Hydrological observation sensor data real-time correction method based on motion attitude of unmanned ship

Abstract

The invention discloses a hydrological observation sensor data real-time correction method based on the motion attitude of an unmanned ship, and the method comprises the following steps: S1, collecting the real-time attitude data and static installation position vector data of the unmanned ship in real time, and determining the equivalent sound velocity profile of an ADCP wave beam pointing region; s2, outputting the actual longitude, latitude and depth of the sampling point of which the sound ray bending and attitude coupling error is eliminated; s3, carrying out geometric correction on the ADCP beam based on the instantaneous incidence angle, and generating normalized echo intensity; s4, calculating a dynamic signal-to-noise ratio, and comparing the dynamic signal-to-noise ratio with a preset threshold to generate a quality flag bit Q; and S5, carrying out weighted least square solution, and outputting a final corrected velocity vector to be solved. According to the method, the systematic deviation of sound ray tracking is effectively eliminated by generating the equivalent sound velocity profile matched with the actual pointing region of the ADCP beam.

Description

Technical Field

[0001] This invention relates to the field of sensor data correction technology, and more specifically to a method for real-time correction of hydrological observation sensor data based on the motion attitude of unmanned surface vessels. Background Technology

[0002] Unmanned surface vessels (USVs) are widely used for hydrological observation tasks in inland rivers, estuaries, and nearshore waters due to their maneuverability, flexibility, and low cost. A core operational task is measuring water flow velocity profiles using an onboard Acoustic Doppler Current Profiler (ADCP). The ADCP transducer emits a fixed-frequency acoustic pulse into the water. When the sound waves encounter suspended particles moving with the current, they are scattered. A portion of the scattered signal returns along its original path and is received by the transducer. Because of the relative motion between the particles and the transducer, a frequency shift occurs between the received and transmitted signals—the Doppler shift. By measuring this shift and combining it with the speed of sound propagation in water, the velocity of the particles along the beam direction can be calculated, and thus the water flow velocity can be derived.

[0003] When unmanned surface vessels (USVs) are equipped with acoustic Doppler current profilers (ADCPs) for hydrological observation, sensor data calibration is a crucial step in ensuring data accuracy. Existing sensor data calibration methods mainly focus on the impact of hull motion (roll, pitch, and bow) on the sensor coordinate system, using a rotation matrix to transform the current velocity data measured in the instrument coordinate system to the geodetic coordinate system.

[0004] However, in actual operations, especially in complex hydrological environments such as inland river estuaries and nearshore areas, unmanned surface vessels (USVs) are affected by wind, waves, and currents, resulting in high-frequency rolling, pitching, and bowing motions, which leads to the following problems:

[0005] 1. It is commonly assumed that the sound velocity profile (measured by CTD or SVP) is uniform in the horizontal direction, and this profile is directly applied to ADCP beam path calculation. In reality, when the attitude of the unmanned surface vessel changes, the CTD sensor and the ADCP transducer beam emission point experience relative displacement in the horizontal direction. In waters with significant horizontal sound velocity gradients (such as estuarine salt wedge fronts), this relative displacement causes the sound velocity data used by the ADCP to differ from the actual sound velocity along the beam's path, resulting in refraction calculation errors due to spatial mismatch in sound velocity sampling.

[0006] 2. Changes in the attitude of unmanned surface vessels (USVs) alter the angle between the transducer beam and the water surface, as well as the effective scattering cross section, leading to dynamic fluctuations in echo signal strength. Existing technologies directly calculate the signal-to-noise ratio (SNR) using the original intensity and then discard the data, which may inadvertently discard normally attenuated signals, resulting in distorted assessments of the echo SNR affected by attitude. Summary of the Invention

[0007] To address this, the present invention provides a real-time correction method for hydrological observation sensor data based on the motion attitude of unmanned surface vessels (USVs) to solve the problems of sound velocity profile usage errors caused by the spatial asynchrony between the temperature, salinity, and depth (CTD) sensor and the ADCP transducer in the prior art, as well as the distortion of echo signal-to-noise ratio assessment due to the influence of USV attitude.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A real-time correction method for hydrological observation sensor data based on the motion attitude of unmanned surface vessels includes the following steps:

[0010] S1: Real-time acquisition of unmanned surface vessel attitude data and static installation position vector data to construct the rotation matrix from the hull to the geographic coordinate system; then calculate the difference in horizontal position between the CTD sensor entry point and the ADCP transducer center in the geographic coordinate system, and determine the equivalent sound velocity profile of the ADCP beam pointing area.

