Unmanned aerial vehicle measurement system based on low-altitude meteorological high-precision positioning

Abstract

The invention relates to the technical field of unmanned aerial vehicle measurement, in particular to an unmanned aerial vehicle measurement system based on low-altitude meteorological high-precision positioning, which comprises a low-altitude meteorological sensing unit, a positioning fusion calculation unit, a cooperative calibration control unit and a self-adaptive measurement unit. The positioning fusion calculation unit uses an atmospheric delay correction model and fuzzy adaptive Kalman filtering to compensate an atmospheric delay error and adjust a positioning noise matrix to output accurate track coordinates, and cooperates with the calibration control unit to establish a relation between a humidity gradient and a positioning height error, obtain an optimal compensation coefficient and convert the optimal compensation coefficient into a flight control instruction, and finally control the flight control instruction. And the adaptive measurement unit controls the multispectral imager to switch wavebands and adjust the scanning angle of the laser radar according to the flight path and the meteorological parameters, so that the aerosol vertical profile synchronous inversion is realized, and the positioning accuracy of the unmanned aerial vehicle and the reliability of measurement data under the complex meteorological condition are improved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) measurement technology, and in particular to an UAV measurement system based on low-altitude meteorological high-precision positioning. Background Art

[0002] Drone measurement is an important technology. In the field of drone measurement technology, the positioning accuracy of drones plays a decisive role in the accuracy of measurement results. However, traditional drone positioning methods have serious defects in low-altitude meteorological environments.

[0003] The low-altitude meteorological environment is complex and changeable. Meteorological factors such as temperature and humidity gradients and pressure gradients will have a significant impact on the positioning of drones. Atmospheric delays in the ionosphere and troposphere will cause positioning errors. Traditional positioning systems lack algorithms and models that can fully consider low-altitude meteorological factors. When processing positioning data, they cannot effectively integrate meteorological parameters such as temperature and humidity gradients and pressure gradients into the calculation process. When calculating positioning coordinates, traditional algorithms do not make targeted corrections to atmospheric delays, resulting in an inability to adapt to complex meteorological changes, making it difficult to accurately compensate for this error, causing the actual trajectory of the drone to deviate from the expected trajectory. In addition, the noise in the positioning process The interference problem is prominent. Conventional positioning algorithms usually rely on pre-set parameters and fixed processing procedures, and lack the ability to monitor and adjust meteorological parameters in real time. They cannot obtain real-time change information of meteorological parameters such as the horizontal pressure gradient field in a timely manner, and cannot dynamically adjust the positioning noise matrix according to these changes. When the horizontal pressure gradient field changes suddenly, the traditional algorithm cannot perceive and respond quickly, and continues to use the original noise matrix for positioning calculations, thereby affecting the positioning accuracy and causing deviations in the measurement data. In order to solve this technical problem, we provide a drone measurement system based on low-altitude meteorological high-precision positioning. Summary of the Invention

[0004] The object of the present invention is to provide a UAV measurement system based on low-altitude meteorological high-precision positioning to solve the problems raised in the above background technology.

[0005] Traditional drone positioning methods in low-altitude meteorological environments lack algorithms and models that consider low-altitude meteorological factors, making it difficult to compensate for positioning errors caused by atmospheric delay, which can cause deviations between the drone's actual and expected tracks. Therefore, this case uses a layered atmospheric delay correction model through the positioning fusion calculation unit. This model divides the low-altitude meteorological environment into layers by altitude, calculates and accumulates the ionospheric / tropospheric delay corrections for each layer, and accurately calculates the impact of atmospheric delay on positioning and effectively compensates for it, improving drone positioning accuracy and reducing track deviations.

[0006] Conventional positioning algorithms rely on preset parameters and fixed processes, lacking the ability to monitor and adjust meteorological parameters in real time and provide feedback. They are unable to dynamically adjust the positioning noise matrix based on changes in meteorological parameters such as the horizontal pressure gradient field, which affects positioning accuracy. Therefore, this case introduces a fuzzy adaptive mechanism into the positioning fusion calculation unit when using the Kalman filter. Based on the range and rate of change of the horizontal pressure gradient field parameters, the noise covariance matrix of the Kalman filter is dynamically adjusted through a fuzzy rule base. This allows the positioning noise matrix to be adjusted in real time based on changes in meteorological parameters, improving positioning accuracy and ensuring the accuracy of drone measurement data.

