Dynamic adaptive attitude-position collaborative compensation system and method for wind lidar

Abstract

The application provides an airborne wind measurement laser radar dynamic adaptive attitude-position collaborative compensation system and method which can dynamically adapt to flight states and accurately couple attitude and position interference. The system comprises a GNSS positioning module (1), a MEMS-IMU attitude sensing module (2), a two-stage coordinate conversion module (3), a dynamic speed solving module (4), an adaptive collaborative compensation module (5) and a laser radar wind measurement module (6). The method comprises the following steps: S1, multi-source data synchronous acquisition, S2, ECEF coordinate system conversion, S3, ENU coordinate system conversion, S4, dynamic speed solving, S5, dynamic weight adaptive compensation, S6, data output and feedback storage. The application is applied to the technical field of wind measurement laser radar.

Description

Technical Field

[0001] This invention relates to the field of wind measurement lidar technology, specifically to a dynamic adaptive attitude-position cooperative compensation system and method for wind measurement lidar, which is particularly suitable for airborne mobile wind measurement scenarios with extremely high requirements for wind measurement accuracy, such as offshore wind power site selection, atmospheric boundary layer observation, and meteorological disaster early warning. Background Technology

[0002] With the deep integration of UAV technology and LiDAR technology, airborne wind-measuring LiDAR has become a core technology solution to replace traditional ground-based wind-measuring equipment due to its advantages of flexible deployment, wide coverage, and high spatiotemporal resolution. However, the dynamic movement of UAVs in airborne scenarios (including attitude deflection and position movement) can cause compound interference to wind measurement accuracy, which has become a key bottleneck restricting its high-precision application.

[0003] Existing solutions to this problem have the following core flaws:

[0004] 1. Lack of adaptability in compensation mechanism: Traditional solutions use fixed-weight attitude or position single-dimensional compensation, without considering the dynamic changes in the UAV's flight state. When the UAV is undergoing drastic attitude adjustments (such as tilting or pitching), the proportion of attitude interference increases significantly. When the UAV is performing high-speed maneuvering flight, the speed interference generated by position movement becomes the main source of error. Fixed weights cannot adapt to the interference characteristics of complex scenarios, resulting in unstable compensation accuracy.

[0005] 2. Simplified coordinate transformation model: Some schemes ignore the calculation of the curvature radius of the WGS84 ellipsoid and directly use the spherical approximation model for coordinate transformation. The sign or coefficient of the superimposed ECEF→ENU rotation matrix is ​​incorrect, which leads to the amplification of position data transformation error and thus affects the accuracy of velocity calculation and subsequent compensation.

[0006] 3. Insufficient noise suppression in velocity calculation: The UAV velocity is solved using only a single difference method without smoothing the random noise in the GNSS data. High-frequency noise in the velocity component will be transmitted to the compensation model, introducing additional errors.

[0007] 4. Insufficient utilization of attitude parameters: Only pitch angle, roll angle and yaw angle are used for static correction, without combining angular acceleration to reflect the severity of attitude changes, and it is impossible to predict the dynamic trend of attitude disturbances.

[0008] The aforementioned technical deficiencies result in wind measurement errors in existing airborne wind lidar typically exceeding 0.8 m / s, making it difficult to meet the application requirements of high-precision scenarios such as offshore wind power site selection (requiring wind measurement errors ≤ 0.3 m / s) and precise meteorological observation. Therefore, there is an urgent need for an adaptive compensation technology that can dynamically adapt to flight conditions and accurately couple attitude and position interference to fundamentally solve the problem of wind measurement distortion caused by composite interference. Summary of the Invention

[0009] The purpose of this invention is to overcome the shortcomings of the prior art and provide an airborne wind-measuring lidar dynamic adaptive attitude-position cooperative compensation system and method that can dynamically adapt to flight conditions and accurately couple attitude and position interference.

[0010] The technical solution adopted in this invention is a dynamic adaptive attitude-position cooperative compensation system for an airborne wind-measuring lidar, the system comprising:

[0011] The GNSS positioning module is used to collect the real-time longitude L, latitude B, and elevation H of the UAV;

[0012] The MEMS-IMU attitude perception module is used to collect real-time attitude parameters of the UAV, including pitch angle α, roll angle β, yaw angle γ and angular acceleration (α', β', γ').

[0013] The dual-stage coordinate transformation module is pre-stored with WGS84 ellipsoid standard parameters. It is used to convert the UAV's position data (L,B,H) into coordinates (X,Y,Z) in the geocentric-Earth-fixed coordinate system-ECEF coordinate system, and then convert it into coordinates (Xe,Ye,Zu) in the East-North-Sky ENU coordinate system through a standard rotation matrix.

