Time-varying gain adaptive control method for multi-beam bathymetric sonar and related products

Abstract

The application relates to the technical field of sonar signal processing, and provides a time-varying gain adaptive control method of a multi-beam bathymetric sonar and related products. The method comprises the following steps: obtaining a salinity profile according to the sound velocity profile data and environmental parameters of a current measurement position, and calculating a continuously depth-varying absorption coefficient profile; based on the absorption coefficient profile, combining sound ray tracking results, and layering a water body along the depth to obtain accurate absorption loss of the beam signal caused by water absorption; based on the absorption coefficient profile and a curved sound ray path integral model, a two-dimensional lookup table is established; based on the two-dimensional lookup table, according to the incident angle and one-way propagation time of the current beam, the two-dimensional lookup table is queried or interpolated to obtain an absorption loss correction value and apply the absorption loss correction value to time-varying gain compensation of a return signal; if the propagation time exceeds the maximum range of the two-dimensional lookup table, the absorption coefficient corresponding to the last row node of the two-dimensional lookup table is used for extrapolation calculation.

Description

Technical Field

[0001] This application relates to the fields of underwater acoustic engineering and sonar signal processing technology, and in particular to a time-varying gain adaptive control method and related products for multibeam bathymetry sonar. Background Technology

[0002] Multibeam echo sounders are one of the mainstream devices for ocean depth measurement, widely used in seabed topography mapping, channel dredging, and marine engineering surveys. When multibeam echo sounders are mounted on unmanned platforms (e.g., unmanned vessels and autonomous underwater vehicles), the limitations of platform size and power consumption place higher demands on the automation level and real-time processing capabilities of the sonar system.

[0003] When sound waves propagate in water, their energy attenuates due to geometric expansion and water absorption. Absorption loss is primarily caused by factors such as seawater viscosity and ion relaxation effects, and is closely related to water temperature, salinity, and sound wave frequency. To ensure the detectability of echo signals at different distances, multibeam echo sounders typically employ time-varying gain control techniques to dynamically amplify and compensate for the echo signal. However, most existing time-varying gain control methods calculate absorption loss based on fixed empirical absorption coefficients or simplified linear propagation models, failing to adequately consider the variations in absorption coefficients with depth and geographical location, or the impact of sound ray bending in non-uniform water bodies on the actual propagation path length. This leads to significant deviations in the estimation of absorption loss in areas with significant thermohaline transitions and drastic changes in sound velocity profiles, thus affecting the accuracy of time-varying gain compensation and reducing the reliability of bathymetry data. Summary of the Invention

[0004] Based on this, it is necessary to address the shortcomings of the existing technologies, such as reliance on fixed absorption coefficients or linear propagation models leading to large deviations in absorption loss estimation, and propose a time-varying gain adaptive control method and related products for multibeam bathysound sonar.

[0005] In a first aspect, a time-varying gain adaptive control method for multibeam bathysound sonar is provided, the method comprising: Based on the sound velocity profile data and environmental parameters at the current measurement location, the corresponding salinity profile is obtained, and the temperature profile is obtained by inversion based on the empirical formula for sound velocity. The absorption coefficient profile that changes continuously with depth is then calculated. Based on the absorption coefficient profile, the water body is layered along the depth by combining the sound ray tracking results. The sound velocity and absorption coefficient within each layer are regarded as equal gradient changes. The absorption coefficient is integrated along the actual curved sound ray path of each layer, and the absorption loss of all layers is accumulated to obtain the accurate absorption loss of the beam signal caused by water absorption. Based on the absorption coefficient profile and the curved sound path integral model, during the system initialization or route planning stage, a two-dimensional lookup table is pre-established using the beam incident angle and signal one-way propagation time as indexes to store the accurate absorption loss values ​​under the corresponding beam incident angle and one-way propagation time. Based on the two-dimensional lookup table, during the actual depth sounding process, the absorption loss correction value is obtained by querying or interpolating the two-dimensional lookup table according to the current beam's incident angle and one-way propagation time, and then applied to the time-varying gain compensation of the echo signal. If the one-way propagation time of the current beam exceeds the maximum range of the two-dimensional lookup table, then the absorption coefficient corresponding to the last node of the two-dimensional lookup table is used to extrapolate the calculation according to the straight-line propagation model.