[0007] To achieve the above objectives, a UAV measurement system based on low-altitude meteorological high-precision positioning is provided, which includes the following collaborative units: The low-altitude meteorological sensing unit collects three-dimensional wind speed vectors, temperature stratification, absolute humidity, and horizontal pressure gradient fields through a sensor array; The positioning fusion calculation unit receives the three-dimensional wind speed vector and temperature stratification parameters, substitutes the temperature and humidity gradient parameters into the refractive index calculation formula through the atmospheric delay correction model, calculates the ionosphere / troposphere joint correction, adjusts the positioning noise matrix based on the horizontal pressure gradient field parameters using Kalman filtering, and outputs track coordinates; The collaborative calibration control unit establishes a linear relationship between the humidity gradient and the positioning height error based on the track coordinates and the absolute humidity parameters, constructs a cost function including the heading angle deviation and the humidity gradient observation value, obtains the optimal compensation coefficient through the QR decomposition method, and then converts the optimal compensation coefficient into a flight control instruction; The adaptive measurement unit controls the multispectral imager to automatically switch bands based on the track coordinates and temperature stratification parameters, and adjusts the lidar scanning angle based on the horizontal pressure gradient field parameters to achieve synchronous inversion of the aerosol vertical profile.

[0008] As a further improvement of this technical solution, in the positioning fusion calculation unit, the atmospheric delay correction model adopts a hierarchical modeling approach: The low-altitude meteorological environment is divided into multiple layers according to altitude. The temperature and humidity gradient parameters in each layer are considered to be uniformly distributed. For each layer, the temperature and humidity gradient parameters are substituted into the refractive index calculation formula to calculate the ionospheric / tropospheric delay correction for each layer. Finally, the corrections for each layer are accumulated to obtain the total ionospheric / tropospheric joint correction.

[0009] As a further improvement of this technical solution, in the positioning fusion calculation unit, when using Kalman filtering to adjust the positioning noise matrix, a fuzzy adaptive mechanism is introduced: According to the variation range and variation rate of the horizontal pressure gradient field parameters, the process noise covariance matrix and the measurement noise covariance matrix of the Kalman filter are dynamically adjusted through the fuzzy rule base. The fuzzy rule base divides the horizontal pressure gradient field into different fuzzy levels according to its size and variation trend, and sets corresponding adjustment coefficients for each level.

[0010] As a further improvement of this technical solution, the positioning fusion calculation unit adopts a track smoothing algorithm based on particle filtering when outputting track coordinates: A set of particles is generated based on the preliminary track coordinates obtained by Kalman filtering. Each particle represents a possible track state. The weight of each particle is then calculated based on the horizontal pressure gradient field and the ionosphere / troposphere joint correction. Particles whose weights exceed the preset weight threshold are retained through a resampling process. Finally, the retained particles are weighted averaged to obtain the smoothed track coordinates.

[0011] As a further improvement of the present technical solution, in the collaborative calibration control unit, when establishing the linear relationship between the humidity gradient and the positioning height error, the least squares method is used for parameter estimation: Multiple sets of track coordinates and absolute humidity parameters under different flight conditions were collected, the humidity gradient and positioning altitude error were calculated, and then the linear equation between the humidity gradient and positioning altitude error was fitted using the least squares method to obtain the optimal parameter estimation value.

[0012] As a further improvement of the present technical solution, in the collaborative calibration control unit, the constructed cost function adopts the form of weighted least squares method: Cost function ;in is the actual heading angle deviation, is the estimated value of heading angle deviation, is the observed value of humidity gradient, is the estimated value of humidity gradient, is the weight coefficient, which is set according to historical data. is the sample size.