[0014] The dynamic velocity calculation module uses the first-order central difference method combined with the sliding window smoothing algorithm to perform time-domain differentiation on the ENU coordinates (Xe, Ye, Zu) to solve for the real-time velocity (Ve, Vn, Vu) of the UAV.

[0015] The adaptive collaborative compensation module has a built-in dynamic weight calculation unit and attitude-velocity coupled compensation model. It is used to dynamically adjust the compensation weight based on the UAV's angular acceleration (α', β', γ') and velocity change rate (Ve', Vn', Vu') to remove interference from the original line-of-sight wind speed VLos of the lidar and output accurate wind field speed Vw.

[0016] The lidar wind measurement module is used to emit a laser beam and receive atmospheric scattered echo signals, and calculate the original line-of-sight wind speed VLos by using the Doppler frequency shift principle;

[0017] The data storage and feedback module is used to store raw data, intermediate calculation results and final wind field velocity Vw, and to feed back the velocity change rate (Ve', Vn', Vu') to the adaptive collaborative compensation module.

[0018] The output terminals of the GNSS positioning module and the MEMS-IMU attitude sensing module are connected to the dual-stage coordinate transformation module. The output terminal of the dual-stage coordinate transformation module is connected to the dynamic velocity calculation module. The output terminals of the dynamic velocity calculation module and the MEMS-IMU attitude sensing module are both connected to the adaptive collaborative compensation module. The adaptive collaborative compensation module communicates bidirectionally with the lidar wind measurement module and the data storage and feedback module, respectively.

[0019] Furthermore, the pre-stored WGS84 ellipsoid parameters in the dual-stage coordinate transformation module are: semi-major axis a = 6378137m, flattening f = 1 / 298, semi-minor axis b = a(1-f), then we have

[0020] The square of the first eccentricity is e² = (a² - b²) / a².

[0021] The radius of curvature N of the zonal loop is: .

[0022] Furthermore, the adaptive collaborative compensation module also has a built-in outlier detection unit. The outlier detection unit uses the 3σ criterion combined with a sliding window to remove outliers in the original LiDAR data, and the detection frequency is consistent with the LiDAR data output frequency.

[0023] A method for attitude-position cooperative compensation of an airborne wind-measuring lidar using the dynamic adaptive attitude-position cooperative compensation system described above includes the following steps:

[0024] S1. Multi-source data synchronous acquisition: Location data (L,B,H) is acquired at a frequency of ≥20Hz through the GNSS positioning module, and attitude parameters (α,β,γ,α',β',γ') are acquired at a frequency of ≥100Hz through the MEMS-IMU attitude sensing module. Data synchronization is achieved through timestamp alignment.

[0025] S2, ECEF coordinate system transformation: The two-stage coordinate transformation module is based on the WGS84 ellipsoid parameters and uses the coordinate transformation formula to convert (L,B,H) into ECEF coordinates (X,Y,Z).

[0026] S3, ENU coordinate system transformation: The two-stage coordinate transformation module converts the ECEF coordinates (X,Y,Z) into the UAV's local ENU coordinate system coordinates (Xe,Ye,Zu) using a standard rotation matrix.

[0027] S4. Dynamic velocity calculation: The dynamic velocity calculation module uses the first-order central difference method to solve the time first-order partial derivative of the ENU coordinates, combined with a 3-point sliding window smoothing algorithm, to obtain...

[0028] The speed of the drone (Ve, Vn, Vu) and the rate of change of speed (Ve'=dVe / dt, Vn'=dVn / dt, Vu'=dVu / dt);

[0029] S5. Dynamic weight adaptive compensation: The adaptive collaborative compensation module calculates dynamic compensation weights (ω1, ω2) based on angular acceleration (α', β', γ') and velocity change rate (Ve', Vn', Vu'), and corrects the original line-of-sight wind speed VLos through a coupled compensation model to obtain the accurate wind field speed Vw.

[0030] S6. Data output and feedback storage: Output Vw and store all data, and feed back the rate of change of velocity to step S5 to realize real-time dynamic weight update.

[0031] Furthermore, in step S2, the ECEF coordinate system transformation formula is:

[0032] ,

[0033] Where B is latitude, L is longitude, H is elevation, and N is the radius of curvature of the circumpolar region.