[0011] S2: The water body is stratified and the instantaneous incident angle is calculated in combination with the real-time attitude angle. The actual slope distance and horizontal offset of the sound ray accumulated in each layer are calculated, and the actual longitude of the sampling point is output after eliminating the error of sound ray bending and attitude coupling. ,latitude and depth ;

[0012] S3: Based on instantaneous incident angle The ADCP beam is geometrically corrected to eliminate the change in effective scattering volume caused by the ship's tilt, thereby generating a normalized echo intensity. ;

[0013] S4: Normalized echo intensity based on the output of S3 and the actual slant distance output by S2 Calculate the dynamic signal-to-noise ratio The dynamic signal-to-noise ratio is compared with a preset threshold to generate a quality flag bit Q;

[0014] S5: Based on the actual longitude output by S2 ,latitude ,depth We perform weighted least squares calculations on the mass flag and the equivalent sound velocity profile to output the final corrected velocity vector.

[0015] Furthermore, the specific steps of S1 are as follows:

[0016] S1.1: The inertial measurement unit carried by the unmanned surface vessel collects the roll angle, pitch angle and bow angle of the hull in real time at a fixed sampling frequency, and calculates the draft at the same time;

[0017] S1.2: Pre-program the static installation position vectors of the CTD sensor and ADCP transducer in the ship's coordinate system into the memory. ;

[0018] S1.3: Based on the three attitude angles acquired in real time, construct a three-dimensional rotation matrix from the ship coordinate system to the geographic coordinate system. , used to describe the instantaneous attitude of a ship's hull in space at any given moment;

[0019] S1.4: Combine the static mounting vector with the three-dimensional rotation matrix Multiply to generate a relative position vector;

[0020] S1.5: Project the relative position vectors, retaining only the values ​​in the geographic coordinate system. The difference between the y-axis and the x-axis is used to obtain the horizontal displacement vector between the actual water entry point of the CTD sensor and the center point of the ADCP transducer at time t. ;

[0021] S1.6: The horizontal sound velocity gradient estimate is calculated based on the real-time differential values ​​of the main CTD sensor and the sound velocity sensor mounted on this unmanned surface vessel. Combined with the horizontal displacement vector and depth weighting function The vertical sound velocity profile measured by the CTD sensor is corrected to the equivalent sound velocity profile of the ADCP beam pointing region. .

[0022] Furthermore, the specific content of S1.6 is as follows: First, the sound velocity measurement value of the main CTD sensor at the current water entry point is synchronously acquired. Then, based on the auxiliary sound velocity probe installed on the other side of the hull, the horizontal distance is obtained as follows: Sound velocity measurement at another measuring point Then, the estimated horizontal sound speed gradient is calculated. The calculation formula is as follows:

[0023]

[0024] in, It is the unit vector along the line connecting the two measuring points;

[0025] Combined with the horizontal displacement vector and depth weighting function The vertical sound velocity profile measured by the CTD sensor is corrected to the equivalent sound velocity profile of the ADCP beam pointing region. Equivalent sound speed profile The calculation formula is as follows:

[0026]

[0027] in, For CTD sensors in The raw sound speed data obtained from the measurement at any time. For the current moment The horizontal displacement vector.

[0028] Furthermore, the specific content of S2 is as follows:

[0029] 1) Divide the water body into N uniform thin layers along the depth direction;

[0030] 2) Then read the fixed emission angle and azimuth angle parameters of the ADCP beam, and combine them with the real-time attitude angle data to calculate the instantaneous incident angle of the beam in the geographic coordinate system through coordinate transformation. The calculation formula is as follows:

[0031]

[0032] in, for The roll angle at any moment, for The pitch angle at any moment, The fixed emission angle of the beam in the ship's coordinate system. The azimuth angle of the beam in the ship's coordinate system;

[0033] 3) According to the first The sound velocity and ray parameters of the layer are used to infer the glancing angle of the sound ray. The sound wave at the first grazing angle and water layer thickness is calculated. The horizontal and vertical displacement components of the layer are summed to obtain the sound wave propagation to the first layer. Actual slope distance of the layer and cumulative horizontal offset ;

[0034] 4) Based on cumulative horizontal offset and geographical azimuth Output actual longitude ,latitude and depth The calculation formula is as follows:

[0035]

[0036]

[0037]

[0038] in, is the Earth's average radius, used to convert planar displacement into spherical angles. The current distance to the gate corresponds to the layer index, which satisfies , For unmanned surface vessels at all times Real-time draft depth, For unmanned surface vessels at the current moment Real-time location.

[0039] Furthermore, the normalized echo intensity The calculation formula is as follows;

[0040] in, For ADCP sensor in The first time directly measured output, without attitude geometry correction. The echo signal strength of each distance gate, To fix the launch angle, The instantaneous angle of incidence.

[0041] Furthermore, when comparing the dynamic signal-to-noise ratio with a preset threshold, if the dynamic signal-to-noise ratio... If the value is higher than the threshold, the quality flag Q=1, indicating that the layered data at the current depth is valid and marked as valid data; if the dynamic signal-to-noise ratio is... If the data is below the threshold, the quality flag Q=0, marking it as invalid data.

[0042] Furthermore, the specific steps of S5 are as follows:

[0043] S5.1: Receive the actual longitude of S2 ,latitude and depth The equivalent sound velocity profile of S1 is used to calculate the beams. In the Radial flow velocity at a depth of stratification;

[0044] S5.2: Based on the weighted least squares algorithm, the radial velocity observations of N beams are mapped to the three geographic components of east, north and vertical to obtain the velocity vector to be determined.

