Underwater positioning method of deep-sea profile buoy based on multi-source constraint calibration

Abstract

The application discloses a kind of underwater positioning methods of deep-sea profile buoy based on multi-source constraint calibration, belong to deep-sea monitoring technical field.Method is: buoy end gathers multi-source original data;Ground end carries out the pretreatment of noise reduction, filtering, time synchronization and outlier rejection to multi-source original data;Multi-source constraint calibration is carried out, constructs GPS position constraint model, depth constraint model and three-axis magnetometer constraint model, and the drift parameter of nine-axis attitude sensor is solved based on the data after pretreatment;Underwater positioning inversion is carried out, and the spatial position data of buoy underwater at any time is inverted based on the nine-axis attitude sensor data after calibration;Data matching association is carried out, and marine monitoring data and spatial position data are aligned according to time stamp, to generate associated data.The underwater positioning method of deep-sea profile buoy based on multi-source constraint calibration provided by the application has the characteristics of low power consumption, long endurance, small positioning error, high calibration accuracy, accurate data matching and strong scene adaptability.

Description

Technical Field

[0001] This invention belongs to the field of deep-sea monitoring technology, and in particular relates to an underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration. Background Technology

[0002] Deep-sea profiling buoys are core equipment for conducting global three-dimensional ocean observation. They can be equipped with sensors for temperature, salinity, and depth (CTD), dissolved oxygen, and chlorophyll, completing profiling data collection from the sea surface to depths of thousands of meters, providing crucial data support for ocean circulation simulation and climate change research. Accurate acquisition of the buoy's underwater spatial position is a prerequisite for matching monitoring data with geographic coordinates, directly determining the scientific value of the data.

[0003] In existing technologies, underwater positioning of deep-sea profiling buoys mainly relies on built-in nine-axis attitude sensors to collect acceleration and angular velocity data, providing a basis for buoy attitude perception and trajectory calculation. However, the harsh deep-sea operating environment (high pressure, temperature gradient, seawater disturbance) causes significant drift in the nine-axis attitude sensors, with the three-axis accelerometer mainly producing zero-bias drift. The constant deviation between the measured value and the true acceleration accumulates over time, with an error reaching ±0.05 m / s². The three-axis gyroscope exhibits zero-bias drift. ) and scaling factor drift ( The zero-bias drift error is ±0.1° / h, and the scale factor drift disrupts the linear relationship between the output and input angular velocities. Furthermore, existing technologies suffer from four major defects: 1) Disconnected calibration scenarios: Traditional laboratory calibration cannot simulate deep-sea conditions, resulting in calibration results deviating from on-site errors by over 30%; 2) Accumulated positioning errors: Without GPS signals underwater, inertial navigation calculation errors accumulate exponentially with the duration of descent and ascent, with conventional solutions exhibiting horizontal positioning errors on the order of hundreds of meters; 3) Data matching failure: Monitoring data can only be correlated with GPS endpoint intervals, failing to accurately match underwater three-dimensional coordinates and impacting the analysis of microscale ocean phenomena; 4) Significant power consumption contradictions: High-precision inertial measurement units consume high power, while low-power units have large errors, and some solutions embed algorithms into the buoy end, resulting in a runtime of only one month, failing to meet long-term monitoring needs.

[0004] Existing inertial navigation calibration technologies are mostly geared towards aviation and land-based scenarios, and cannot adapt to the characteristics of the deep sea, where there is no continuous GPS and the movement of the carrier is dominated by ocean currents. Therefore, there is an urgent need for a low-power technology solution based on multi-source constraints and integrated ground-based calibration and positioning inversion to solve the core problem of accurately matching deep-sea profiling buoy observation data with underwater spatial location. Summary of the Invention

[0005] In view of the shortcomings of the related technologies, the purpose of this invention is to provide an underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration, so as to solve the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] An underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration includes the following steps:

[0008] S1, Buoy End Data Acquisition

[0009] Before the buoy descends, during the buoy's descent, and after the buoy ascends, multi-source raw data are collected in stages. The multi-source raw data includes GPS location data, depth gauge data, nine-axis attitude sensor data, and ocean monitoring data. The nine-axis attitude sensor data includes three-axis accelerometer data, three-axis gyroscope angular velocity data, and three-axis magnetometer data.

[0010] S2, Ground-based Data Preprocessing

[0011] Noise reduction, filtering, time synchronization, and outlier removal are performed on multi-source raw data.

[0012] S3, Multi-source constraint calibration

[0013] Construct GPS position constraint model, depth constraint model and triaxial magnetometer constraint model, and calculate the drift parameters of the nine-axis attitude sensor based on the preprocessed data;

[0014] S4, Underwater Positioning Inversion

[0015] Based on the calibrated nine-axis attitude sensor data, the spatial position data of the buoy at any time underwater is retrieved through an inertial navigation recursive algorithm. The spatial position data includes the latitude, longitude and depth data of the buoy.