[0013] As a further improvement of the present technical solution, in the collaborative calibration control unit, when obtaining the optimal compensation coefficient by the QR decomposition method, the cost function is expressed in matrix form ,in is the coefficient matrix, is the compensation coefficient vector to be determined, is the observation vector, is the weight matrix, For the transpose operation, the coefficient matrix Perform QR decomposition and get ,in is an orthogonal matrix, For an upper triangular matrix, then solve Get the optimal compensation coefficient vector .

[0014] As a further improvement of the present technical solution, in the collaborative calibration control unit, a proportional-integral-differential controller is used when converting the optimal compensation coefficient into a flight control instruction: The heading angle and altitude adjustment values ​​are calculated based on the optimal compensation coefficient and used as inputs of the proportional-integral-differential controller. The corresponding control value is calculated through the proportional, integral, and differential links of the proportional-integral-differential controller, and the control value is converted into a flight control command and sent to the UAV.

[0015] As a further improvement of the present technical solution, in the adaptive measurement unit, a spectrum matching-based method is adopted when controlling the multispectral imager to automatically switch bands according to the track coordinates and temperature stratification parameters: A spectral library of typical landforms under different temperature stratifications is established. The current measurement area of ​​the UAV is determined according to the track coordinates, and the temperature stratification parameters of the area are obtained. The spectral data collected by the multispectral imager is then matched with the spectra in the spectral library, and the band with the highest matching degree is selected as the current measurement band.

[0016] As a further improvement of this technical solution, in the adaptive measurement unit, a fuzzy control algorithm is used when adjusting the laser radar scanning angle based on the horizontal pressure gradient field parameters: The horizontal pressure gradient field parameters are divided into different fuzzy levels. The scanning angle adjustment amount of the lidar is determined according to the different fuzzy levels through the fuzzy rule base. The fuzzy rule base is designed according to the influence of the horizontal pressure gradient field on the vertical profile distribution of aerosols.

[0017] Compared with the prior art, the present invention has the following beneficial effects: In the UAV measurement system based on low-altitude meteorological high-precision positioning, the positioning fusion calculation unit can effectively compensate for the positioning error caused by atmospheric delay through the hierarchical modeling atmospheric delay correction model and fuzzy adaptive Kalman filter adjustment, dynamically adjust the positioning noise matrix, greatly improve the UAV positioning accuracy, and make the track more accurate. The collaborative calibration control unit establishes a linear relationship between the humidity gradient and the positioning height error, constructs a cost function and obtains the optimal compensation coefficient, converts it into flight control instructions, and can calibrate the positioning deviation in real time to ensure flight stability. The adaptive measurement unit controls the multispectral imager to switch bands and adjust the lidar scanning angle based on the track coordinates, temperature stratification and horizontal pressure gradient field parameters, thereby realizing the synchronous inversion of the aerosol vertical profile and improving the pertinence and accuracy of the measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of the overall workflow of the present invention.