[0034] Furthermore, in step S3, the standard rotation matrix is:

[0035] ,

[0036] Wherein, (X0,Y0,Z0) are the ECEF coordinates of the initial position of the UAV, which are calculated from the initial latitude and longitude (L0,B0) and elevation H0 collected in step S1 using the ECEF coordinate system transformation formula; X e The coordinates are eastward. Y e North coordinates, Z u These are the coordinates for the celestial direction.

[0037] Furthermore, in step S4, the calculation formula for dynamic velocity solution is:

[0038] Ve=[Xe(t+Δt)-Xe(t-Δt)] / (2Δt),

[0039] Vn=[Ye(t+Δt)-Ye(t-Δt)] / (2Δt),

[0040] Vu=[Zu(t+Δt)-Zu(t-Δt)] / (2Δt),

[0041] Ve'=[Ve(t)-Ve(t-Δt)] / Δt,

[0042] Where Δt is the GNSS data sampling interval, and t is the current sampling time; X e (t+Δt), X e (t-Δt) are the eastward coordinates at times t+Δt and t-Δt, respectively, Y e (t+Δt), Y e (t-Δt) represent the northward coordinates at times t+Δt and t-Δt, respectively, Z u (t+Δt), Z u (t-Δt) represents the celestial coordinates at times t+Δt and t-Δt, respectively;

[0043] The sliding window smoothing algorithm uses a 3-point weighted average with weight coefficients of 0.2, 0.6, and 0.2 to eliminate noise in the velocity calculation. The sliding window smoothing algorithm is as follows:

[0044] , i∈(Xe,Ye,Zu).

[0045] Furthermore, in step S5, the calculation model for the dynamic compensation weights (ω1, ω2) is as follows:

[0046] ω1=[k1·(||α',β',γ'||)+ε] / [k1·(||α',β',γ'||)+k2·(||Ve',Vn',Vu'||)+2ε],

[0047] ω2=[k2·(||Ve',Vn',Vu'||)+ε] / [k1·(||α',β',γ'||)+k2·(||Ve',Vn',Vu'||)+2ε],

[0048] The coupling compensation model is as follows:

[0049] Vw=VLos-ω1·[(Ve·cosα·cosγ+Vn·cosα·sinγ-Vu·sinα)·cosθ+(Vn·sinβ-Vu·cosβ·sinγ)·sinθ]-ω2·(Ve·sinθ·cosγ+Vn·sinθ·sinγ),

[0050] Where ω1 is the attitude compensation weight, ω2 is the position compensation weight, and ω1+ω2=1; k1 and k2 are proportional coefficients; ε is the minimum value to avoid the denominator being zero; ||·|| is the 2-norm of the vector; θ is the fixed angle for laser beam installation, 0°≤θ≤90°.

[0051] The beneficial effects of this invention are:

[0052] 1. Dynamic adaptive compensation mechanism improves accuracy and stability: The compensation weight is dynamically adjusted by angular acceleration and velocity change rate, so that the compensation scheme can adapt to the flight state in real time. When the flight is stable, position compensation is emphasized, and when the attitude is drastically adjusted, attitude compensation is emphasized. This avoids the accuracy fluctuation caused by fixed weights. Compared with the traditional scheme, the fluctuation range of wind measurement accuracy is greatly reduced.

[0053] 2. Precise coordinate transformation reduces systematic errors: The curvature radius of the primordial and trochanteric ellipsoids is calculated using the complete WGS84 ellipsoid model. Combined with the standard ECEF→ENU rotation matrix, errors caused by the curvature of the Earth and coordinate transformation errors are eliminated, greatly improving the accuracy of coordinate transformation and providing high-fidelity basic data for velocity calculation.

[0054] 3. Dynamic velocity calculation suppresses noise interference: The combined algorithm of "central difference method + sliding window smoothing" not only ensures the real-time performance of velocity calculation, but also effectively filters out random noise in GNSS data, greatly reducing velocity calculation error and significantly improving noise suppression effect compared to the single difference method.

[0055] 4. Full-parameter coupling model achieves accurate interference removal: Fully utilize multi-dimensional parameters such as attitude angle, angular acceleration, velocity components and velocity change rate to construct a coupling compensation model, which fully covers the composite interference of attitude deflection and position movement. Compared with single-dimensional compensation, the wind measurement error is greatly reduced, meeting the requirements of high-precision scenarios.