[0016] S5, Data Matching and Association

[0017] By aligning marine monitoring data with spatial location data according to timestamps, correlated data of time, latitude and longitude, depth and monitoring parameters are generated.

[0018] In some embodiments, in step S3, the GPS position constraint model is used to constrain the distance of the buoy's underwater horizontal movement trajectory, specifically including:

[0019] Assume the initial GPS position data before the buoy dives. GPS position data after the buoy rises ,but and Horizontal distance between two points for:

[0020]

[0021] in, The radius of the Earth's equator. The initial latitude, The initial longitude, The latitude after surfacing The longitude after surfacing;

[0022] At any time underwater, the buoy The cumulative horizontal distance satisfies:

[0023]

[0024] in, The moment the buoy dives. For any given moment underwater, The moment the buoy rises to the surface. For the buoy in the integration interval At any instant The corresponding eastward velocity, For the buoy in the integration interval At any instant The corresponding northbound speed;

[0025] Define GPS position error for:

[0026]

[0027] By using a GPS position constraint model, the cumulative horizontal distance of the buoy underwater is made equal to the initial GPS position of the buoy before it dives. GPS position after the buoy rises Linear interpolation distance matching between them.

[0028] In some embodiments, in step S3, the depth constraint model is used to correct the vertical position of the buoy, specifically including:

[0029] Computational depth gauge dynamic zero bias :

[0030]

[0031] in, The density of seawater, It is the acceleration due to gravity. For depth measurement data, This is the original water pressure value from the depth gauge;

[0032] The corrected vertical position of the buoy is obtained by using dynamic zero bias of the depth gauge. :

[0033]

[0034] Define depth error for:

[0035]

[0036] in, The vertical position of the buoy is calculated using inertial navigation.

[0037] The vertical position of the buoy, calculated using a depth-constrained model, is constrained by inertial navigation. Vertical position of the modified buoy Matching.

[0038] In some embodiments, in step S3, the triaxial magnetometer constraint model is used for buoy attitude calibration constraints, specifically including:

[0039] Calculate the data after magnetometer calibration :

[0040]

[0041] in, Let be the attitude rotation matrix from the buoy body coordinate system to the ground-fixed coordinate system. This is the raw data from the magnetometer. Zero bias for the magnetometer;

[0042] From calibrated angular velocity Recursive attitude rotation matrix Using matrix exponents and Formula approximation:

[0043]

[0044] in, The angular velocity antisymmetric matrix is ​​of the form: , , , It is the identity matrix;

[0045] Define magnetometer error for:

[0046]

[0047] in, This refers to the standard value of the geomagnetic field intensity in the operating sea area;

[0048] Using a triaxial magnetometer constraint model, the data after magnetometer calibration are constrained. The modulus value and the standard value of the geomagnetic field intensity in the operating sea area match.

[0049] In some embodiments, step S3 specifically includes calculating the drift parameters of the nine-axis attitude sensor:

[0050] To minimize the sum of squared errors from multiple constraints, construct the objective function. :

[0051]

[0052] in, , , These are the weighting coefficients. The drift parameters to be calculated contain 12 dimensions. ,set up , Triaxial accelerometer The axis is initially zero bias. Triaxial accelerometer Axis zero-bias drift rate, Three-axis gyroscope The axis is initially zero bias. Three-axis gyroscope Axis scaling factor;

[0053] Iterative update of drift parameters To minimize The iteration termination condition is the gradient magnitude. Or the number of iterations exceeds 100;

[0054] Based on the calculated drift parameters, the calibrated acceleration is obtained. With angular velocity :

[0055]

[0056]

[0057] in, This is the raw acceleration data from the triaxial accelerometer. The raw angular velocity data from the three-axis gyroscope is used for calculation. The unit is converted from degrees per second to radians per second.

[0058] In some embodiments, in step S4, the underwater positioning inversion uses a preset time step to match the sampling rate of the nine-axis attitude sensor. Combining the initial attitude and initial velocity, the spatial position data of the buoy at any underwater moment is obtained through a three-step recursive process of attitude update, velocity update, and position update. The initial attitude includes the buoy's initial pitch angle. Initial roll angle Initial yaw angle ,and , , All are 0; the initial velocity includes the buoy's eastward velocity. Northbound speed Vertical velocity ,and , , All are 0.

[0059] In some embodiments, in step S4, the attitude update is based on the calibrated angular velocity. The attitude angle quaternions are solved using the fourth-order Runge-Kutta method, and the attitude angles, including pitch angles, are calculated using the trapezoidal integral method. Roll angle and yaw angle The velocity update incorporates a correction term for Earth's rotation angular velocity, based on the calibrated acceleration. The eastward velocity is calculated using the trapezoidal integral method in conjunction with the attitude angle. Northbound speed and vertical velocity Obtain the velocity vector Position update is based on velocity vector. The differential relationship with the buoy's underwater spatial position is calculated by discrete integration of latitude and longitude changes, and the vertical position of the buoy is corrected by combining the depth constraint model. ,get The buoy's underwater position at any given time.