[0045] S5.3: Bind the velocity vector to be determined with the absolute geographic coordinates and timestamps provided in S2 to generate the final hydrological sensor dataset.

[0046] Furthermore, the radial flow velocity The calculation formula is as follows:

[0047]

[0048] in, This is the equivalent sound speed profile. The frequency of the ultrasonic waves emitted by the ADCP transducer. To fix the launch angle, For the first The beam at the 1st The frequency of the received echo signal at each depth of the layer is the same as the transmission frequency. difference.

[0049] Furthermore, the dynamic signal-to-noise ratio The calculation formula is as follows:

[0050] in, The initial intensity constant of the acoustic wave emitted by the ADCP transducer is determined by the factory calibration of the equipment. exist Background ambient noise intensity at any given time To accumulate the actual slope distance, This is the preset absorption coefficient.

[0051] This invention has the following advantages: By acquiring unmanned surface vessel (USV) attitude data in real time and combining it with static installation position vectors, a dynamic rotation model from the hull to the geographic coordinate system is constructed. This model accurately quantifies the horizontal displacement vector between the CTD sensor entry point and the ADCP transducer center caused by hull swaying. Based on this, an equivalent sound velocity profile matching the actual pointing area of ​​the ADCP beam is generated. This solves the problem in traditional processing where the dynamic changes in the sensor's spatial position are ignored, leading to incorrect refraction calculations using sound velocity data in waters with significant horizontal sound velocity gradients. It effectively eliminates systematic biases in ray tracking and avoids flow velocity measurement errors caused by spatial mismatch in sound velocity sampling.

[0052] Simultaneously, a geometric normalization correction based on the instantaneous incident angle is introduced. By normalizing the original echo intensity, the effective scattering volume projection scaling effect is eliminated, effectively removing the interference of attitude changes on signal intensity, so that the echo intensity only reflects the true changes in the water scattering characteristics. Based on this normalized intensity and the actual slant range output by ray tracing, the dynamic signal-to-noise ratio is calculated, which can truly reflect the effective intensity of the signal relative to the background noise, avoiding errors in transmission loss estimation caused by sound ray bending or false rejection caused by attitude fluctuations, and improving the real-time correction effect of sensor data.

[0053] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. Attached Figure Description

[0054] To more intuitively illustrate the prior art and this application, exemplary drawings are provided below. It should be understood that the specific shapes and structures shown in the drawings should not generally be regarded as limiting conditions for implementing this application; for example, based on the technical concept disclosed in this application and the exemplary drawings, those skilled in the art are able to easily make conventional adjustments or further optimizations to the addition / reduction / classification, specific shapes, positional relationships, connection methods, size ratios, etc. of certain units (components).

[0055] Figure 1 This is a flowchart illustrating the implementation of the real-time correction method for hydrological observation sensor data based on the motion attitude of an unmanned surface vessel (USV) according to the present invention.

[0056] Figure 2 This is a flowchart illustrating the implementation of step S1 in the real-time correction method for hydrological observation sensor data based on the motion attitude of an unmanned surface vessel (USV) according to the present invention. Detailed Implementation

[0057] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these embodiments are merely for further explanation of the present invention and should not be construed as limiting the scope of protection of the present invention. Technical engineers in the field can make some non-essential improvements and adjustments to the present invention based on the above-described content. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] Please see Figures 1-2 A real-time correction method based on hydrological observation sensor data of unmanned surface vessel motion attitude includes the following steps:

[0059] S1: First, real-time attitude data and static installation position vector data of the unmanned surface vessel (USV) are acquired to construct a rotation matrix from the hull to the geographic coordinate system. Then, the horizontal position difference between the CTD sensor entry point and the ADCP transducer center in the geographic coordinate system is calculated, and the equivalent sound velocity profile of the ADCP beam pointing area is determined. This facilitates the reduction of spatial mismatch between the CTD measurement point and the ADCP beam path caused by hull swaying, avoids refraction calculations using incorrect sound velocity data in waters with significant horizontal sound velocity gradients, and reduces the systematic bias in ray tracking from the source.

[0060] The static installation location vector data includes the installation locations of the temperature, salinity, and depth (CTD) sensor and the acoustic Doppler current profiler (ADCP) transducer.

[0061] Real-time attitude data includes the unmanned surface vessel's roll, pitch, bow angles, and draft. Data is collected through an inertial measurement unit (IMU) and pressure sensors mounted on the rigid structure of the hull;

[0062] An inertial measurement unit (IMU) mainly consists of a three-axis accelerometer and a three-axis gyroscope. Its working principle is based on Newton's laws of motion: the accelerometer measures the linear acceleration of the carrier along three orthogonal axes by detecting the displacement or deformation of an internal mass under inertial forces; the gyroscope, based on the Coriolis effect, senses the angular velocity of the carrier around the three orthogonal axes by measuring the capacitance changes generated by the rotating vibrating structure. By measuring the raw angular velocity and acceleration data, the inertial measurement unit can output real-time attitude angles, including roll, pitch, and yaw angles.