[0019] The meaning of each number in the figure is: 1. Low-altitude meteorological sensing unit; 2. Positioning fusion calculation unit; 3. Collaborative calibration control unit; 4. Adaptive measurement unit. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] The present invention provides a UAV measurement system based on low-altitude meteorological high-precision positioning, please refer to Figure 1 As shown, it includes the following units working together: The low-altitude meteorological sensing unit 1 collects three-dimensional wind speed vectors, temperature stratification, absolute humidity, and horizontal pressure gradient fields through a sensor array; Positioning fusion calculation unit 2 receives the three-dimensional wind speed vector and temperature stratification parameters, substitutes the temperature and humidity gradient parameters into the refractive index calculation formula through the atmospheric delay correction model, calculates the ionosphere / troposphere joint correction, uses Kalman filtering to adjust the positioning noise matrix based on the horizontal pressure gradient field parameters, and outputs the track coordinates; In the positioning fusion calculation unit 2, the atmospheric delay correction model adopts a hierarchical modeling approach: The low-altitude meteorological environment is divided into multiple layers according to the height. The temperature and humidity gradient parameters in each layer are considered to be uniformly distributed, which simplifies the subsequent calculation and improves the accuracy and efficiency of the calculation. For each layer, the temperature and humidity gradient parameters measured by multiple sensors in the layer are collected, and then the average value of these parameters is calculated as the representative value of the temperature and humidity gradient parameters of the layer. The temperature and humidity gradient parameters are respectively substituted into the refractive index calculation formula ;in is the refractive index, is the air pressure, is the temperature, is the water vapor pressure, The refractive index of the layer is calculated using the empirical constant. Then, the ionospheric / tropospheric delay correction for each layer is calculated based on the signal propagation path length and refractive index within the layer. Finally, the corrections for each layer are accumulated to obtain the total ionospheric / tropospheric joint correction. In actual positioning calculations, using this total correction can significantly improve positioning accuracy and reliability and reduce positioning errors caused by atmospheric delay. In the positioning fusion calculation unit 2, a fuzzy adaptive mechanism is introduced when using Kalman filtering to adjust the positioning noise matrix: The sensor array in the low-altitude meteorological sensor unit 1 collects air pressure data in real time, calculates the air pressure difference between adjacent positions, and obtains the horizontal air pressure gradient field parameters. At the same time, the horizontal air pressure gradient field parameters at different times are recorded, and their change rate is calculated, that is, the difference between the horizontal air pressure gradient field parameters at adjacent times is divided by the time interval. According to the change range and change rate of the horizontal air pressure gradient field parameters, the process noise covariance matrix and the measurement noise covariance matrix of the Kalman filter are dynamically adjusted through the fuzzy rule base. The fuzzy rule base divides the horizontal air pressure gradient field into different fuzzy levels according to the size and change trend of the horizontal air pressure gradient field, and sets a corresponding adjustment coefficient for each level. For example, the size of the horizontal air pressure gradient field can be divided into three levels of "small", "medium" and "large", and the change trend can be divided into three levels of "slow", "medium" and "fast". The corresponding adjustment coefficient is set for each fuzzy level combination. For example, when the size of the horizontal air pressure gradient field is "small" and the change trend is "slow", the process noise covariance matrix is ​​set. and the measurement noise covariance matrix The adjustment coefficient is 0.8. When the horizontal pressure gradient field size is "large" and the change trend is "fast", the adjustment coefficient is set to 1.2. According to different horizontal pressure gradient field conditions, the noise matrix can be adjusted quickly and accurately, and the adaptability of the Kalman filter to different meteorological environments can be improved. The current horizontal pressure gradient field size and change rate are mapped to the corresponding fuzzy level to obtain the membership of each fuzzy level. For example, the current horizontal pressure gradient field size is 5hPa / km. According to the threshold value of the fuzzy level division, the membership of the "small" level is 0.2, the membership of the "medium" level is 0.8, and the membership of the "large" level is 0. According to the membership obtained by fuzzification processing, the corresponding rules are found in the fuzzy rule base, and reasoning is performed according to the weight of the rules to obtain the final adjustment coefficient. For example, the process noise covariance matrix is ​​calculated by the weighted average method. The adjustment coefficient is 0.9. Under different meteorological conditions, the noise matrix is ​​adjusted through the fuzzy adaptive mechanism, which can significantly improve the positioning accuracy and reduce the positioning error. In the positioning fusion calculation unit 2, a track smoothing algorithm based on particle filtering is used when outputting track coordinates: Although Kalman filtering can give preliminary track coordinates, the actual environment is uncertain, and a single coordinate cannot fully reflect the track state. A set of particles is generated based on the preliminary track coordinates obtained by Kalman filtering. Each particle represents a possible track state. Assume that the preliminary track coordinates obtained by Kalman filtering are , with the coordinate as the center, randomly generated according to Gaussian distribution within a certain range particles , each particle Represents a possible track state. Using particle swarms to describe the track state can more comprehensively consider the possible uncertainty of the track and provide a basis for subsequent more accurate track estimation. The weight of each particle is then calculated based on the horizontal pressure gradient field and the ionosphere / troposphere joint correction. For each particle , combined with the horizontal pressure gradient field information and the combined ionosphere / troposphere correction Construct a likelihood function ,in Calculate the weight of each particle based on the position information obtained by the sensor , initial weight ,Then the weights of all particles are normalized, so that the particle filter algorithm can adaptively adjust the degree of trust in different track states according to environmental factors, improving the accuracy of track estimation, and retain particles whose weights exceed the preset weight threshold through the resampling process. Finally, the retained particles are weighted averaged to obtain the smoothed track coordinates, and a preset weight threshold is set. , traverse all particles and filter out weights The particles are copied and added to the particle set, so that the number of particles in the set is restored to This avoids the problem of inaccurate track estimation caused by particle degradation, improves the stability and reliability of the algorithm, and , respectively calculate their The weighted average in the direction is used to mark the smoothed track coordinates as In practical applications, it can significantly reduce the fluctuation of the track, improve the continuity and accuracy of the track, and provide more reliable positioning information for subsequent UAV control and measurement.