[0060] In some embodiments, in step S5, the marine monitoring data is aligned with the spatial location data based on the timestamp. When the sampling rates of the marine monitoring data and the spatial location data are inconsistent, linear interpolation is used to supplement the missing data.

[0061] In some embodiments, step S1, the buoy end data acquisition specifically includes:

[0062] Before the buoy dives, it remains stationary on the sea surface for a preset time. The GPS collects initial GPS position data, and the nine-axis attitude sensor simultaneously collects the initial values ​​of the three-axis accelerometer and the initial values ​​of the three-axis gyroscope angular velocity.

[0063] During the buoy's descent, the GPS enters a dormant state, the depth gauge collects depth data, the nine-axis attitude sensor collects three-axis accelerometer data, three-axis gyroscope angular velocity data, and three-axis magnetometer data, and simultaneously collects ocean monitoring data including at least temperature and salinity.

[0064] After the buoy rises to the surface, the GPS resumes operation and collects GPS position data of the buoy after it rises to the surface;

[0065] The raw data collected from multiple sources was transmitted to the ground via Iridium satellite.

[0066] In some embodiments, step S2, the ground-side data preprocessing specifically includes: using a wavelet denoising algorithm to denoise the triaxial accelerometer data and the triaxial gyroscope angular velocity data; using a Kalman filter to filter the triaxial magnetometer data; and setting a state noise matrix. Observation noise matrix Time synchronization is performed based on the buoy's internal clock; outliers are eliminated using the 3σ criterion.

[0067] Compared with the prior art, the beneficial effects of the present invention are:

[0068] 1. The underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration provided by this invention has the advantages of excellent calibration and positioning accuracy. Based on multi-source raw data, the drift parameters of the nine-axis attitude sensor are calculated through GPS position constraint model, depth constraint model and three-axis magnetometer constraint model, which effectively reduces the drift correction error of three-axis accelerometer and three-axis gyroscope, and the calibration accuracy is significantly improved compared with single constraint scheme. Combined with the calibrated data for inertial navigation recursive inversion, the underwater positioning error is far better than related schemes.

[0069] 2. The underwater positioning method of deep-sea profile buoy based on multi-source constraint calibration provided by the present invention has the characteristics of low power consumption and long endurance. The buoy end only collects raw data from multiple sources such as nine-axis attitude sensors and depth gauges (low power consumption working mode), and the calculation tasks such as positioning inversion are completed by high-performance equipment on the ground end, which significantly extends the underwater endurance of the buoy.

[0070] 3. The underwater positioning method of deep-sea profiling buoy based on multi-source constraint calibration provided by this invention has the characteristics of accurate data matching. It achieves accurate alignment between marine monitoring data and spatial location data based on timestamps. In the case of inconsistent sampling rates of the two types of data, linear interpolation is used to supplement missing data and establish four-dimensional correlation data including time, latitude and longitude, depth and monitoring parameters. This effectively solves the problem of disconnect between observation data and positioning information and ensures the accuracy of the correlation between observation data and positioning information.

[0071] 4. The underwater positioning method based on multi-source constraint calibration of deep-sea profiling buoys provided by this invention has the characteristics of strong scene adaptability, no need for laboratory pre-calibration, compatibility with various devices such as Argo buoys and deep-sea profiling buoys, and can achieve accurate matching of observation data with underwater spatial position, which is suitable for observation scenarios in global deep-sea areas. Attached Figure Description

[0072] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0073] Figure 1 This is a flowchart illustrating an embodiment of the underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration according to the present invention. Detailed Implementation

[0074] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0075] In the description of this invention, it should be understood that the terms "center", "lateral", "longitudinal", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0076] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0077] See appendix Figure 1 This paper presents an illustrative embodiment of the underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration proposed in this invention. The underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration includes the following steps:

[0078] S1, Buoy End Data Acquisition

[0079] Before the buoy dives, during the buoy's descent, and after the buoy rises, multi-source raw data are collected in stages. The multi-source raw data includes GPS location data, depth gauge data, nine-axis attitude sensor data, and ocean monitoring data. The nine-axis attitude sensor data includes three-axis accelerometer data, three-axis gyroscope angular velocity data, and three-axis magnetometer data.

[0080] S2, Ground-based Data Preprocessing

[0081] Noise reduction, filtering, time synchronization, and outlier removal are performed on multi-source raw data.

[0082] S3, Multi-source constraint calibration

[0083] Construct GPS position constraint model, depth constraint model and triaxial magnetometer constraint model, and calculate the drift parameters of the nine-axis attitude sensor based on the preprocessed data;

[0084] S4, Underwater Positioning Inversion

[0085] Based on the calibrated nine-axis attitude sensor data, the spatial position data of the buoy at any time underwater is retrieved through an inertial navigation recursive algorithm. The spatial position data includes the latitude, longitude and depth data of the buoy.