[0056] 5. System robustness adaptable to complex environments: The combination of outlier detection unit and real-time feedback mechanism enables the system to work stably in complex environments such as strong winds and multipath interference, significantly improving data efficiency and adapting to diverse airborne scenarios such as offshore wind power and meteorological observation. Attached Figure Description

[0057] Figure 1 This is a simplified structural block diagram of the system of the present invention;

[0058] Figure 2 This is a simplified flowchart of the method of the present invention;

[0059] Figure 3 This is a schematic diagram illustrating the ECEF→ENU coordinate system transformation principle;

[0060] Figure 4 This is the first schematic diagram of the lidar attitude measurement principle;

[0061] Figure 5 This is the second schematic diagram of the lidar attitude measurement principle;

[0062] Figure 6 This is the third schematic diagram of the lidar attitude measurement principle;

[0063] Figure 7 This is the fourth schematic diagram illustrating the principle of lidar attitude measurement;

[0064] Figure 8 It is a graph showing the relationship between dynamic compensation weight and flight status;

[0065] Figure 9 This is a comparison chart of the wind measurement errors of the present invention and the traditional compensation scheme. Detailed Implementation

[0066] like Figures 1-9 As shown, an airborne wind-measuring lidar dynamic adaptive attitude-position cooperative compensation system includes:

[0067] GNSS positioning module 1 is used to collect the real-time longitude L, latitude B and elevation H of the UAV. The elevation H is the height difference between the UAV deck and the sea surface. It supports GPS + Beidou + GLONASS tri-mode positioning.

[0068] MEMS-IMU attitude perception module 2 is used to collect real-time attitude parameters of the UAV, including pitch angle α, roll angle β, yaw angle γ and angular acceleration (α', β', γ').

[0069] The dual-stage coordinate transformation module 3 is pre-stored with WGS84 ellipsoid standard parameters, which is used to convert the UAV's position data (L,B,H) into coordinates (X,Y,Z) in the geocentric-Earth-fixed coordinate system-ECEF coordinate system, and then convert it into coordinates (Xe,Ye,Zu) in the East-North-Sky ENU coordinate system through a standard rotation matrix.

[0070] The dynamic velocity calculation module 4 uses the first-order central difference method combined with the sliding window smoothing algorithm to perform time-domain differentiation on the ENU coordinates (Xe, Ye, Zu) to solve for the real-time velocity (Ve, Vn, Vu) of the UAV.

[0071] The adaptive collaborative compensation module 5 has a built-in dynamic weight calculation unit and attitude-velocity coupled compensation model. It is used to dynamically adjust the compensation weight based on the UAV's angular acceleration (α', β', γ') and velocity change rate (Ve', Vn', Vu') to remove interference from the original line-of-sight wind speed VLos of the lidar and output accurate wind field speed Vw.

[0072] The lidar wind measurement module 6 is used to emit a laser beam and receive atmospheric scattered echo signals, and calculate the original line-of-sight wind speed VLos through the Doppler frequency shift principle.

[0073] The data storage and feedback module 7 is used to store the raw data, intermediate calculation results and the final wind field velocity Vw, and to feed back the velocity change rate (Ve', Vn', Vu') to the adaptive collaborative compensation module 5;

[0074] The output terminals of the GNSS positioning module 1 and the MEMS-IMU attitude sensing module 2 are connected to the dual-stage coordinate transformation module 3. The output terminal of the dual-stage coordinate transformation module 3 is connected to the dynamic velocity calculation module 4. The output terminals of the dynamic velocity calculation module 4 and the MEMS-IMU attitude sensing module 2 are both connected to the adaptive cooperative compensation module 5. The adaptive cooperative compensation module 5 communicates bidirectionally with the lidar wind measurement module 6 and the data storage and feedback module 7, respectively.

[0075] Specifically, the GNSS positioning module 1 has a positioning error ≤0.2m, an elevation measurement accuracy ≤0.05m, a data update rate configured to 20~50Hz, and supports anti-multipath interference algorithms. The MEMS-IMU attitude sensing module 2 has an attitude measurement error ≤0.03°, an angular acceleration measurement range of ±2000° / s², and supports real-time temperature drift calibration (calibration accuracy ≤0.01° / ℃).

[0076] The pre-stored WGS84 ellipsoid parameters in the dual-stage coordinate transformation module 3 are: semi-major axis a = 6378137m, flattening f = 1 / 298, semi-minor axis b = a(1-f), then...

[0077] The square of the first eccentricity is e² = (a² - b²) / a².

[0078] The radius of curvature N of the zonal loop is: .

[0079] The adaptive collaborative compensation module 5 also has a built-in outlier detection unit. The outlier detection unit uses the 3σ criterion combined with a sliding window (window size N=5) to remove outliers in the original lidar data. The detection frequency is consistent with the lidar data output frequency (≥10Hz).