[0063] The pressure sensor is installed on the bottom of the unmanned surface vessel to measure water pressure. Through formula Calculate draft ,in The density of water, The acceleration due to gravity is used to determine the water surface reference.

[0064] The specific steps of S1 are as follows:

[0065] S1.1: The inertial measurement unit carried by the unmanned surface vessel collects the roll angle (the angle of rotation around the longitudinal axis of the hull), pitch angle (the angle of rotation around the transverse axis of the hull), and bow angle (the angle of rotation around the vertical axis of the hull) of the hull in real time at a fixed sampling frequency, and calculates the draft at the same time.

[0066] S1.2: Pre-program the static installation position vectors of the CTD sensor and ADCP transducer in the ship's coordinate system into the memory. .in, and These represent the longitudinal and lateral mounting distances of the CTD sensor relative to the center of the ADCP transducer, respectively. This indicates the vertical installation height difference.

[0067] The ship's coordinate system is defined as follows: the origin is located at the center of the ADCP transducer, the X-axis points to the bow, the Y-axis points to the starboard side, and the Z-axis is perpendicular to the direction of travel.

[0068] S1.3: Based on the three attitude angles (roll angle R, pitch angle P, and yaw angle H) collected in real time, construct a three-dimensional rotation matrix from the hull coordinate system to the geographic coordinate system according to the Euler angle rotation sequence of first roll, then pitch, and finally yaw. This is used to describe the instantaneous attitude of the ship's hull in space at any given moment, and the formula is as follows:

[0069]

[0070] in, For the bow roll matrix, For the pitch matrix, This is the roll matrix. The calculation formula is as follows:

[0071]

[0072] in, The roll angle indicates the hull's rotation around the waterline. The left and right tilt angles of the axis, The ship is listing to starboard (when viewed from stern to bow, rotate clockwise). The ship is listing to port (rotating counterclockwise).

[0073]

[0074] in, The pitch angle represents the angle at which the ship rolls around the waterline. The pitch angle of the shaft. The bow is raised (rotated counterclockwise, viewed from the starboard side to the port side). The bow is pointing downwards (rotating clockwise).

[0075]

[0076] in, The bow angle represents the change in heading of the ship about the z-axis. Turn the bow to the right (clockwise, looking down at the ship from the zenith). Turn the bow to the left (counter-clockwise).

[0077] S1.4: Combine the static mounting vector with the three-dimensional rotation matrix Multiplication generates a relative position vector. This vector converts the fixed mounting vector of the CTD sensor relative to the ADCP transducer into a relative position vector of the two sensors in the geographic coordinate system at the current attitude. It quantifies the spatial positional changes of the sensor caused by hull rolling.

[0078] S1.5: Project the relative position vector, ignoring the vertical (z-axis) component and retaining only the component in the geographic coordinate system. The difference between the y-axis and the x-axis is used to obtain the horizontal displacement vector between the actual water entry point of the CTD sensor and the center point of the ADCP transducer at time t. When the unmanned surface vessel tilts, the actual water entry point of the CTD sensor, which was originally aligned vertically or had a fixed horizontal spacing, and the center point of the ADCP transducer will shift relative to each other on the water surface.

[0079] S1.6: The horizontal sound velocity gradient estimate is calculated based on the real-time differential values ​​of the main CTD sensor and the sound velocity sensor mounted on this unmanned surface vessel. ;

[0080] First, the sound velocity measurement value of the main CTD sensor at the current water entry point is acquired simultaneously. Then, based on the auxiliary sound velocity probe installed on the other side of the hull, the horizontal distance is obtained as follows: Sound velocity measurement at another measuring point Then, the estimated horizontal sound speed gradient is calculated. The calculation formula is as follows:

[0081]

[0082] in, It is the unit vector of the direction of the line connecting the two measuring points.

[0083] Combined with the horizontal displacement vector and depth weighting function The vertical sound velocity profile measured by the CTD sensor is corrected to the equivalent sound velocity profile of the ADCP beam pointing region. The function is defined as water depth. The monotonically decreasing function, with its value limited to the interval [0, 1], is used to quantify the attenuation effect of the horizontal sound velocity gradient from the water surface to deeper layers (i.e., the horizontal gradient has a significant impact on the surface layer, while its impact weakens in deeper layers). Equivalent sound velocity profile. The calculation formula is as follows:

[0084]

[0085] in, For CTD sensors in Time-dependent measurements The changing raw sound speed data, For the current moment The horizontal displacement vector.

[0086] S2: The water body is stratified and the instantaneous incident angle is calculated in combination with the real-time attitude angle. The actual slope distance and horizontal offset of the sound ray accumulated in each layer are calculated, and the actual longitude of the sampling point is output after eliminating the error of sound ray bending and attitude coupling. ,latitude and depth This method facilitates the calculation of the actual propagation path of sound waves in density-stratified water bodies through layered ray tracing, thus solving the spatial positioning error of flow velocity caused by the traditional straight-line assumption.