[0086] S5, Data Matching and Association

[0087] By aligning marine monitoring data with spatial location data according to timestamps, correlated data of time, latitude and longitude, depth and monitoring parameters are generated.

[0088] The underwater positioning method based on multi-source constraint calibration of deep-sea profiling buoys is implemented using an underwater positioning system based on multi-source constraint calibration of deep-sea profiling buoys. This underwater positioning system consists of a buoy-end hardware module and a ground-end processing module.

[0089] The buoy-end hardware module includes a low-power nine-axis attitude sensor, a high-precision depth gauge, a low-power GPS module, and a data storage / transmission module. The nine-axis attitude sensor collects raw three-axis acceleration data. ), raw data from the three-axis gyroscope ( ), triaxial magnetometer data ( In this embodiment, the ICM-20948 model can be selected, with a sampling rate of 0.5Hz and power consumption of <1mA. The depth gauge collects underwater depth data. This provides vertical position constraints; in this embodiment, the Series33X model can be selected, with a range of 0-6000m, an accuracy of ±0.02%FS, and a sampling rate of 1Hz. The GPS module collects data before diving. Collect after surfacing In this embodiment, the underwater hibernation mode can be selected from the UBLOXNEO-7M model, with a positioning accuracy of ±2.5mCEP and underwater shutdown capability. The data storage / transmission module stores the raw data and transmits it to the ground via Iridium satellite after surfacing. The storage capacity is 16GB, the transmission power consumption is <8mA, and it supports breakpoint resumption; Iridium 9602 module.

[0090] The ground-based processing module includes a data preprocessing unit, a multi-source constraint construction unit, a drift calibration unit, and an underwater positioning inversion unit. The data preprocessing unit is used for noise reduction, time synchronization, and outlier removal of raw data, and runs wavelet thresholding and Kalman filtering algorithms. The multi-source constraint construction unit is used to establish GPS position constraint models, depth constraint models, and triaxial magnetometer constraint models, fuse constraint equations, and generate an optimization objective function. The drift calibration unit is used to solve for the drift parameters of the triaxial accelerometer and triaxial gyroscope, employing... (LM) Nonlinear optimization algorithm. The underwater positioning inversion unit recursively derives the underwater position based on calibrated data, correlates with monitoring data, runs an inertial navigation recursive algorithm, and supports multi-parameter binding.

[0091] In step S3, the GPS position constraint model is used to constrain the distance of the buoy's underwater horizontal movement trajectory, specifically including:

[0092] Assume the initial GPS position data before the buoy dives. GPS position data after the buoy rises ,but and Horizontal distance between two points for:

[0093]

[0094] in, The radius of the Earth's equator. , The initial latitude, The initial longitude, The latitude after surfacing The longitude after surfacing;

[0095] At any time underwater, the buoy The cumulative horizontal distance satisfies:

[0096]

[0097] in, The moment the buoy dives. For any given moment underwater, The moment the buoy rises to the surface. For the buoy in the integration interval At any instant The corresponding eastward velocity, For the buoy in the integration interval At any instant Corresponding northward velocity; buoy eastward velocity buoy northward speed Acceleration after calibration Integral to obtain (initial velocity) );

[0098] Define GPS position error for:

[0099]

[0100] By using a GPS position constraint model, the cumulative horizontal distance of the buoy underwater is made equal to the initial GPS position of the buoy before it dives. GPS position after the buoy rises Linear interpolation distance matching between them.

[0101] In this embodiment, the depth constraint model is based on depthmeter data. The relationship with hydrostatic pressure is used to correct the vertical position of the buoy. ( (Axial direction downwards is positive). Specifically, in step S3, the depth constraint model is used to correct the vertical position of the buoy, including:

[0102] Computational depth gauge dynamic zero bias :

[0103]

[0104] in, The density of seawater, , For gravitational acceleration (corrected by the WGS-84 model): , For the buoy in (real-time latitude of the moment) For depth measurement data, This is the original water pressure value from the depth gauge;

[0105] The corrected vertical position of the buoy is obtained by using dynamic zero bias of the depth gauge. :

[0106]

[0107] Define depth error for:

[0108]

[0109] in, The vertical position (i.e., depth) of the buoy calculated for inertial navigation, under ideal conditions ;

[0110] The vertical position of the buoy, calculated using a depth-constrained model, is constrained by inertial navigation. Vertical position of the modified buoy Matching.