[0022] The collaborative calibration control unit 3 establishes a linear relationship between the humidity gradient and the positioning height error based on the track coordinates and the absolute humidity parameters, and constructs a cost function that includes the heading angle deviation and the humidity gradient observation value. The optimal compensation coefficient is obtained through the QR decomposition method, and then the optimal compensation coefficient is converted into a flight control instruction. In the collaborative calibration control unit 3, the least squares method is used to estimate parameters when establishing the linear relationship between the humidity gradient and the positioning height error: Under different flight states of the UAV, the track coordinates and absolute humidity parameters are continuously recorded, and multiple sets of track coordinates and absolute humidity parameters under different flight states are collected. The track coordinates can be obtained by the output of the positioning fusion calculation unit 2, and the absolute humidity parameters are measured by the humidity sensor in the low-altitude meteorological sensing unit 1. The recorded data should contain a timestamp for subsequent accurate matching and analysis. For the collected absolute humidity parameters, two adjacent measurement points are selected, the humidity difference between them is calculated, and divided by the vertical distance between the two points to obtain the humidity gradient. The height value in the track coordinates is subtracted from the known true height value to obtain the positioning height error, which provides accurate data for the subsequent least squares fitting and helps to obtain a more accurate linear equation. The linear equation between the humidity gradient and the positioning height error is then fitted by the least squares method to obtain the optimal parameter estimate. In the subsequent UAV flight process, the positioning height error can be predicted based on the real-time measured humidity gradient using this equation, and corresponding calibration can be performed. In practical applications, the positioning height error caused by the humidity gradient can be effectively reduced, thereby improving the overall performance of the UAV measurement system. In the collaborative calibration control unit 3, the constructed cost function adopts the form of weighted least squares method: In order to comprehensively consider the errors of heading angle deviation and humidity gradient observation value, it is necessary to construct a cost function. According to system requirements and error analysis, the cost function is determined. ;in is the actual heading angle deviation, is the estimated value of heading angle deviation, is the observed value of humidity gradient, is the estimated value of humidity gradient, is the weight coefficient, which is set according to historical data. is the number of samples, the actual heading angle deviation The humidity gradient observation value can be obtained by comparing the gyroscope with the current heading angle of the system. The heading angle deviation estimate is measured and calculated by the humidity sensor in the low-altitude meteorological sensing unit 1. and humidity gradient estimates The prediction can be made based on the previously established linear relationship model between humidity gradient and positioning height error. The weight coefficient is determined according to factors such as the source of the data, measurement accuracy and the degree of impact on system performance. The actual data collected is used to 、 and estimated data 、 The determined weight coefficient is substituted into the cost function for calculation, and the cost function is optimized using an optimization algorithm. According to the derivative information of the cost function with respect to the parameter, the parameter value is continuously updated until the cost function reaches the minimum value or meets the preset convergence condition. After the parameter adjustment, the system's heading angle deviation and humidity gradient estimation error are effectively reduced, thereby improving the overall performance of the UAV measurement system. In order to facilitate the use of QR decomposition method for solving, the cost function needs to be converted into a matrix form. In the collaborative calibration control unit 3, when obtaining the optimal compensation coefficient by QR decomposition method, the cost function is expressed in matrix form ,in is the coefficient matrix, is the compensation coefficient vector to be determined, is the observation vector, is the weight matrix, For the transpose operation, the coefficient matrix Perform QR decomposition and get ,in is an orthogonal matrix, For an upper triangular matrix, then solve Get the optimal compensation coefficient vector , the optimal compensation coefficient vector obtained is Bring it back to the cost function, calculate the value of the cost function, and check whether the value meets the preset convergence conditions. At the same time, through actual flight tests or simulation experiments, observe whether the system performance is improved after calibration using the compensation coefficient, avoid using the wrong compensation coefficient for calibration, and ensure that the UAV measurement system can operate stably and accurately; In the collaborative calibration control unit 3, a proportional-integral-differential controller is used to convert the optimal compensation coefficient into flight control instructions: The optimal compensation coefficient is obtained to correct the deviation of the UAV in heading and altitude. It needs to be converted into specific heading angle and altitude adjustment to provide a clear target for subsequent control. The heading angle and altitude adjustment are calculated based on the optimal compensation coefficient. The heading angle and altitude adjustment are used as the input of the proportional-integral-differential controller. The corresponding control quantity is calculated through the proportional, integral and differential links of the proportional-integral-differential controller, and the control quantity is converted into flight control instructions and sent to the UAV. For heading angle control, the output of the proportional link ,in is the heading angle scale factor, Is the heading angle error, that is, the difference between the desired heading angle and the actual heading angle. For altitude control, the output of the proportional link is ,in is the height scale factor, It is the height error, that is, the difference between the expected height and the actual height. The integral link accumulates the error, and the output of the heading angle integral link is ,in is the heading angle integral coefficient, It is The heading angle error at the moment, the output of the altitude integration link ,in is the height integration coefficient, It is The height error at the moment, the differential link is adjusted according to the rate of change of the error, and the output of the heading angle differential link ,in is the heading angle differential coefficient, is the sampling time interval, the output of the highly differential link ,in is the height differential coefficient. Finally, the outputs of the three links are added together to obtain the total control amount, the heading angle control amount , height control amount The three links of the proportional-integral-differential controller cooperate with each other, which can comprehensively consider the current value, historical accumulated value and change trend of the error to achieve more accurate and stable control. According to the interface protocol and requirements of the UAV's flight control system, the heading angle control quantity and the altitude control quantity are converted, and the converted flight control instructions are sent to the UAV's flight controller. After receiving the instructions, the flight controller drives the corresponding actuator to adjust the UAV's heading and altitude, so that the UAV can fly according to the desired heading and altitude, thereby improving the overall performance of the UAV measurement system.