[0006] Secondly, a time-varying gain adaptive control device for a multibeam bathymetry sonar is provided, the device comprising: The first acquisition module is used to obtain the corresponding salinity profile based on the sound velocity profile data and environmental parameters at the current measurement location, and to obtain the temperature profile based on the empirical formula of sound velocity, and to calculate the absorption coefficient profile that changes continuously with depth. The first calculation module is used to divide the water body into layers along the depth based on the absorption coefficient profile and the sound ray tracking results, treat the sound velocity and absorption coefficient within the layer as equal gradient changes, integrate the absorption coefficient along the actual curved sound ray path of each layer, and accumulate the absorption loss of all layers to obtain the accurate absorption loss of the beam signal caused by water absorption. The table building module is used to build a two-dimensional lookup table based on the absorption coefficient profile and the curved sound path integral model, in advance during the system initialization or route planning stage, using the beam incident angle and signal one-way propagation time as indexes to store the accurate absorption loss values ​​under the corresponding beam incident angle and one-way propagation time. The second acquisition module is used to, based on the two-dimensional lookup table, in the actual depth sounding process, perform a query or interpolation in the two-dimensional lookup table according to the current beam's incident angle and one-way propagation time to obtain the absorption loss correction value and apply it to the time-varying gain compensation of the echo signal. The second calculation module is used to extrapolate the calculation using the absorption coefficient corresponding to the last row node of the two-dimensional lookup table according to the straight-line propagation model if the one-way propagation time of the current beam exceeds the maximum range of the two-dimensional lookup table.

[0007] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the aforementioned time-varying gain adaptive control method for multibeam echo sounder.

[0008] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the aforementioned time-varying gain adaptive control method for multibeam echo sounder.

[0009] As can be seen from the technical solution provided in this application: Firstly, by obtaining an absorption coefficient profile that continuously varies with depth, the limitations of traditional methods using fixed absorption coefficients are overcome, enabling the estimation of absorption loss to truly reflect the water absorption characteristics of the current sea area. Furthermore, integrating the absorption coefficient along the actual curved acoustic ray path avoids path length errors caused by treating acoustic rays as straight-line propagation, thus obtaining a more accurate beam signal absorption loss value. Secondly, by pre-generating a two-dimensional lookup table indexed by the beam incident angle and one-way propagation time, the complex curved acoustic ray integration calculation is completed offline, reducing the online computational burden and meeting the real-time processing requirements of unmanned platforms. Thirdly, by quickly querying or interpolating the absorption loss correction value during actual depth sounding and applying it to time-varying gain compensation, a balance between high precision and high efficiency is achieved, effectively improving the depth sounding data quality of multibeam bathysound sonar in different hydrological environments. In summary, the technical solution of this application can adaptively provide accurate time-varying gain compensation based on the actual water temperature and salinity distribution and acoustic ray bending effect, enhancing the environmental adaptability and measurement reliability of multibeam bathysound sonar. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart of a time-varying gain adaptive control method for a multibeam echo sounder in one embodiment; Figure 2 This is a graph showing the absorption coefficient under different conditions in one embodiment; Figure 3 This is a two-dimensional lookup table for absorption loss at an acoustic frequency of 26.5 kHz in one embodiment; Figure 4 This is a structural block diagram of a time-varying gain adaptive control device for a multibeam echo sounder in one embodiment. Figure 5 This is a structural block diagram of a computer device in one embodiment. Detailed Implementation

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0012] When sound waves propagate in water, their energy attenuates due to geometric expansion and water absorption. Absorption loss is primarily caused by factors such as seawater viscosity and ion relaxation effects, and is closely related to water temperature, salinity, and sound wave frequency. To ensure the detectability of echo signals at different distances, multibeam echo sounders typically employ time-varying gain control techniques to dynamically amplify and compensate for the echo signal. However, most existing time-varying gain control methods calculate absorption loss based on fixed empirical absorption coefficients or simplified linear propagation models, failing to adequately consider the variations in absorption coefficients with depth and geographical location, or the impact of sound ray bending in non-uniform water bodies on the actual propagation path length. This leads to significant deviations in the estimation of absorption loss in areas with significant thermohaline transitions and drastic changes in sound velocity profiles, thus affecting the accuracy of time-varying gain compensation and reducing the reliability of bathymetry data.