[0111] In this embodiment, the triaxial magnetometer constraint model utilizes the constant magnitude of the deep-sea geomagnetic field to constrain attitude calculations. Specifically, in step S3, the triaxial magnetometer constraint model is used for buoy attitude calibration constraints, including:

[0112] Calculate the data after magnetometer calibration :

[0113]

[0114] in, This is the attitude rotation matrix (3×3 orthogonal matrix) from the buoy body coordinate system to the ground-fixed coordinate system. This is the raw data from the magnetometer. Zero bias of the magnetometer (calculated synchronously with drift parameters). ;

[0115] From calibrated angular velocity Recursive attitude rotation matrix Using matrix exponents and Formula approximation:

[0116]

[0117] in, The angular velocity antisymmetric matrix is ​​of the form: , , , It is the identity matrix;

[0118] Define magnetometer error for:

[0119]

[0120] in, The standard value of the geomagnetic field intensity in the operating sea area (obtained from the China Geomagnetic Database, such as a certain area in the South China Sea) ); Ideally ;

[0121] Using a triaxial magnetometer constraint model, the data after magnetometer calibration are constrained. The modulus value and the standard value of the geomagnetic field intensity in the operating sea area match.

[0122] In step S3, the drift parameters of the nine-axis attitude sensor are calculated using a drift parameter calibration algorithm (LM optimization), specifically including:

[0123] To minimize the sum of squared errors from multiple constraints, construct the objective function. :

[0124]

[0125] in, , , These are the weighting coefficients. The drift parameters to be calculated contain 12 dimensions. ,set up , Triaxial accelerometer The axis is initially zero bias. Triaxial accelerometer Axis zero-bias drift rate (unit: ), Three-axis gyroscope Initial zero offset of shaft (unit: ), Three-axis gyroscope Axis scaling factor (dimensionless, ideal value 1).

[0126] In this embodiment, the weighting coefficient The weighting coefficients are adaptively adjusted based on the data accuracy.

[0127] Solving using the LM algorithm, iteratively updating the drift parameters. To minimize The iteration termination condition is the gradient magnitude. Or the number of iterations exceeds 100;

[0128] Based on the calculated drift parameters, the calibrated acceleration is obtained. With angular velocity :

[0129]

[0130]

[0131] in, This is the raw acceleration data from the triaxial accelerometer. The raw angular velocity data from the three-axis gyroscope is used for calculation. The unit is converted from degrees per second to radians per second, specifically from... Convert to : .

[0132] In step S4, the underwater positioning inversion uses a preset time step to match the sampling rate of the nine-axis attitude sensor. Combining the initial attitude and initial velocity, the spatial position data of the buoy at any underwater moment is obtained through three recursive steps: attitude update, velocity update, and position update. The initial attitude includes the buoy's initial pitch angle. Initial roll angle Initial yaw angle ,and , , All are 0; the initial velocity includes the buoy's eastward velocity. Northbound speed Vertical velocity ,and , , All are 0.

[0133] In this embodiment, the preset time step is: To match the three-axis gyroscope Sampling rate.

[0134] In step S4, the attitude update is based on the calibrated angular velocity. The attitude angle quaternions are solved using the fourth-order Runge-Kutta method, and the attitude angles, including pitch angles, are calculated using the trapezoidal integral method. Roll angle and yaw angle The velocity update incorporates a correction term for Earth's rotation angular velocity, based on the calibrated acceleration. The eastward velocity is calculated using the trapezoidal integral method in conjunction with the attitude angle. Northbound speed and vertical velocity Obtain the velocity vector Position update is based on velocity vector. The differential relationship with the buoy's underwater spatial position is calculated by discrete integration of latitude and longitude changes, and the vertical position of the buoy is corrected by combining the depth constraint model. ,get The buoy's underwater position at any given time.

[0135] The attitude update specifically includes:

[0136] In deep-sea environments, the raw data from a nine-axis attitude sensor is subject to drift and noise. Based on calibrated angular velocity data... The attitude angle differential equations are constructed using rigid body kinematics theory. Let the attitude angle quaternion of the carrier coordinate system relative to the navigation coordinate system be... ,but:

[0137]

[0138] in, , , To calibrate the angular velocity data of the three-axis gyroscope in the carrier coordinate system The angular velocity component of the axis.

[0139] Using the fourth-order Runge-Kutta method in each Numerical solution is performed within the sampling period:

[0140] Calculate intermediate variables :

[0141]

[0142] Calculate intermediate variables :

[0143]

[0144] Calculate intermediate variables :

[0145]

[0146] Calculate intermediate variables :

[0147]

[0148] Final updated attitude angle quaternions:

[0149]

[0150] Meanwhile, the pitch angle is solved using the following attitude angle differential equation and trapezoidal integral method. Roll angle Yaw angle (Initial posture) ):

[0151]

[0152]

[0153] Speed ​​updates specifically include:

[0154] Introducing the Earth's rotational angular velocity into the velocity differential equation Correction items:

[0155]

[0156] in, To calibrate the triaxial accelerometer data The measured value on the corresponding axis, Latitude For the Earth's radius, This is the acceleration due to gravity.

[0157] The trapezoidal integration method is used to solve the problem, where the initial velocity... .