[0080] The attitude-position cooperative compensation method for airborne wind-measuring lidar using the aforementioned dynamic adaptive attitude-position cooperative compensation system includes the following steps:

[0081] S1. Multi-source data synchronous acquisition: Location data (L,B,H) is acquired through GNSS positioning module 1 at a frequency of ≥20Hz, and attitude parameters (α,β,γ,α',β',γ') are acquired through MEMS-IMU attitude sensing module 2 at a frequency of ≥100Hz. Data synchronization is achieved through timestamp alignment.

[0082] S2, ECEF coordinate system transformation: The two-stage coordinate transformation module 3, based on the WGS84 ellipsoid parameters, transforms (L,B,H) into ECEF coordinates (X,Y,Z) using the coordinate transformation formula.

[0083] S3, ENU coordinate system transformation: The two-stage coordinate transformation module 3 transforms the ECEF coordinates (X,Y,Z) into the UAV's local ENU coordinate system coordinates (Xe,Ye,Zu) using a standard rotation matrix.

[0084] S4. Dynamic velocity calculation: The dynamic velocity calculation module 4 uses the first-order central difference method to solve the time first-order partial derivative of the ENU coordinates, combined with a 3-point sliding window smoothing algorithm, to obtain...

[0085] The speed of the drone (Ve, Vn, Vu) and the rate of change of speed (Ve'=dVe / dt, Vn'=dVn / dt, Vu'=dVu / dt);

[0086] S5. Dynamic weight adaptive compensation: The adaptive collaborative compensation module 5 calculates dynamic compensation weights (ω1, ω2) based on angular acceleration (α', β', γ') and velocity change rate (Ve', Vn', Vu'), and corrects the original line-of-sight wind speed VLos through the coupled compensation model to obtain the accurate wind field speed Vw.

[0087] S6. Data output and feedback storage: Output Vw and store all data, and feed back the rate of change of velocity to step S5 to realize real-time dynamic weight update.

[0088] Specifically, in step S2, the ECEF coordinate system transformation formula is:

[0089] ,

[0090] Where B is latitude, L is longitude, H is elevation, and N is the radius of curvature of the circumpolar region.

[0091] In step S3, the standard rotation matrix is:

[0092] ,

[0093] Wherein, (X0,Y0,Z0) are the ECEF coordinates of the initial position of the UAV, which are calculated from the initial latitude and longitude (L0,B0) and elevation H0 collected in step S1 using the ECEF coordinate system transformation formula; X e The coordinates are eastward. Y e North coordinates, Z u These are the coordinates for the celestial direction.

[0094] In step S4, the calculation formula for dynamic velocity is:

[0095] Ve=[Xe(t+Δt)-Xe(t-Δt)] / (2Δt),

[0096] Vn=[Ye(t+Δt)-Ye(t-Δt)] / (2Δt),

[0097] Vu=[Zu(t+Δt)-Zu(t-Δt)] / (2Δt),

[0098] Ve'=[Ve(t)-Ve(t-Δt)] / Δt,

[0099] Where Δt is the GNSS data sampling interval (Δt = 1 / update rate), and t is the current sampling time; X e (t+Δt), X e (t-Δt) are the eastward coordinates at times t+Δt and t-Δt, respectively, Y e (t+Δt), Y e (t-Δt) represent the northward coordinates at times t+Δt and t-Δt, respectively, Z u (t+Δt), Z u (t-Δt) represents the celestial coordinates at times t+Δt and t-Δt, respectively;

[0100] The sliding window smoothing algorithm uses a 3-point weighted average with weight coefficients of 0.2, 0.6, and 0.2 to eliminate noise in the velocity calculation. The sliding window smoothing algorithm is as follows:

[0101] , i∈(Xe,Ye,Zu).

[0102] In step S5, the calculation model for the dynamic compensation weights (ω1, ω2) is as follows:

[0103] ω1=[k1·(||α',β',γ'||)+ε] / [k1·(||α',β',γ'||)+k2·(||Ve',Vn',Vu'||)+2ε],

[0104] ω2=[k2·(||Ve',Vn',Vu'||)+ε] / [k1·(||α',β',γ'||)+k2·(||Ve',Vn',Vu'||)+2ε],

[0105] The coupling compensation model is as follows:

[0106] Vw=VLos-ω1·[(Ve·cosα·cosγ+Vn·cosα·sinγ-Vu·sinα)·cosθ+(Vn·sinβ-Vu·cosβ·sinγ)·sinθ]-ω2·(Ve·sinθ·cosγ+Vn·sinθ·sinγ),

[0107] Where ω1 is the attitude compensation weight, ω2 is the position compensation weight, and ω1+ω2=1; k1 and k2 are proportional coefficients (k1=0.8, k2=0.6); ε is the minimum value (ε=10). -6 ), to avoid the denominator being zero; ||·|| is the 2-norm of the vector; θ is the fixed angle for laser beam installation, 0°≤θ≤90°.