[0013] To address the aforementioned problems in the prior art, this application proposes a time-varying gain adaptive control method for multibeam bathysound sonar, the main process of which is as follows: Figure 1 As shown, it mainly includes steps S1 to S5, which are detailed below: Step S1: Based on the sound velocity profile data and environmental parameters at the current measurement location, obtain the corresponding salinity profile, and invert the temperature profile based on the empirical formula for sound velocity to calculate the absorption coefficient profile that changes continuously with depth.

[0014] In traditional multibeam sonar systems, the absorption coefficient is typically obtained by acquiring conductivity, temperature, and depth (CTD) profile data in the field. However, in autonomous operation scenarios using unmanned platforms, it is difficult to carry CTD sensors due to limitations in platform size, power consumption, and cost. To address this issue, this application employs a small, platform-mountable sound velocity profiler (such devices are small, low-power, and suitable for unmanned platforms) to acquire sound velocity profile data and environmental parameters. These are then combined with global salinity model inversion to obtain temperature and salinity profiles, thereby obtaining the absorption coefficient profile without relying on CTD. It should be noted that the aforementioned environmental parameters, in addition to the sound velocity profile data, are essential positioning information for spatial interpolation in the global salinity model, including at least the latitude and longitude coordinates of the current measurement location. Furthermore, if the global salinity model includes a time dimension (e.g., monthly average datasets), it can further include the operation time (e.g., month) to reflect the seasonal variations in the salinity profile.

[0015] Before or during bathymetry operations on the unmanned platform, the system first receives sound velocity profile data at the current measurement location. This sound velocity profile data can be obtained through various means, such as real-time measurement by a small sound velocity profiler on the platform, acquisition via an XBT (Expendable Bathythermograph) drop-out probe, or estimation using sea surface temperature and climatological data retrieved from satellite remote sensing. In this embodiment, the sound velocity profile data is obtained by measuring at the operation's starting point using a sound velocity profiler on the platform, with a depth resolution of 1 meter, covering the entire water column from the surface to the seabed.

[0016] After obtaining the sound velocity profile data, the system calls the Argo global salinity model (e.g., WOA2018, World Ocean Atlas 2018) and uses inverse distance weighted interpolation (IDW) to interpolate the salinity profile at the corresponding location from the model based on the latitude and longitude coordinates of the current measurement location. The basic principle of IDW is that the attribute value of the point to be interpolated is obtained by taking the attribute values ​​of known points around it as a weighted average of the inverse distance, with the weights inversely proportional to the distance. In this embodiment, the four nearest grid points around the current measurement location are selected, and their salinity values ​​and the inverse of the square of the distance are used as weights for interpolation, thereby obtaining a salinity profile at the same depth as the sound velocity profile.

[0017] Next, the system selects the Leroy empirical formula for the speed of sound as the model for calculating the speed of sound. The Leroy empirical formula for the speed of sound is a widely used empirical formula, and its expression is: c = 1402.5 + 5T - 5.44×10- ²T² + 2.19×10 -4 T³ + 1.35(S - 35) - 1.56×10 - ²z Where c is the speed of sound (m / s), T is the temperature (°C), S is the salinity (‰), and z is the depth (m). Given the sound velocity profile c(z) and the salinity profile S(z), the above formula can be considered a cubic equation about temperature T. The system solves this cubic equation to derive the corresponding temperature value at each depth point, thus obtaining the complete temperature profile T(z). Specifically, the Newton-Raphson method can be used for numerical solution, with the initial guess set to the temperature value of the previous depth point, and the convergence condition set to a temperature change of less than 0.001°C. Since the sound velocity profile has high depth resolution, the temperature change between adjacent depth points is usually small, so the Newton-Raphson method converges quickly, generally requiring only 3-5 iterations to obtain a solution that meets the accuracy requirements. For regions with drastic temperature changes, such as strata, if the Newton-Raphson method fails to converge after 10 iterations, the system automatically switches to the bisection method for solution. The initial interval of the bisection method is set to [-2℃, 35℃]. Iteration continues until the interval width is less than 0.001℃. It usually takes 15 to 20 iterations to converge, and it has the advantage of being unconditionally stable.