[0087] First, based on ray acoustics theory, the water body is divided into N uniform thin layers along the depth direction. The thickness of each layer is determined according to the maximum measurement depth and the number of layers, with no fewer than 20 layers to ensure calculation accuracy. The equivalent sound velocity profile output in step S1 is then... The depth is interpolated and assigned to each water layer, assuming that the sound velocity within each layer is constant.

[0088] Then, the fixed emission angle and azimuth angle parameters of the ADCP beam are read, and combined with real-time attitude angle data (R, P, H), the instantaneous incident angle of the beam in the geographic coordinate system is calculated through coordinate transformation. Using the surface sound velocity and incident angle as initial conditions, the ray parameters are determined, which remain constant throughout the sound ray propagation process. The calculation formula is as follows:

[0089]

[0090] in, for The roll angle at any moment, for The pitch angle at any moment, The fixed emission angle of the beam in the ship's coordinate system. The azimuth angle of the beam in the ship's coordinate system.

[0091] According to the The sound velocity and ray parameters of the layer are used to infer the glancing angle of the sound ray. The horizontal and vertical displacement components of the sound wave in the water layer are calculated based on the grazing angle and the water layer thickness, and then summed to obtain the sound wave propagation to the [missing information - likely a specific point or dimension]. Actual slope distance of the layer and cumulative horizontal offset The calculation formula is as follows:

[0092]

[0093] in, The ray parameter is a constant that remains unchanged during sound wave propagation and is used to describe the propagation characteristics of sound rays. For the first The speed of sound in a layer of water.

[0094] Actual slope distance and cumulative horizontal offset The calculation formula is as follows:

[0095]

[0096] in, For the first The thickness of the water layer, For the sound wave at the 1st Horizontal displacement of the layer, . The current distance to the gate corresponds to the layer index (integer), which satisfies .

[0097] The horizontal offset vectors of each distance gate relative to the transducer are obtained based on ray tracing accumulation. and its corresponding geographical azimuth , , For the bow roll angle, the relative rectangular coordinate displacement is converted into latitude and longitude increments based on the principle of spherical projection, and then superimposed on the unmanned surface vessel's position at the current moment. Real-time location This process maps the ship's relative coordinate system to its absolute geographic coordinate system. Simultaneously, it uses the accumulated vertical layer thickness to determine the true depth, thereby outputting the actual longitude to eliminate errors caused by acoustic curvature and attitude coupling. ,latitude and depth The calculation formula is as follows:

[0098]

[0099]

[0100]

[0101] in, is the Earth's average radius, used to convert planar displacement into spherical angles. The current distance to the gate corresponds to the layer index (integer), which satisfies = , For unmanned surface vessels at all times Real-time draft depth.

[0102] S3: Based on instantaneous incident angle The ADCP beam is geometrically corrected to eliminate the change in effective scattering volume caused by the ship's tilt, thereby generating a normalized echo intensity. The calculation formula is as follows;

[0103] in, For ADCP sensor in The first time directly measured output, without attitude geometry correction. The echo signal strength of each distance gate, To fix the launch angle, The instantaneous angle of incidence.

[0104] When the ship tilts, the angle between the ADCP beam axis and the direction perpendicular to gravity (instantaneous angle of incidence) changes, causing a geometric projection scaling of the volume of water irradiated per unit depth (effective scattering cross-section) relative to the stationary horizontal state. Directly using the original echo intensity would incorrectly interpret this geometric gain or loss as fluctuations in the signal-to-noise ratio of the water scattering characteristics. Geometric normalization eliminates the attitude-induced signal strength reference drift.

[0105] The physical meaning of this formula is: when the ship tilts, the beam becomes steeper ( < When the beam flattens, the effective scattering volume decreases, and the original signal becomes weaker. This is compensated by subtracting a negative value (i.e., adding a gain); conversely, when the beam flattens... When the effective scattering volume increases, the original signal becomes stronger. This is suppressed by subtracting a positive value, thus normalizing the output echo intensity. This method eliminates geometric projection errors caused by roll and pitch, and the numerical fluctuations only reflect the true changes in the concentration of suspended particles in the water, sound wave transmission loss, and the influence of environmental noise, no longer including systematic deviations introduced by the ship's motion attitude. This ensures the comparability of echo intensity data under different sea states and attitude conditions, providing clean input data free from attitude interference for accurate signal-to-noise ratio calculation in subsequent steps.

[0106] S4: Normalized echo intensity based on the output of S3 and the actual slant distance output by S2 Calculate the dynamic signal-to-noise ratio It accurately reflects the effective strength of the signal relative to noise, unaffected by attitude geometry factors and acoustic ray bending path errors. The dynamic signal-to-noise ratio is then compared with a preset threshold to generate a quality flag bit Q.

