly, the rigid update mechanism still uses fixed parameters or preset update strategies, and cannot adjust parameters such as meteorological correction coefficients in the model in a timely manner. This results in the adjustment calculation failing to effectively eliminate errors, residuals accumulating continuously, and the deviation between observed values ​​and model predictions gradually increasing. Ultimately, this makes the network adjustment model unable to accurately reflect the actual measurement situation. By reducing the weight of the observation values ​​of the measurement robot, the interference of errors caused by untimely meteorological correction on the adjustment results can be effectively weakened, the continuous accumulation of residuals can be avoided, and the deviation between observed values ​​and model predictions can be reduced. This allows the network adjustment model to still be as close as possible to the actual measurement situation in complex meteorological environments, further improving the operational stability of the network adjustment model.

[0109] Specifically, determining whether to increase the minimum effective signal-to-noise ratio of the optimized observation data based on the response delay of the robot's sensors includes:

[0110] The response delay of the robot's sensors is compared with the preset delay.

[0111] If the response delay of the measurement robot's sensor is less than or equal to the preset delay, then it is determined that there is no need to increase the minimum effective signal-to-noise ratio of the optimized observation data, and it is determined whether the weight of the measurement robot's observations meets the requirements.

[0112] If the response delay of the measurement robot sensor is greater than the preset delay, then it is determined that the minimum effective signal-to-noise ratio of the optimized observation data needs to be increased, and the minimum effective signal-to-noise ratio of the optimized observation data is increased.

[0113] Specifically, when the response delay of the measurement robot's sensor is less than or equal to the preset delay, it is determined that the accuracy of the real-time weather correction meets the requirements. However, if the adaptability of the previously determined network adjustment model does not meet the requirements, it is necessary to further determine whether the weights of the measurement robot's observations meet the requirements.

[0114] In practice, the weight of the actual measurement robot's observations is compared with a predetermined sampling interval threshold to determine whether the weight of the actual measurement robot's observations meets the requirements. If the weight of the actual measurement robot's observations is greater than or equal to the predetermined sampling interval threshold, the weight of the measurement robot's observations is determined to be unacceptable. The predetermined weight threshold for the measurement robot's observations is the average weight of the measurement robot's observations monitored in the previous three months of the historical period.

[0115] If the weight of the robot's observations does not meet the requirements, the weight of the robot's observations will be reduced; if the weight of the robot's observations meets the requirements, the response delay of the robot's sensors will be re-acquired, and the accuracy of the real-time weather correction will be reassessed.

[0116] When the response delay of the robot sensor exceeds the preset delay, it can be determined that the reason why the adaptability of the network adjustment model does not meet the requirements is that the accuracy of the real-time meteorological correction does not meet the requirements. Therefore, it is necessary to increase the minimum effective signal-to-noise ratio of the optimized observation data.

[0117] It is understandable that the two pre-defined intervals for efficient division correspond to two different scenarios:

[0118] The first interval is when the response delay of the robot sensor is less than the preset delay, which corresponds to the situation where the accuracy of the real-time weather correction meets the requirements.

[0119] The second interval is when the response delay of the robot sensor is greater than or equal to the preset delay time. The corresponding situation is that as the years of use increase, the physical performance of the core components of the sensor deteriorates, resulting in decreased measurement accuracy, data drift and response delay. It is unable to capture rapid changes in meteorological parameters in time, which in turn leads to systematic errors in the meteorological correction of distance observations.

[0120] Understandably, the preset delay duration can be set according to actual working conditions, aiming to ensure the accuracy and practicality of the test results. Optionally, a dynamic compensation mechanism can be established for the preset delay duration, embedding a delay compensation algorithm in the robot control software and periodically updating the preset value according to changes in working conditions to ensure that the delay control matches the actual operating state. For example, the preset delay duration is generally selected in the range of [6ms, 10ms].

[0121] Preferably, the preset delay duration is 8ms in the preferred embodiment.

[0122] In practice, the method of the present invention determines the accuracy of real-time weather correction by setting a preset delay time, thereby reducing the impact of inaccurate determination of the accuracy of real-time weather correction on the operational stability of the network adjustment model and further improving the operational stability of the network adjustment model.

[0123] Specifically, the increase in the minimum effective signal-to-noise ratio of the optimized observation data is determined by measuring the difference between the response delay of the robot sensor and the preset delay.

[0124] Specifically, when the difference between the response delay of the robot sensor and the preset delay is within 2ms, the minimum effective signal-to-noise ratio (SNR) of the optimized observation data is increased to 1.2 times the original value. When the difference exceeds 2ms, in addition to increasing to 1.2 times the original value, the minimum effective SNR of the optimized observation data increases by 2dB for every 1ms exceeding the original value. For example, when the difference between the response delay of the robot sensor and the preset delay is 3ms, the current minimum effective SNR of the optimized observation data is 20dB. The increased minimum effective SNR of the optimized observation data is 20×1.2+1×2=26dB.

[0125] In practice, the method described in this invention adjusts the minimum effective signal-to-noise ratio of optimized observation data by setting a preset delay time. As the service life of the sensor increases, the core components experience physical performance degradation, leading to decreased measurement accuracy, data drift, and response delay. This results in the inability to capture rapid changes in meteorological parameters in a timely manner, which in turn causes systematic errors in the meteorological correction of distance observations. By increasing the minimum effective signal-to-noise ratio of optimized observation data, low-efficiency data can be filtered out, unreliable observations caused by aging can be eliminated, and the contamination of adjustment results can be avoided, thereby improving data quality, suppressing error propagation, and further enhancing the operational stability of the network adjustment model.