[0023] The adaptive measurement unit 4 controls the multispectral imager to automatically switch bands based on the track coordinates and temperature stratification parameters, and adjusts the lidar scanning angle based on the horizontal pressure gradient field parameters to achieve synchronous inversion of the aerosol vertical profile; In the adaptive measurement unit 4, a spectrum matching-based method is used to control the automatic switching of the multispectral imager bands according to the track coordinates and temperature stratification parameters: Establish a spectral library of typical landforms under different temperature stratifications, determine the current measurement area of ​​the UAV based on the track coordinates, obtain the temperature stratification parameters of the area, and then match the spectral data collected by the multispectral imager with the spectra in the spectral library. After completing the spectral matching calculation, sort all the calculated similarity values, find the spectral record in the spectral library corresponding to the maximum value, extract the multispectral imager band information associated with the record, send a control command to the multispectral imager to switch to the selected band for subsequent measurement operations. At the same time, the currently switched band information is displayed on the UAV control system interface for easy monitoring by the operator. The band with the highest matching degree is selected as the current measurement band. In actual measurement tasks, clearer and more distinctive images of landforms can be obtained, which can significantly improve the quality of the final results, whether used in agricultural monitoring, environmental assessment, or geographic surveying and mapping. In the adaptive measurement unit 4, a fuzzy control algorithm is used to adjust the laser radar scanning angle based on the horizontal pressure gradient field parameters: First, the value range of the horizontal pressure gradient field parameters is determined. According to the value range, the horizontal pressure gradient field parameters are divided into different fuzzy levels. Each level corresponds to a fuzzy subset, which can be represented by fuzzy linguistic variables. The number of levels and interval boundaries can be adjusted according to the accuracy requirements and system complexity of the actual application scenario. According to different fuzzy levels, the scanning angle adjustment amount of the lidar is determined by the fuzzy rule base. The fuzzy rule base is designed according to the influence of the horizontal pressure gradient field on the distribution of the vertical profile of the aerosol. In actual atmospheric detection tasks, the lidar can flexibly adjust the scanning angle according to the real-time horizontal pressure gradient field and effectively obtain the aerosol vertical profile data. Whether in stable meteorological conditions or complex and changeable weather, the quality and integrity of data acquisition can be guaranteed.