[0158] Position updates are based on inertial navigation theory, comprehensively considering the effects of buoy velocity, Earth's rotation angular velocity, and Earth's curvature. Underwater position inversion is achieved through differential equation establishment, discrete integration, and unit correction. Specifically, this includes:

[0159] Step 1: Establishing the differential equation (fundamentals of dynamics)

[0160] The velocity change of the buoy in the navigation coordinate system (East-North-Sky, ENU) needs to be compensated for the additional acceleration caused by the Earth's rotation. Its differential equation is:

[0161]

[0162] in, The velocity vector of the buoy in the ENU coordinate system ( , The eastward velocity of the buoy. The buoy's northward speed. Vertical velocity of the buoy, unit: / ); Specific force vector (i.e., calibrated acceleration) measured by a triaxial accelerometer ,unit: / ); The vector of Earth's rotational angular velocity ( / The direction is along the Earth's rotation axis pointing towards the North Pole, and in the ENU coordinate system, it needs to be converted into local components through the attitude matrix. The position vector of the buoy relative to the center of the Earth (approximately) , The radius of the Earth's equator. , The underwater depth of the buoy, in units of: ); The gravitational acceleration vector (corrected by the WGS-84 model). (Direction along the negative direction of the sky). The zero bias vector of the triaxial accelerometer (already solved by the drift parameter calibration algorithm in step S3 and obtained from...) (Removed from the middle section; retained here for formula integrity). Noise vector measured for triaxial accelerometer (after wavelet denoising and Kalman filtering suppression, unit: ).

[0163] Key additional acceleration description: To compensate for the coupling effect between the Earth's rotation and the relative motion of the buoy by using Coriolis acceleration; Centripetal acceleration is used to compensate for the effect of centrifugal force generated by the Earth's rotation on velocity.

[0164] The second step is to update the differential relation (parameter mapping) at specific locations.

[0165] velocity vector With respect to the underwater spatial position of the buoy (latitude) ,longitude ,depth Direct correlation is established to create discrete update differential equations (ignoring higher-order terms for engineering calculations):

[0166]

[0167] in, Rate of change of latitude, unit: , northbound speed Decision, denominator Compensating for the effect of underwater depth on the Earth's radius of curvature; Rate of change of longitude, unit: , Eastward speed Decision, denominator Additional multiplication is required The circumference of the Earth's longitude circle decreases as latitude increases (it is largest at the equator and zero at the poles). The rate of change of depth is directly equal to the vertical velocity. (The z-axis is positive downwards, which is consistent with the direction of depth gauge data).

[0168] Step 3: Solve using the integral method (discretization calculation)

[0169] Because the buoy data acquisition is discrete sampling (three-axis gyroscope 0.5), Hz Depth gauge 1 Hz The above continuous differential equation needs to be converted into discrete integral form. The steps are as follows:

[0170] First, define the discretization parameters:

[0171] (1) Start time: (GPS positioning time before buoy descent, corresponding to initial position) );

[0172] (2) Current moment: (At any time underwater);

[0173] (3) Number of sampling points: (from arrive Total number of samples, adjacent sampling interval during uniform sampling ,unit: );

[0174] (4) Discrete sampling time: .

[0175] Secondly, the correlation between angular velocity components and rate of change:

[0176] Define the angular velocity components in the latitudinal direction ,Right now Rate of change of latitude at any given time, in units of: Define the angular velocity components in the longitude direction. ,Right now Unit of rate of change of longitude at any given time: .

[0177] Then, the position change is calculated using discrete integration:

[0178] The calculation is performed by approximating integration using rectangular summation (which satisfies the accuracy requirements in engineering; trapezoidal integration can be used to further improve accuracy). arrive Latitude variation Longitude variation :

[0179]

[0180] Finally, unit conversion (radians → degrees):

[0181] The inertial navigation calculation output is in radians, which needs to be converted to the geographically commonly used degree system. ):

[0182]

[0183] Step 4: Final Position Update Formula (Complete Positioning Result)

[0184] Combined with initial GPS location data ( The depth correction results of the depth constraint model are obtained. Full underwater position of the buoy :

[0185] Latitude update (degree system):

[0186]

[0187] in, Initial GPS latitude, The latitude increment obtained by integration;

[0188] Longitude update (degree system):

[0189]

[0190] in, Initial GPS longitude, The longitude increment obtained by integration;

[0191] Deep update (corrected, unit: ):

[0192]

[0193] in, For inertial navigation to calculate depth, The depth gauge dynamic zero bias is calculated for the depth constraint model to ensure that the depth accuracy is consistent with the depth gauge.

[0194] In step S5, the marine monitoring data is aligned with the spatial location data based on the timestamp. When the sampling rates of the marine monitoring data and the spatial location data are inconsistent, linear interpolation is used to supplement the missing data. In this embodiment, a monitoring data and location matching algorithm is used, specifically, the monitoring data (temperature) is aligned based on the timestamp. ,salinity ( ) and location data, using linear interpolation to supplement missing values ​​(e.g., monitoring data sampling rate 1) Hz Location data 0.5 Hz hour):

[0195]

[0196]

[0197] in, , , To monitor adjacent sampling times, a "time-latitude-longitude-depth-monitoring parameter" association table is generated, supporting NetCDF ocean data format output.