[0126] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for real-time meteorological correction and network adjustment model based on collaborative measurement robots, characterized in that, include: Meteorological observation data is collected by meteorological sensors and the distance between measurement points is obtained by several measurement robots. The number of measurement robots connected to the network is determined according to the actual measurement needs. The meteorological observation data is preprocessed to output optimized observation data. A meteorological correction model is constructed based on the optimized observation data. Error correction is performed on the meteorological observation data to output corrected observation data. A network adjustment model is constructed based on the corrected observation data. The meteorological correction model is updated based on the network adjustment data generated by the network adjustment model. Obtain the effective number and total number of network adjustment data respectively, and calculate the effectiveness rate of the network adjustment data; The efficiency of network adjustment data is used to determine whether the operational stability of the network adjustment model meets the requirements. If the operational stability of the network adjustment model does not meet the requirements, then determine whether it is necessary to increase the time window length of the network adjustment model. If it is not necessary to increase the time window length of the network adjustment model, then the suitability of the network adjustment model is determined based on the parameter update frequency of the meteorological correction model. If the adaptability of the network adjustment model does not meet the requirements, then determine whether it is necessary to reduce the weight of the observations of the measurement robot. If it is not necessary to reduce the weight of the observations of the measurement robot, then determine whether it is necessary to increase the minimum effective signal-to-noise ratio of the optimized observation data based on the response delay of the measurement robot's sensors; The parameter update frequency of the meteorological correction model is compared with the preset second update frequency; If the parameter update frequency of the meteorological correction model is greater than the preset second update frequency, then it is determined that the adaptability of the network adjustment model meets the requirements, and it is determined whether the time window length of the network adjustment model meets the requirements. If the parameter update frequency of the meteorological correction model is less than or equal to the preset second update frequency, then the adaptability of the network adjustment model is determined to be unacceptable. The response delay of the robot's sensors is compared with the preset delay. If the response delay of the measurement robot's sensor is less than or equal to the preset delay, then it is determined that there is no need to increase the minimum effective signal-to-noise ratio of the optimized observation data, and it is determined whether the weight of the measurement robot's observations meets the requirements. If the response delay of the measurement robot sensor is greater than the preset delay, then it is determined that the minimum effective signal-to-noise ratio of the optimized observation data needs to be increased, and the minimum effective signal-to-noise ratio of the optimized observation data is increased.

2. The method for real-time meteorological correction and network adjustment model based on measurement robot collaboration according to claim 1, characterized in that, Determining whether the operational stability of the network adjustment model meets the requirements based on the effectiveness of the network adjustment data includes: The effectiveness of the network adjustment data is compared with the preset second effectiveness. If the effectiveness of the network adjustment data is greater than the preset second effectiveness, then the operational stability of the network adjustment model is determined to meet the requirements. If the effectiveness of the network adjustment data is less than or equal to the preset second effectiveness, then the operational stability of the network adjustment model is determined to be unsatisfactory.

3. The method for real-time meteorological correction and network adjustment model based on measurement robot collaboration according to claim 2, characterized in that, Determine whether the time window length of the network adjustment model needs to be increased, including: The effectiveness of the network adjustment data is compared with the preset first effectiveness and the preset second effectiveness, respectively. If the effectiveness of the network adjustment data is less than the preset first effectiveness, then it is determined that the time window length of the network adjustment model needs to be increased. If the effectiveness of the network adjustment data is greater than or equal to the preset first effectiveness and less than or equal to the preset second effectiveness, then it is determined that it is not necessary to increase the time window length of the network adjustment model.

4. The real-time meteorological correction and network adjustment model method based on measurement robot collaboration according to claim 3, characterized in that, The effectiveness rate of the network adjustment data is the ratio of the effective number of network adjustment data to the total number of data.

5. The real-time meteorological correction and network adjustment model method based on measurement robot collaboration according to claim 4, characterized in that, The increase in the time window length of the network adjustment model is determined by the difference between the effectiveness of the network adjustment data and the preset first effectiveness.

6. The real-time meteorological correction and network adjustment model method based on measurement robot collaboration according to claim 5, characterized in that, Determine whether the weights of the robot's observations need to be reduced, including: The parameter update frequency of the meteorological correction model is compared with the preset first update frequency and the preset second update frequency, respectively; If the parameter update frequency of the meteorological correction model is greater than the preset first update frequency and less than or equal to the preset second update frequency, then it is determined that the weight of the observation value of the measurement robot does not meet the requirements, and the weight of the observation value of the measurement robot is reduced. If the parameter update frequency of the meteorological correction model is less than or equal to the preset first update frequency, then it is determined that there is no need to reduce the weight of the observations of the measuring robot.

7. The method for real-time meteorological correction and network adjustment model based on measurement robot collaboration according to claim 6, characterized in that, The reduction in the weight of the observations of the measuring robot is determined by the difference between the parameter update frequency of the meteorological correction model and the preset first update frequency.

8. The method for real-time meteorological correction and network adjustment model based on collaborative measurement robots according to claim 7, characterized in that, The increase in the minimum effective signal-to-noise ratio of the optimized observation data is determined by measuring the difference between the response delay of the robot sensor and the preset delay.

Images