[0024] In the present invention, meteorological parameters such as three-dimensional wind speed vectors are collected through the low-altitude meteorological sensor unit 1, and the positioning fusion calculation unit 2 uses the atmospheric delay correction model and fuzzy adaptive Kalman filter to compensate for the atmospheric delay error and adjust the positioning noise matrix to output accurate track coordinates. The collaborative calibration control unit 3 establishes the relationship between the humidity gradient and the positioning height error, obtains the optimal compensation coefficient and converts it into a flight control instruction. The adaptive measurement unit 4 controls the multispectral imager to switch bands and adjust the lidar scanning angle according to the track and meteorological parameters, thereby realizing the synchronous inversion of the aerosol vertical profile, thereby improving the positioning accuracy of the UAV and the reliability of the measurement data under complex meteorological conditions.

[0025] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. The UAV measurement system based on low-altitude meteorological high-precision positioning is characterized by: It includes the following units working together: The low-altitude meteorological sensing unit (1) collects three-dimensional wind speed vectors, temperature stratification, absolute humidity, and horizontal pressure gradient fields through a sensor array; The positioning fusion calculation unit (2) receives the three-dimensional wind speed vector and the temperature stratification parameter, substitutes the temperature and humidity gradient parameter into the refractive index calculation formula through a preset atmospheric delay correction model, calculates the ionosphere / troposphere joint correction amount, adjusts the positioning noise matrix based on the horizontal pressure gradient field using Kalman filtering, and outputs the track coordinates; The collaborative calibration control unit (3) establishes a linear relationship between the humidity gradient and the positioning height error according to the track coordinates and the absolute humidity parameters, constructs a cost function including the heading angle deviation and the humidity gradient observation value, obtains the optimal compensation coefficient by the QR decomposition method, and then converts the optimal compensation coefficient into a flight control instruction; The adaptive measurement unit (4) controls the multispectral imager to automatically switch bands based on the track coordinates and temperature stratification parameters, adjusts the lidar scanning angle based on the horizontal pressure gradient field parameters, and realizes the synchronous inversion of the aerosol vertical profile.

2. The UAV measurement system based on low-altitude meteorological high-precision positioning according to claim 1 is characterized in that: In the positioning fusion calculation unit (2), the atmospheric delay correction model adopts a hierarchical modeling method: The low-altitude meteorological environment is stratified according to altitude, and the temperature and humidity gradient parameters within each layer are considered to be uniformly distributed. For each layer, the temperature and humidity gradient parameters are substituted into the refractive index calculation formula to calculate the ionospheric / tropospheric delay correction for each layer. Finally, the corrections for each layer are accumulated to obtain the total ionospheric / tropospheric joint correction.