[0107] If the dynamic signal-to-noise ratio If the value is higher than the threshold, then Q=1, indicating that the layered data at the current depth is valid and marked as valid data; if the dynamic signal-to-noise ratio is... If the data is below the threshold, Q=0, and it is marked as invalid data, forming a data quality mask. This facilitates the dynamic determination of the signal-to-noise ratio through the corrected normalized echo intensity and path data, avoiding errors in transmission loss estimation caused by ray bending and false rejections caused by attitude fluctuations, thus improving the reliability of velocity profile data.

[0108] Dynamic signal-to-noise ratio The calculation formula is as follows:

[0109] in, The initial intensity constant of the acoustic wave emitted by the ADCP transducer is determined by the factory calibration of the equipment. In order to be in The background ambient noise intensity at any given time. This is obtained by statistically analyzing the mean value of the near-field region with no effective scattered signal. To accumulate the actual slope distance, This is the preset absorption coefficient.

[0110] Background ambient noise intensity During the calculation, the near-field dead zone (usually the first 1-3 range gates) close to the ADCP transducer, where effective echoes cannot be received due to transmit / receive switching and pulse length limitations, and the far-field region, where the acoustic energy has completely attenuated or exceeded the maximum measurement range, are first selected. The near-field dead zone and the far-field region are then combined into a calculation set. And output. In At any given time, the normalized echo intensity of all range gates within the set is considered. The background environmental noise intensity was calculated using robust statistics such as the median or truncated mean. The formula for calculating transient pulse interference and other outliers is as follows:

[0111]

[0112] in, A robust statistical calculator is a statistical calculation method that is insensitive to outliers, referring to the median or truncated mean. For the median: the data in the calculation set are sorted from smallest to largest, and the median value is taken. For the truncated mean: a certain percentage (e.g., 10% each) of the largest and smallest data points in the calculation set is removed, and the arithmetic mean of the remaining data is calculated. This prevents transient strong interference (such as bubble bursting, electron spike pulses, or sidelobe leakage) from occasionally occurring in the near-field blind zone from increasing the noise estimate, ensuring the background environmental noise intensity is maintained. It accurately reflects the stable background noise level, rather than instantaneous interference.

[0113] S5: Based on the actual longitude output by S2 ,latitude and depth Using the mass flags provided by S4, a weighted least squares solution is performed to output the final corrected velocity vector. This achieves a complete transformation from relative measurements in the ship's coordinate system to absolute values ​​in the geographic coordinate system, ensuring the dual accuracy of the unmanned surface vessel's flow velocity data in terms of both numerical precision and spatial positioning.

[0114] The specific steps for S5 are as follows:

[0115] S5.1: Receive the actual longitude of S2 ,latitude and depth The equivalent sound velocity profile of S1 is used to calculate the k of each beam at the t1. The radial flow velocity at each depth stratification; the calculation formula is as follows:

[0116]

[0117] in, This is the equivalent sound speed profile. The frequency of the ultrasonic waves emitted by the ADCP transducer. To fix the launch angle, For the first The beam at the 1st The frequency of the received echo signal at each depth of the layer is the same as the transmission frequency. difference.

[0118] S5.2: Based on the weighted least squares algorithm, the radial velocity observations of N beams are mapped to the three geographic components of east, north and vertical to obtain the velocity vector to be determined.

[0119] Firstly, regarding the first... Stratify the system of linear equations at different depths. ,in observation vector , The velocity vector to be determined . For an N×3 matrix, the first... element , Indicates the first Actual radial velocity observation vectors of each beam in each layer Compared with the velocity vector to be determined Through geometric projection matrix The difference between the calculated theoretical radial flow velocities.

[0120] Then introduce a diagonal weight matrix ,like =1 (data is valid), then =1. If =0 (invalid data), then =0.

[0121] Finally, the solution is obtained based on the least squares normal equation. This effectively prevents abnormal radial velocity contamination of the overall solution caused by excessively low signal-to-noise ratio, ensuring that a highly reliable velocity vector can still be output even in cases of partial beam failure (such as in 3-beam mode). The equations are as follows:

[0122]

[0123] S5.3: Bind the velocity vector to be determined with the absolute geographic coordinates and timestamps provided in S2 to generate the final hydrological sensor dataset. Each data point includes a complete spatiotemporal label and a quality score (the confidence level is determined by the number of effective beams involved in the solution).

[0124] This invention constructs a dynamic rotation model from the hull to the geographic coordinate system by real-time acquisition of unmanned surface vessel (USV) attitude data and combining it with static installation position vectors. This model accurately quantifies the horizontal displacement vector between the CTD sensor entry point and the ADCP transducer center caused by hull swaying. Based on this, an equivalent sound velocity profile matching the actual pointing area of ​​the ADCP beam is generated. This solves the problem in traditional processing where the dynamic changes in the sensor's spatial position are ignored, leading to incorrect refraction calculations using sound velocity data in waters with significant horizontal sound velocity gradients. It effectively eliminates systematic biases in ray tracking and avoids flow velocity measurement errors caused by spatial mismatch in sound velocity sampling.