[0198] In step S1, the data acquisition at the buoy end specifically includes: before the buoy dives ( After the sea surface remains stationary for a preset time (e.g., 30 minutes), the GPS acquires initial GPS position data. The nine-axis attitude sensor synchronously acquires the initial values ​​of the three-axis accelerometer. Initial value of angular velocity of three-axis gyroscope During the buoy's descent (underwater operations), GPS enters sleep mode, depth gauge at 1 Hz Sampling rate for collecting depth gauge data The nine-axis attitude sensor has a speed of 0.5 Hz The sampling rate is used to acquire triaxial accelerometer data. Angular velocity data from a three-axis gyroscope and triaxial magnetometer data and simultaneously collect at least temperature data. and salinity Ocean monitoring data; after the buoy rises ( GPS resumed operation and collected GPS position data after the buoy rose to the surface. The collected raw data from multiple sources is transmitted to the ground via Iridium satellite.

[0199] In this embodiment, in step S2, the ground-end data preprocessing specifically includes: employing a wavelet noise reduction algorithm ( threshold Noise reduction is performed on the triaxial accelerometer data and triaxial gyroscope angular velocity data. Error from Down to , Error from Down to Kalman filtering was used to filter the triaxial magnetometer data, and a state noise matrix was set. Observation noise matrix , Fluctuations from Down to Time synchronization is achieved by aligning data with the buoy's internal clock (±1ms accuracy), with a deviation of ≤5ms, thus realizing millisecond-level time synchronization; outliers are removed using the 3σ criterion (removal rate <0.5%).

[0200] In the above illustrative embodiments, the underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration can accurately solve the drift calibration problem of the low-power nine-axis attitude sensor built into the deep-sea profiling buoy. It constructs multi-source constraints using the initial GPS position before diving, the GPS position after surfacing, high-precision depth gauge data, and three-axis magnetometer data. Relying on ground-based processing, it achieves coordinated drift calibration and underwater positioning inversion. This effectively addresses the pain point of accurately matching deep-sea observation data with underwater spatial position, providing a reliable spatial coordinate benchmark for marine environmental research, resource development, disaster early warning, and engineering applications. It is adaptable to scenarios such as marine environmental monitoring and deep-sea resource exploration, possessing clear practical value and application relevance.

[0201] Finally, it should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0202] The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them; although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific implementation of the present invention or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solutions of the present invention, and all such modifications and substitutions should be covered within the scope of the technical solutions claimed in the present invention.

Claims

1. An underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration, characterized in that, Includes the following steps: S1, Buoy End Data Acquisition Before the buoy dives, during the buoy's descent, and after the buoy rises, multi-source raw data are collected in stages. The multi-source raw data includes GPS location data, depth gauge data, nine-axis attitude sensor data, and ocean monitoring data. The nine-axis attitude sensor data includes three-axis accelerometer data, three-axis gyroscope angular velocity data, and three-axis magnetometer data. S2, Ground-based Data Preprocessing The multi-source raw data is subjected to noise reduction, filtering, time synchronization, and outlier removal. S3, Multi-source constraint calibration Construct GPS position constraint model, depth constraint model and triaxial magnetometer constraint model, and calculate the drift parameters of the nine-axis attitude sensor based on the preprocessed data; S4, Underwater Positioning Inversion Based on the calibrated nine-axis attitude sensor data, the spatial position data of the buoy at any time underwater is retrieved through an inertial navigation recursive algorithm. The spatial position data includes the latitude, longitude and depth data of the buoy. S5, Data Matching and Association The ocean monitoring data and spatial location data are aligned with timestamps to generate associated data of time, latitude and longitude, depth and monitoring parameters; In step S3, the GPS position constraint model is used to constrain the distance of the buoy's underwater horizontal movement trajectory, specifically including: Assume the initial GPS position data before the buoy dives. GPS position data after the buoy rises ,but and Horizontal distance between two points for: in, The radius of the Earth's equator. The initial latitude, The initial longitude, The latitude after surfacing The longitude after surfacing; At any time underwater, the buoy The cumulative horizontal distance satisfies: in, The moment the buoy dives. For any given moment underwater, The moment the buoy rises to the surface. For the buoy in the integration interval At any instant within The corresponding eastward velocity, For the buoy in the integration interval At any instant within The corresponding northbound speed; Define GPS position error for: The GPS position constraint model ensures that the buoy's cumulative underwater horizontal distance is equal to its initial GPS position before descent. GPS position after the buoy rises Linear interpolation distance matching between them.