[0018] After obtaining the temperature profile T(z), salinity profile S(z), and depth profile z, and combining this with the sonar's current operating frequency f (26.5 kHz in this embodiment), the system calls the Francois-Garrison absorption coefficient model to calculate the absorption coefficient profile that continuously varies with depth. α (z). The Francois-Garrison model is currently the internationally recognized high-precision seawater acoustic absorption model, which comprehensively considers three main absorption mechanisms: viscous absorption of pure water, relaxation absorption of MgSO4, and relaxation absorption of B(OH)3. The calculation formula of this model is: α = A1P1f1f² / (f²+f1²) + A2P2f2f² / (f²+f2²) + A3P3f² Where A1, A2, and A3 are the relaxation absorption coefficient amplitude factors related to temperature and salinity, respectively; P1, P2, and P3 are pressure correction factors; and f1 and f2 are the relaxation frequencies of B(OH)3 and MgSO4, respectively. The specific values ​​of these parameters can be calculated using the standard calculation formula of the Francois-Garrison model, based on the temperature, salinity, and pressure values ​​at the current depth point. By substituting the temperature, salinity, and pressure values ​​at each depth point, the absorption coefficient at that depth point can be calculated. Figure 2As shown, the absorption coefficient varies significantly with frequency under different temperature, salinity, and depth conditions. The method of this application can accurately fit these curves, thereby obtaining an accurate absorption coefficient profile. Figure 2 Typical curves showing the absorption coefficient as a function of frequency under different combinations of temperature, salinity, and depth are presented. As can be seen from the figures, in the low-frequency range (e.g., 10–50 kHz), the absorption coefficient increases approximately quadratically with frequency; in the high-frequency range, the growth trend gradually slows down. Meanwhile, increasing temperature significantly increases the absorption coefficient, while the effect of salinity is relatively small but not negligible. Through the above steps, this application can automatically and accurately obtain the absorption coefficient profile at the current measurement location, laying a solid foundation for subsequent accurate calculations of absorption loss.

[0019] As one embodiment of this application, obtaining the corresponding salinity profile based on the sound velocity profile data and environmental parameters at the current measurement location, and then inverting the temperature profile based on the empirical formula for sound velocity, can be achieved through the following steps S1a and S1b, which are detailed below: Step S1a: The system calls the Argo global salinity model WOA2018 (World Ocean Atlas 2018) and uses the inverse distance weighted interpolation method to interpolate the salinity profile based on the latitude and longitude of the current measurement location.

[0020] The Argo Global Salinity Model (WOA2018, World Ocean Atlas 2018) is a global ocean climatological dataset released by the National Ocean Data Center (NODC) of the United States. It provides monthly and annual average salinity fields with a spatial resolution of 0.25 × 0.25, covering 102 standard layers vertically from 0 to 5500 meters. Since the WOA2018 model provides gridded data, and the current measurement location is typically located between grid points, spatial interpolation is needed to obtain the salinity profile at that location. This embodiment uses inverse distance-weighted interpolation (IDW). The core idea is that the attribute value of the point to be interpolated is obtained by taking the attribute values ​​of the surrounding known grid points as a weighted average of the inverse distances. The weights are inversely proportional to the p-th power of the distance (usually p = 2, i.e., the inverse of the square of the distance). In practice, the four nearest grid points (i.e., the smallest rectangular grid surrounding the current measurement location) are selected, and the Euclidean distance d_i from the measurement point to each grid point is calculated. Then, the interpolation result is calculated using the following formula: in, Let be the salinity value at depth z for the i-th grid point. Since the standard layer depth of the WOA2018 model may differ from the depth sampling points of the sound velocity profile, vertical linear interpolation of the WOA2018 salinity data is required before interpolation to align it with the depth grid of the sound velocity profile. Through these steps, a high-precision salinity profile sampled at the same depth as the sound velocity profile can be obtained.

[0021] Step S1b: Using Leroy's empirical formula for the speed of sound, the temperature profile is obtained by solving the cubic equation using Newton's iterative method.