[0108] The present invention will be further described below with reference to more specific embodiments.

[0109] The following describes the implementation process of this invention in detail, using the wind measurement scenario for offshore wind power site selection as an example:

[0110] System Deployment: The compensation system of this invention and the lidar wind measurement module are integrated into a DJI M300 drone. The drone flies at an altitude of H=1000m, and the flight area is the sea area between 120° and 120.5° east longitude and 30° and 30.5° north latitude. The target wind measurement range is the height layer of 50 to 200m above the ground.

[0111] Data acquisition and processing:

[0112] Step S1: The GNSS module collects position data at a frequency of 50Hz. For example, the position data at a certain time t=10.00s is L=120.2345°E, B=30.1234°N, H=1002.35m; the IMU module collects attitude parameters at a frequency of 200Hz, and the parameters interpolated to the time t=10.00s are α=0.8°, β=0.5°, γ=352.7°, α'=0.3° / s², β'=0.2° / s², γ'=0.1° / s².

[0113] Step S2: ECEF coordinate transformation. First, calculate the radius of curvature of the ramusoidal circle N = a / √(1-e²sin²B) = 6378137 / √(1-0.00669437999013²×sin²(30.1234°)) ≈ 6379283.65m; substitute into the formula to calculate X = (6379283.65+1002.35)×cos(30.1234°)×cos(120.23) 45°)≈-3195682.45m, Y=(6379283.65+1002.35)×cos(30.1234°)×sin(120.2345°)≈552834 1.78m, Z=(6379283.65×(1-0.00669437999013²)+1002.35)×sin(30.1234°)≈3289765.32m;

[0114] Step S3: ENU coordinate transformation. Initial ECEF coordinates: X0 = -3192456.18m, Y0 = 5523124.36m, Z0 = 3287654.19m; calculate coordinate differences: X - X0 = -3226.27m, Y - Y0 = 5217.42m, Z - Z0 = 2111.13m; substitute into the rotation matrix to calculate Xe = -sin(120.2345°) × (-3226.27) + cos(120.2345°) × 5217.42 + 0 × 2111.13 ≈ 4568.32m, Ye = -sin(30.12) × ( ... 34°)×cos(120.2345°)×(-3226.27)-sin(30.1234°)×sin(120.2345°)×5217.42+cos(30.1234°)×2111.13≈1892.45m, Zu=cos (30.1234°)×cos(120.2345°)×(-3226.27)+cos(30.1234°)×sin(120.2345°)×5217.42+sin(30.1234°)×2111.13≈3215.68m;

[0115] Step S4: Dynamic velocity calculation. The ENU coordinates at t-Δt=9.98s are Xe=4559.21m, Ye=1887.32m, Zu=3213.45m; the ENU coordinates at t+Δt=10.02s are Xe=4577.43m, Ye=1897.58m, Zu=3217.91m; calculate the original velocity Ve=(4577.43-4559.21) / (2×0.02)=455.5m / s, Vn=(1897.58-1887.32) / (2×0.02)=455.5m / s. 2) = 256.5 m / s, Vu = (3217.91 - 3213.45) / (2 × 0.02) = 111.5 m / s; after smoothing by the sliding window, Ve = 452.3 m / s, Vn = 254.8 m / s, Vu = 110.2 m / s; the rate of change of velocity Ve' = (452.3 - 448.7) / 0.02 = 180 m / s², Vn' = (254.8 - 251.2) / 0.02 = 180 m / s², Vu' = (110.2 - 108.5) / 0.02 = 85 m / s²;

[0116] Step S5: Dynamic weight adaptive compensation, calculate ||α',β',γ'||=√(0.3²+0.2²+0.1²)=0.3742, ||Ve',Vn',Vu'||=√(180²+180²+85²)=267.8; dynamic weight ω1=(0.8×0.3742+10 -6 ) / (0.8×0.3742+0.6×267.8+2×10 -6 )≈0.0018, ω2≈0.9982; the original line-of-sight wind speed of the lidar is VLos=523.6m / s; substituting into the coupling compensation model, the interference component is calculated as: ω1×[(452.3×cos0.8°×cos352.7°+254.8×cos0.8°×sin352.7°-110.2×sin0.8°)×cos45°+(254.8×sin0.5°-1 10.2×cos0.5°×sin352.7°)×sin45°]+ω2×(452.3×sin45°×cos352.7°+254.8×sin45°×sin352.7°)≈0.0018×420.3+0.9982×356.8≈357.4m / s; therefore, the precise wind field velocity Vw=523.6-357.4=166.2m / s;

[0117] Step S6: Data output and storage. Output Vw=166.2m / s to the ground receiver via RS485 interface, and synchronously store the full amount of data in flash memory with a timestamp of 10.00s.