2. The underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration according to claim 1, characterized in that, In step S3, the depth constraint model is used to correct the vertical position of the buoy, specifically including: Computational depth gauge dynamic zero bias : in, The density of seawater, It is the acceleration due to gravity. For depth measurement data, This is the original water pressure value from the depth gauge; The corrected vertical position of the buoy is obtained through the dynamic zero bias of the depth gauge. : Define depth error for: in, The vertical position of the buoy is calculated using inertial navigation. The vertical position of the buoy, calculated using the depth constraint model, is constrained by inertial navigation. Vertical position of the modified buoy Matching.

3. The underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration according to claim 2, characterized in that, In step S3, the triaxial magnetometer constraint model is used for buoy attitude calibration constraints, specifically including: Calculate the data after magnetometer calibration : in, Let be the attitude rotation matrix from the buoy body coordinate system to the ground-fixed coordinate system. This is the raw data from the magnetometer. Zero bias for the magnetometer; From calibrated angular velocity Recursive attitude rotation matrix Using matrix exponents and Formula approximation: in, The angular velocity antisymmetric matrix is ​​of the form: , , , It is the identity matrix; Define magnetometer error for: in, This refers to the standard value of the geomagnetic field intensity in the operating sea area; The triaxial magnetometer constraint model is used to constrain the calibrated magnetometer data. The modulus value and the standard value of the geomagnetic field intensity in the operating sea area match.

4. The underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration according to claim 3, characterized in that, In step S3, calculating the drift parameters of the nine-axis attitude sensor specifically includes: To minimize the sum of squared errors from multiple constraints, construct the objective function. : in, , , These are the weighting coefficients. The drift parameters to be calculated contain 12 dimensions. ,set up , Triaxial accelerometer The axis is initially zero bias. Triaxial accelerometer Axis zero-bias drift rate, Three-axis gyroscope The axis is initially zero bias. Three-axis gyroscope Axis scaling factor; Iterative update of drift parameters To minimize The iteration termination condition is the gradient magnitude. Or the number of iterations exceeds 100; Based on the calculated drift parameters, the calibrated acceleration is obtained. With angular velocity : in, This is the raw acceleration data from the triaxial accelerometer. The raw angular velocity data from the three-axis gyroscope is used for calculation. The unit is converted from degrees per second to radians per second.

5. The underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration according to claim 4, characterized in that, In step S4, the underwater positioning inversion uses a preset time step to match the sampling rate of the nine-axis attitude sensor. Combining the initial attitude and initial velocity, the spatial position data of the buoy at any underwater moment is obtained through a three-step recursive process of attitude update, velocity update, and position update. The initial attitude includes the buoy's initial pitch angle. Initial roll angle Initial yaw angle And the , , All are 0; the initial velocity includes the buoy's eastward velocity. Northbound speed Vertical velocity And the , , All are 0.

6. The underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration according to claim 5, characterized in that, In step S4, the attitude update is based on the calibrated angular velocity. The attitude angle quaternions are solved using the fourth-order Runge-Kutta method, and the attitude angles, including pitch angles, are calculated using the trapezoidal integral method. Roll angle and yaw angle The velocity update incorporates a correction term for the Earth's rotation angular velocity, based on the calibrated acceleration. The eastward velocity is calculated using the trapezoidal integral method in conjunction with the attitude angle. Northbound speed and vertical velocity Obtain the velocity vector The position update is based on the velocity vector. The differential relationship with the buoy's underwater spatial position is calculated by discrete integration of latitude and longitude changes, and the vertical position of the buoy is corrected by combining the depth constraint model. ,get The buoy's underwater position at all times.

7. The underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration according to claim 1, characterized in that, In step S5, the marine monitoring data is aligned with the spatial location data based on the timestamp. When the sampling rates of the marine monitoring data and the spatial location data are inconsistent, the missing data is supplemented by linear interpolation.

8. The underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration according to claim 1, characterized in that, In step S1, the data acquisition at the buoy end specifically includes: Before the buoy dives, it remains stationary on the sea surface for a preset time. The GPS collects initial GPS position data, and the nine-axis attitude sensor simultaneously collects the initial values ​​of the three-axis accelerometer and the initial values ​​of the three-axis gyroscope angular velocity. During the buoy's descent, the GPS enters a dormant state, the depth gauge collects depth data, the nine-axis attitude sensor collects three-axis accelerometer data, three-axis gyroscope angular velocity data, and three-axis magnetometer data, and simultaneously collects ocean monitoring data including at least temperature and salinity. After the buoy rises to the surface, the GPS resumes operation and collects GPS position data of the buoy after it rises to the surface; The raw data collected from multiple sources was transmitted to the ground via Iridium satellite.

9. The underwater positioning method for deep-sea profiling buoys based on multi-source constraint calibration according to claim 1, characterized in that, In step S2, the ground-side data preprocessing specifically includes: using wavelet denoising algorithm to denoise the triaxial accelerometer data and triaxial gyroscope angular velocity data; using Kalman filtering to filter the triaxial magnetometer data; and setting the state noise matrix. Observation noise matrix Time synchronization is performed based on the buoy's internal clock; outliers are eliminated using the 3σ criterion.

Images