[0125] Simultaneously, a geometric normalization correction based on the instantaneous incident angle is introduced. By normalizing the original echo intensity, the effective scattering volume projection scaling effect is eliminated, effectively removing the interference of attitude changes on signal intensity, so that the echo intensity only reflects the true changes in the water scattering characteristics. Based on this normalized intensity and the actual slant range output by ray tracing, the dynamic signal-to-noise ratio is calculated, which can truly reflect the effective intensity of the signal relative to the background noise, avoiding errors in transmission loss estimation caused by sound ray bending or false rejection caused by attitude fluctuations, and improving the real-time correction effect of sensor data.

[0126] An example is given below:

[0127] In a salt wedge frontal zone at a river estuary, an unmanned surface vessel (USV) equipped with an ADCP and CTD system was conducting operations. The system first collected data from a pressure sensor at a frequency of 10 Hz, including roll angle of 5°, pitch angle of -2°, yaw angle of 90°, and other parameters, and calculated the draft to be 1.5 meters.

[0128] Given that the static installation vector of the CTD relative to the center of the ADCP is [0, 2, -1] meters, after constructing the rotation matrix, the horizontal displacement vector of the CTD entry point relative to the center of the ADCP at the current moment is calculated to be [0.17, 1.98] meters.

[0129] Velocity of sound measured by the main CTD =1480m / s, measured by the auxiliary probe at a horizontal distance of 2m. =1470m / s, the horizontal sound velocity gradient is calculated. The horizontal sound velocity gradient and the superimposed horizontal displacement vector are then applied to the original profile using a depth-weighted function to generate an equivalent sound velocity profile.

[0130] Subsequently, the system divides the water body into 50 layers. Combining the ADCP beam with a fixed emission angle of 20° and the real-time attitude, it calculates the instantaneous incident angle and performs ray tracing to obtain the actual slant distance of the 10th layer as 15.2m and the cumulative horizontal offset, and then calculates the absolute geographic coordinates of the layer.

[0131] Given that the original echo intensity of this layer is -60dB, the normalized intensity is obtained by correcting it according to the formula. The system selects near-field blind zone data to calculate the background noise as -85dB, and substitutes it into the dynamic signal-to-noise ratio formula to obtain the dynamic signal-to-noise ratio. =15, which is higher than the preset threshold of 10, so the quality flag is set to 1. Finally, based on this valid data, the weighted least squares solution is used to output the precise east, north, and vertical velocity vectors for this depth, and the spatiotemporal label binding is completed.

[0132] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A real-time correction method based on hydrological observation sensor data of unmanned surface vessel motion attitude, characterized in that, Includes the following steps: S1: Real-time acquisition of unmanned surface vessel attitude data and static installation position vector data to construct the rotation matrix from the hull to the geographic coordinate system; then calculate the difference in horizontal position between the CTD sensor entry point and the ADCP transducer center in the geographic coordinate system, and determine the equivalent sound velocity profile of the ADCP beam pointing area. S2: Stratify the water body and calculate the instantaneous incident angle by combining it with the real-time attitude angle. It calculates the actual slant distance and horizontal offset of the sound ray accumulated in each layer, and outputs the actual longitude of the sampling points after eliminating the errors of sound ray bending and attitude coupling. ,latitude and depth ; S3: Based on instantaneous incident angle The ADCP beam is geometrically corrected to eliminate the effective scattering volume variation caused by the ship's tilt, thereby generating a normalized echo intensity. ; S4: Normalized echo intensity based on the output of S3 and the actual slant distance output by S2 Calculate dynamic signal-to-noise ratio The dynamic signal-to-noise ratio is compared with a preset threshold to generate a quality flag bit Q; S5: Based on the actual longitude output by S2 ,latitude ,depth We perform weighted least squares calculations on the mass flag and the equivalent sound velocity profile, and output the final corrected velocity vector to be determined.

2. The real-time correction method for hydrological observation sensor data based on the motion attitude of an unmanned surface vessel as described in claim 1, characterized in that, The specific steps of S1 are as follows: S1.1: The inertial measurement unit carried by the unmanned surface vessel collects the roll angle, pitch angle and bow angle of the hull in real time at a fixed sampling frequency, and calculates the draft at the same time; S1.2: Pre-program the static installation position vectors of the CTD sensor and ADCP transducer in the ship's coordinate system into the memory. ; S1.3: Based on the three attitude angles acquired in real time, construct a three-dimensional rotation matrix from the ship coordinate system to the geographic coordinate system. , used to describe the instantaneous attitude of a ship's hull in space at any given moment; S1.4: Combine the static mounting vector with the three-dimensional rotation matrix Multiply to generate a relative position vector; S1.5: Project the relative position vectors, retaining only the values ​​in the geographic coordinate system. The difference between the y-axis and the x-axis is used to obtain the horizontal displacement vector between the actual water entry point of the CTD sensor and the center point of the ADCP transducer at time t. ; S1.6: The horizontal sound velocity gradient estimate is calculated based on the real-time differential values ​​of the main CTD sensor and the sound velocity sensor mounted on this unmanned surface vessel. Combined with the horizontal displacement vector and depth weighting function The vertical sound velocity profile measured by the CTD sensor is corrected to the equivalent sound velocity profile of the ADCP beam pointing region. .