[0022] The Leroy empirical formula for the speed of sound, as before, is a cubic equation in terms of temperature T. At each depth point z, substituting the known c(z) and S(z) into the formula, we obtain the cubic equation in terms of temperature T: 2.19×10 -4 T³ - 5.44×10 - ² T² + 5T + [1402.5 + 1.35(S-35) - 1.56×10 - ²z - c] = 0 Let a = 2.19 × 10 -4 b = -5.44 × 10 - ², c=5, d=1402.5+1.35(S-35)-1.56×10 - If ²z - c, then the equation becomes aT³ + bT² + cT + d = 0. The system uses Newton's iteration method to solve this equation. The iterative format of Newton's iteration method is: Where f(T) = aT³ + bT² + cT + d, f'(T) = 3aT² + 2bT + c. The initial guess value T0 is taken as the temperature value of the previous depth point (15℃ is taken as the initial value for the surface layer), and the iteration termination condition is set as | - |<0.001℃. Because the sound velocity profile has high depth resolution (typically 1 meter), the temperature change between adjacent depth points is usually less than 0.1℃. Therefore, the Newton-Raphson iteration method can converge quickly, generally requiring only 3-5 iterations to obtain a solution that meets the accuracy requirements. For individual depth points where convergence is difficult (e.g., areas near mesa with drastic temperature changes), the upper limit of the number of iterations can be increased to 20, and a bisection method can be used as a backup to ensure convergence.

[0023] Through steps S1a and S1b described above, this application achieves automatic acquisition of temperature and salinity profiles without the need for CTD sensors, providing a reliable data source for subsequent absorption coefficient calculations. Compared to directly using the global average temperature and salinity model, this method significantly improves the accuracy of temperature inversion by introducing measured sound velocity profiles as constraints, especially in sea areas with thermoclines, where the deviation between the inverted temperature and the measured temperature can be controlled within 0.5℃.

[0024] In another embodiment of this application, the empirical formula for the speed of sound can be the Medwin formula, and a bisection method is used for iteration when solving the cubic equation to accelerate the convergence speed. The Medwin formula is in the form of: c = 1449.2 + 4.6T - 5.5×10 - ² T² + 2.9 × 10 -4 T³ + (1.34 - 10 - ² T)(S -35) + 1.6×10 - ² z Compared to Leroy's empirical formula for the speed of sound, the Medwin formula shows slightly different fitting accuracy under low-temperature and low-salinity conditions, but it still meets engineering requirements in most marine environments. After rearranging the Medwin formula into a cubic equation about temperature (T), the system uses a bisection method for iterative solution. The basic idea of ​​the bisection method is to first determine the possible range of temperature values; for example, by setting a lower limit based on historical data. Upper limit Then calculate the midpoint. and calculate ;like and If the sign is different, the upper limit will be updated to... Otherwise, update the lower limit to Repeat the above process until the interval length is less than a preset threshold. The advantages of the bisection method are good stability, no divergence problem, and predictable convergence speed (the interval is halved with each iteration). For cubic equations, it typically requires 15-20 iterations to achieve an accuracy of 0.001℃. In this embodiment, when the system detects that the Newton iteration method converges slowly at a specific depth point (failure to converge after more than 10 iterations), it automatically switches to the bisection method as an alternative to ensure the robustness of temperature inversion.

[0025] It should be noted that, in addition to the Leroy empirical formula for the speed of sound and the Medwin formula, those skilled in the art may also choose other empirical formulas for the speed of sound as needed, such as the Chen-Millero formula, the Del Grosso formula, etc., as long as the formula can deduce the temperature from the speed of sound and salinity.

[0026] Step S2: Based on the absorption coefficient profile and combined with the sound ray tracking results, the water body is divided into layers along the depth. The sound velocity and absorption coefficient within each layer are considered to change at equal gradients. The absorption coefficient is integrated along the actual curved sound ray path of each layer, and the absorption loss of all layers is accumulated to obtain the accurate absorption loss of the beam signal caused by water absorption.

[0027] Traditional methods for calculating absorption loss typically assume that sound rays propagate in a straight line and use a constant average absorption coefficient. However, in real-world marine environments, the sound ray bends due to the variation in sound speed with depth, resulting in an actual propagation path length greater than the straight-line distance. Simultaneously, the absorption coefficient varies continuously with depth, introducing systematic biases when using an average value. The combination of these two factors leads to absorption loss estimation errors of tens of decibels in deep-sea or strongly stratified environments using traditional methods, severely impacting the accuracy of time-varying gain compensation.