[0118] Implementation effect verification:

[0119] Under the same flight scenario, the wind measurement error of the traditional fixed weight compensation scheme is 0.6 to 1.1 m / s, while the wind measurement error of the scheme of this invention is 0.15 to 0.23 m / s, which meets the accuracy requirement of ≤0.3 m / s for offshore wind power site selection. Moreover, during the UAV attitude adjustment phase (α'=5.6° / s²), the wind measurement error of this invention remains stable within 0.25 m / s, which is more than 70% lower than the error of the traditional scheme.

[0120] Finally, it should be emphasized that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A dynamic adaptive attitude-position cooperative compensation system for wind-measuring lidar, characterized in that, The system includes: GNSS positioning module (1) is used to collect the real-time longitude L, real-time latitude B and real-time elevation H of the UAV; MEMS-IMU attitude perception module (2) is used to collect real-time attitude parameters of the UAV, including pitch angle α, roll angle β, yaw angle γ and angular acceleration (α', β', γ'). The dual-stage coordinate transformation module (3) is pre-stored with WGS84 ellipsoid standard parameters, which is used to convert the UAV's position data (L,B,H) into coordinates (X,Y,Z) in the geocentric-Earth-fixed coordinate system-ECEF coordinate system, and then convert them into coordinates in the East-North-Sky ENU coordinate system through a standard rotation matrix. X e , Y e , Z u ), X e The coordinates are eastward. Y e North coordinates, Z u The coordinates are celestial coordinates; The dynamic velocity calculation module (4) uses the first-order central difference method combined with the sliding window smoothing algorithm to calculate the ENU coordinates ( X e , Y e , Z u Perform time-domain differentiation to solve for the real-time velocity of the UAV. V e , V n , V u The calculation formula for dynamic velocity is: V e =[ X e (t+Δt)- X e (t-Δt)] / (2Δt), V n =[ Y e (t+Δt)- Y e (t-Δt)] / (2Δt), V u =[ Z u (t+Δt)- Z u (t-Δt)] / (2Δt), V e ’=[ V e (t)- V e (t-Δt)] / Δt, in, V e ' represents the velocity change rate component of the UAV, Δt represents the GNSS data sampling interval, and t represents the current sampling time; X e (t+Δt), X e (t-Δt) represent the eastward coordinates at times t+Δt and t-Δt, respectively. Y e (t+Δt), Y e (t-Δt) represent the northward coordinates at times t+Δt and t-Δt, respectively. Z u (t+Δt), Z u (t-Δt) represents the celestial coordinates at times t+Δt and t-Δt, respectively; The sliding window smoothing algorithm uses a 3-point weighted average with weight coefficients of 0.2, 0.6, and 0.2 to eliminate noise in the velocity calculation. The sliding window smoothing algorithm is as follows: ，i∈( e , n , u )； The adaptive collaborative compensation module (5) has a built-in dynamic weight calculation unit and attitude-velocity coupled compensation model, which is used to calculate the angular acceleration (α', β', γ') and velocity change rate of the UAV. V e '、 V n '、 V u The compensation weights are dynamically adjusted to remove interference from the original line-of-sight wind speed (VLos) of the lidar, resulting in a precise output wind field velocity. V w The calculation model for the dynamic compensation weights (ω1, ω2) is as follows: ω1=[k1·(||α',β',γ'||)+ε] / [k1·(||α',β',γ'||)+k2·(|| V e ', V n ', V u '||)+2ε], ω2=[k2·(|| V e ', V n ', V u '||)+ε] / [k1·(||α',β',γ'||)+k2·(|| V e ', V n ', V u '||)+2ε], The coupling compensation model is as follows: V w =VLos-ω1·[( V e ·cosα·cosγ+ V n ·cosα·sinγ- V u ·sinα)·cosθ+( V n ·sinβ- V u ·cosβ·sinγ)·sinθ]-ω2·( V e ·sinθ·cosγ+ V n ·sinθ·sinγ), Where ω1 is the attitude compensation weight, ω2 is the position compensation weight, and ω1+ω2=1; k1 and k2 are proportional coefficients; ε is the minimum value to avoid the denominator being zero; ||·|| is the 2-norm of the vector; θ is the fixed angle for laser beam installation, 0°≤θ≤90°; The lidar wind measurement module (6) is used to emit a laser beam and receive atmospheric scattered echo signals, and calculate the original line-of-sight wind speed VLos by using the Doppler frequency shift principle; The data storage and feedback module (7) is used to store raw data, intermediate calculation results and final wind field velocity. V w and the rate of change of velocity ( V e '、 V n '、 V u Feedback is sent to the adaptive collaborative compensation module (5); The output ends of the GNSS positioning module (1) and the MEMS-IMU attitude perception module (2) are connected to the dual-stage coordinate transformation module (3). The output end of the dual-stage coordinate transformation module (3) is connected to the dynamic velocity calculation module (4). The output ends of the dynamic velocity calculation module (4) and the MEMS-IMU attitude perception module (2) are both connected to the adaptive collaborative compensation module (5). The adaptive collaborative compensation module (5) communicates bidirectionally with the lidar wind measurement module (6) and the data storage and feedback module (7).