3. The real-time correction method for hydrological observation sensor data based on the motion attitude of an unmanned surface vessel according to claim 2, characterized in that, The specific content of S1.6 is as follows: First, the sound velocity measurement value of the main CTD sensor at the current water entry point is acquired synchronously. Then, based on the auxiliary sound velocity probe installed on the other side of the hull, the horizontal distance is obtained as follows: Sound velocity measurement at another measuring point Then, the estimated horizontal sound speed gradient is calculated. The calculation formula is as follows: in, It is the unit vector along the line connecting the two measuring points; Combined with the horizontal displacement vector and depth weighting function The vertical sound velocity profile measured by the CTD sensor is corrected to the equivalent sound velocity profile of the ADCP beam pointing region. Equivalent sound speed profile The calculation formula is as follows: in, For CTD sensors in The raw sound speed data obtained from the measurement at any time. For the current moment The horizontal displacement vector.

4. The real-time correction method for hydrological observation sensor data based on the motion attitude of an unmanned surface vessel according to claim 1, characterized in that, The specific content of S2 is as follows: 1) Divide the water body into N uniform thin layers along the depth direction; 2) Then read the fixed emission angle and azimuth angle parameters of the ADCP beam, and combine them with the real-time attitude angle data to calculate the instantaneous incident angle of the beam in the geographic coordinate system through coordinate transformation. The calculation formula is as follows: in, for The roll angle at any moment, for The pitch angle at any moment, The fixed emission angle of the beam in the ship's coordinate system. The azimuth angle of the beam in the ship's coordinate system; 3) According to the first The sound velocity and ray parameters of the layer are used to infer the glancing angle of the sound ray. The sound wave at the first grazing angle and water layer thickness is calculated. The horizontal and vertical displacement components of the layer are summed to obtain the sound wave propagation to the first layer. Actual slope distance of the layer and cumulative horizontal offset ; 4) Based on cumulative horizontal offset and geographical azimuth Output actual longitude ,latitude and depth The calculation formula is as follows: in, is the Earth's average radius, used to convert planar displacement into spherical angles. The index of the layer corresponding to the current distance gate, satisfying = , For unmanned surface vessels at all times Real-time draft depth, For unmanned surface vessels at the current moment Real-time location.

5. The real-time correction method for hydrological observation sensor data based on the motion attitude of an unmanned surface vessel according to claim 1, characterized in that, The normalized echo intensity The calculation formula is as follows; in, For ADCP sensor in The first time directly measured output, without attitude geometry correction. The echo signal strength of each distance gate, To fix the launch angle, The instantaneous angle of incidence.

6. The real-time correction method for hydrological observation sensor data based on the motion attitude of an unmanned surface vessel according to claim 1, characterized in that, When the dynamic signal-to-noise ratio is compared with a preset threshold, if the dynamic signal-to-noise ratio... If the value is higher than the threshold, the quality flag will be set. =1, indicating that the current depth of layered data is valid and marked as valid data; if the dynamic signal-to-noise ratio is... If it falls below the threshold, the quality flag bit... =0, marked as invalid data.

7. The real-time correction method for hydrological observation sensor data based on the motion attitude of an unmanned surface vessel according to claim 1, characterized in that, The specific steps of S5 are as follows: S5.1: Receive the actual longitude of S2 ,latitude and depth The equivalent sound velocity profile of S1 is used to calculate the k of each beam at the t1. Radial flow velocity at a depth of stratification; S5.2: Based on the weighted least squares algorithm, the radial velocity observations of N beams are mapped to the three geographic components of east, north and vertical directions to obtain the velocity vector to be determined. S5.3: Bind the velocity vector to be determined with the absolute geographic coordinates and timestamps provided in S2 to generate the final hydrological sensor dataset.

8. The real-time correction method for hydrological observation sensor data based on the motion attitude of an unmanned surface vessel according to claim 7, characterized in that, The radial flow velocity The calculation formula is as follows: in, This is the equivalent sound speed profile. The frequency of the ultrasonic waves emitted by the ADCP transducer. To fix the launch angle, For the first The beam at the 1st The frequency of the received echo signal at each depth of the layer is the same as the transmission frequency. difference.

9. The real-time correction method for hydrological observation sensor data based on the motion attitude of an unmanned surface vessel according to claim 1, characterized in that, The dynamic signal-to-noise ratio The calculation formula is as follows: in, The initial intensity constant of the acoustic wave emitted by the ADCP transducer is determined by the factory calibration of the equipment. exist Background ambient noise intensity at any given time To accumulate the actual slope distance, This is the preset absorption coefficient.

Images