[0028] To address this issue, this application establishes a layer integral model along the curved acoustic ray path, namely, the curved acoustic ray path integral model. First, based on the absorption coefficient profile obtained in step S1... α Using the existing sound velocity profile c(z) and the ray-tracing results from multibeam sonar, the water body is divided into several thin layers along the depth direction. Ray-tracing is a routine step in multibeam sonar signal processing, and its purpose is to calculate the actual propagation path of each beam based on the sound velocity profile. In this embodiment, the ray-tracing uses the classic Snell's law ray tracing method, dividing the water body into layers according to the principle of equal sound velocity gradient. The thickness of each layer does not exceed a preset sound velocity change threshold (set to 1 m / s in this embodiment), so that the sound velocity and absorption coefficient within the layer can be considered to change linearly.

[0029] For each beam, the propagation path length of its sound ray within each layer is: The path length can be calculated geometrically based on the sound velocity gradient and the beam's incident angle within the layer. Since both the sound velocity and absorption coefficient within the layer are considered linearly variable, the integral of the absorption coefficient within that layer can be calculated precisely. Specifically, for the i-th layer, assuming its top and bottom absorption coefficients are respectively... and The absorption loss within this layer It can be represented as: This method employs the trapezoidal numerical integration method, assuming that the absorption coefficient varies linearly with the path within each layer. By summing the absorption losses of all layers, the precise total loss of the beam signal due to water absorption during propagation can be obtained. : This layered integration method along the actual curved sound ray path simultaneously considers the effects of path length changes caused by sound ray curvature and vertical variations in the absorption coefficient, significantly improving the accuracy of absorption loss correction in complex sound velocity profile environments. In the simulation verification in the lake environment of this embodiment, compared with the traditional straight-line propagation model, the absorption loss calculated by this method at the maximum beam incidence angle (approximately 60°) differs by about 2.3 dB. This difference will significantly affect the consistency of the echo signal in subsequent time-varying gain compensation.

[0030] The aforementioned curved acoustic path integral model is based on acoustic ray tracking results. It divides the water body into several thin layers along the depth direction, treating the sound velocity and absorption coefficient as linear changes within each layer. Then, it integrates the absorption coefficient of each layer along the actual curved acoustic path, accumulating the integral results of all layers to obtain a calculation model of the total loss of the beam signal caused by water absorption. In the curved acoustic path integral model of step S2, the fineness of the water layering directly affects the integration accuracy. Traditional methods typically use fixed depth intervals (e.g., 1 meter) for layering. This method may lead to excessively large changes in sound velocity within a layer in regions with drastic sound velocity gradient changes (e.g., strata), thus violating the isogradability assumption; while in regions with gentle sound velocity gradients (e.g., deep-sea isotherms), it causes unnecessary computational redundancy. To solve this problem, this embodiment uses the isogradability principle for adaptive layering. That is, in the above embodiment, when layering the water body along the depth direction based on acoustic ray tracking results, it is divided according to the isogradability principle, ensuring that the sound velocity change within each layer does not exceed a preset sound velocity change threshold, making the sound velocity and absorption coefficient within the layer approximately linear. Specifically, the system traverses the sound speed profile downwards from the surface, and whenever the cumulative change in sound speed reaches a preset threshold... (In this embodiment of the application, it can be set) When the sound velocity is 1 m / s, a layer is formed, and the top and bottom depths of that layer, along with the corresponding sound velocity and absorption coefficient, are recorded. In this way, the change in sound velocity within each layer does not exceed [a certain value]. This ensures that the sound velocity within the layer can be approximated as a linear change; correspondingly, since the absorption coefficient is a continuous function of temperature, salinity, and depth, it can also be considered a linear change within such a small layer thickness. This adaptive stratification method can automatically densify stratification in the thermocline region (where the layer thickness may only be tens of centimeters) and sparsely stratify stratification in the isothermal region (where the layer thickness can reach tens of meters), minimizing computational load while ensuring integration accuracy. In the lake measurement data of this embodiment, the following method is used... The stratification strategy with a speed of 1 m / s generated a total of about 80 strata, while using a fixed interval of 1 meter would require 200 strata, reducing the computational load by about 60%.