3. The UAV measurement system based on low-altitude meteorological high-precision positioning according to claim 2 is characterized in that: In the positioning fusion calculation unit (2), a fuzzy adaptive mechanism is introduced when the Kalman filter is used to adjust the positioning noise matrix: According to the variation range and variation rate of the horizontal pressure gradient field parameters, the process noise covariance matrix and the measurement noise covariance matrix of the Kalman filter are dynamically adjusted through the fuzzy rule base. The fuzzy rule base divides the horizontal pressure gradient field into different fuzzy levels according to its size and variation trend, and sets corresponding adjustment coefficients for each level.

4. The UAV measurement system based on low-altitude meteorological high-precision positioning according to claim 3 is characterized in that: In the positioning fusion calculation unit (2), a track smoothing algorithm based on particle filtering is adopted when outputting track coordinates: A set of particles is generated based on the preliminary track coordinates obtained by Kalman filtering. Each particle represents a track state. The weight of each particle is then calculated based on the horizontal pressure gradient field and the ionosphere / troposphere joint correction. Particles whose weights exceed the preset weight threshold are retained through a resampling process. Finally, the retained particles are weighted averaged to obtain the smoothed track coordinates.

5. The UAV measurement system based on low-altitude meteorological high-precision positioning according to claim 1 is characterized in that: In the collaborative calibration control unit (3), when establishing the linear relationship between the humidity gradient and the positioning height error, the least square method is used for parameter estimation: The track coordinates and absolute humidity parameters under different flight conditions are collected, the humidity gradient and positioning height error are calculated, and then the linear equation between the humidity gradient and positioning height error is fitted by the least squares method to obtain the optimal parameter estimation value.

6. The UAV measurement system based on low-altitude meteorological high-precision positioning according to claim 5 is characterized in that: In the collaborative calibration control unit (3), the cost function constructed adopts the form of weighted least square method: Cost function ;in is the actual heading angle deviation, is the estimated value of heading angle deviation, is the observed value of humidity gradient, is the estimated humidity gradient, is the weight coefficient, which is set according to historical data. is the sample size.

7. The UAV measurement system based on low-altitude meteorological high-precision positioning according to claim 6 is characterized in that: In the collaborative calibration control unit (3), when obtaining the optimal compensation coefficient by the QR decomposition method, the cost function is expressed in the form of a matrix ,in is the coefficient matrix, is the compensation coefficient vector to be determined, is the observation vector, is the weight matrix, For the transpose operation, the coefficient matrix Perform QR decomposition and get ,in is an orthogonal matrix, For an upper triangular matrix, then solve Get the optimal compensation coefficient vector .

8. The UAV measurement system based on low-altitude meteorological high-precision positioning according to claim 7 is characterized in that: In the collaborative calibration control unit (3), a proportional-integral-differential controller is used when converting the optimal compensation coefficient into a flight control instruction: The heading angle and altitude adjustment values ​​are calculated based on the optimal compensation coefficient and used as inputs of the proportional-integral-differential controller. The corresponding control value is calculated through the proportional, integral, and differential links of the proportional-integral-differential controller, and the control value is converted into a flight control command and sent to the UAV.

9. The UAV measurement system based on low-altitude meteorological high-precision positioning according to claim 1 is characterized in that: In the adaptive measurement unit (4), a spectrum matching-based method is used to control the multispectral imager to automatically switch bands according to the track coordinates and temperature stratification parameters: A spectral library of typical landforms under different temperature stratifications is established. The current measurement area of ​​the UAV is determined according to the track coordinates, and the temperature stratification parameters of the area are obtained. The spectral data collected by the multispectral imager is then matched with the spectra in the spectral library, and the band with the highest matching degree is selected as the current measurement band.

10. The UAV measurement system based on low-altitude meteorological high-precision positioning according to claim 9 is characterized in that: In the adaptive measurement unit (4), a fuzzy control algorithm is used to adjust the laser radar scanning angle based on the horizontal pressure gradient field parameters: The horizontal pressure gradient field parameters are divided into different fuzzy levels. The scanning angle adjustment amount of the lidar is determined according to the different fuzzy levels through the fuzzy rule base. The fuzzy rule base is designed according to the influence of the horizontal pressure gradient field on the vertical profile distribution of aerosols.

Images