2. The system according to claim 1, characterized in that, The pre-stored WGS84 ellipsoid parameters in the dual-stage coordinate transformation module (3) are: semi-major axis a = 6378137m, flattening f = 1 / 298, semi-minor axis b = a(1-f), then... The square of the first eccentricity is e² = (a² - b²) / a². The radius of curvature N of the zonal loop is: .

3. The system according to claim 1, characterized in that, The adaptive collaborative compensation module (5) also has an outlier detection unit built in. The outlier detection unit uses the 3σ criterion combined with a sliding window to remove outliers in the original laser radar data. The detection frequency is consistent with the laser radar data output frequency.

4. A method for attitude-position cooperative compensation of an airborne wind-measuring lidar using the dynamic adaptive attitude-position cooperative compensation system for wind-measuring lidar as described in any one of claims 1 to 3, characterized in that, The method includes the following steps: S1. Multi-source data synchronous acquisition: Location data (L,B,H) is acquired at a frequency of ≥20Hz through the GNSS positioning module (1), and attitude parameters (α,β,γ,α',β',γ') are acquired at a frequency of ≥100Hz through the MEMS-IMU attitude sensing module (2). Data synchronization is achieved through timestamp alignment. S2, ECEF coordinate system transformation, the two-stage coordinate transformation module (3) based on WGS84 ellipsoid parameters, transforms (L,B,H) into ECEF coordinates (X,Y,Z) through coordinate transformation formula; S3, ENU coordinate system transformation, the two-stage coordinate transformation module (3) converts the ECEF coordinates (X,Y,Z) into the UAV's local ENU coordinate system coordinates using a standard rotation matrix. X e , Y e , Z u ); S4. Dynamic velocity calculation: The dynamic velocity calculation module (4) uses the first-order central difference method to solve the time first-order partial derivative of the ENU coordinates, combined with the 3-point sliding window smoothing algorithm, to obtain... Drone speed ( V e , V n , V u ) and rate of change of velocity ( V e '=d V e / dt, V n '=d V n / dt, V u '=d V u / dt); S5, Dynamic weight adaptive compensation, the adaptive collaborative compensation module (5) is based on angular acceleration (α', β', γ') and velocity change rate ( V e '、 V n '、 V u ') Calculate the dynamic compensation weights (ω1, ω2), and correct the original line-of-sight wind speed VLos through a coupled compensation model to obtain the accurate wind field velocity. V w ; S6. Data output and feedback storage, output V w It stores all data and feeds back the rate of change of speed to step S5 to achieve real-time dynamic weight updates.

5. The method according to claim 4, characterized in that, In step S2, the ECEF coordinate system transformation formula is: ， Where B is the real-time latitude, L is the real-time longitude, H is the real-time elevation, and N is the real-time radius of curvature of the circumpolar region.

6. The method according to claim 5, characterized in that, In step S3, the standard rotation matrix is: ， Wherein, (X0,Y0,Z0) are the ECEF coordinates of the initial position of the UAV, which are calculated from the initial latitude and longitude (L0,B0) and elevation H0 collected in step S1 using the ECEF coordinate system transformation formula; X e The coordinates are eastward. Y e North coordinates, Z u These are the coordinates for the celestial direction.

Images