[0031] It should be noted that, in the above embodiments, the uniform gradient change of the sound velocity and absorption coefficient within the layer means that within each layer, the sound velocity and absorption coefficient change linearly with depth, that is, the sound velocity gradient and absorption coefficient gradient remain constant within the layer.

[0032] Step S3: Based on the absorption coefficient profile obtained in step S1 and the curved acoustic path integral model used in step S2, during the system initialization or route planning stage, a two-dimensional lookup table is pre-established using the beam incident angle and signal one-way propagation time as indexes to store the accurate absorption loss values ​​under the corresponding beam incident angle and one-way propagation time.

[0033] Although the curved ray path integral model in step S2 can accurately calculate the absorption loss of each beam, the computational cost is substantial. For each beam, ray tracing, path integration, and accumulation operations need to be performed layer by layer. With a large number of beams (modern multibeam sonar typically has 128-512 beams) and updates multiple times per second, executing this model in real-time online would consume significant computational resources, making it difficult to meet the stringent real-time processing requirements (typically requiring millisecond-level response) of unmanned platforms. Traditional solutions either have to simplify the computational model (e.g., adopting the straight-line propagation assumption), sacrificing accuracy for speed; or rely on high-performance computing hardware, increasing platform power consumption and cost. Neither of these solutions is suitable for autonomous operation scenarios of unmanned platforms.

[0034] To resolve the aforementioned contradictions, this application innovatively adopts an architecture of "offline precise calculation + online rapid lookup". Specifically, during the system initialization or flight path planning stage, the system pre-calculates and establishes a two-dimensional lookup table with the beam incidence angle and signal one-way propagation time as key indexes based on the absorption coefficient profile obtained in step S1 and the curved acoustic ray path integral model used in step S2. The table stores the precise absorption loss values ​​for the corresponding beam incidence angle and one-way propagation time.

[0035] In this embodiment, the construction process of the two-dimensional lookup table is as follows: First, the index range of the two-dimensional lookup table is determined. The range of beam incident angle is determined by the sector opening angle of the multibeam sonar. In this embodiment, the sector opening angle of the sonar is ±60°, so the incident angle range is 0° to 60°, the angle step is set to 0.5°, and there are a total of 121 angle grid points. The range of signal one-way propagation time is determined by the maximum depth sounding and the sound velocity profile. In this embodiment, the maximum depth sounding is about 200 meters, corresponding to a one-way propagation time of about 0.13 seconds (calculated based on an average sound velocity of 1500 m / s). Considering the safety margin, the time range is set to 0 to 0.3 seconds, the time step is set to 1 millisecond, and there are a total of 301 time grid points. Therefore, the total number of grid points in the lookup table is 121 × 301 ≈ 36421.

[0036] For each grid point The system calls the absorption coefficient profile obtained in step S1. α ( z The integral model of the curved acoustic path in step S2 is used to calculate the beam at the incident angle. Dissemination time Precise absorption loss value under the conditions In specific calculations, the propagation time is first taken into account. The sound velocity profile c(z) is used to deduce the corresponding propagation distance and depth range through ray tracing. Then, the absorption coefficient is integrated layer by layer along this curved ray path to obtain the accurate absorption loss value. The calculation results of all grid points are stored in a two-dimensional lookup table, forming a two-dimensional array.

[0037] The two-dimensional lookup table can be constructed during the system initialization phase (e.g., during the unmanned platform's power-on self-test) or the route planning phase (e.g., before the platform enters a new survey area). Since the absorption coefficient profile changes slowly over a short period (unless the platform crosses a significant ocean front or stratospheric region), a lookup table constructed once can be reused for a considerable time without frequent reconstruction. This "one-time calculation, multiple uses" model decouples complex integration calculations from the real-time processing chain, significantly reducing the burden of online computation.

[0038] It should be noted that the step size of the two-dimensional lookup table has a significant impact on both accuracy and efficiency. An excessively large step size will reduce interpolation accuracy, while an excessively small step size will increase storage space and construction time. In this embodiment, the 0.5° angle step size and 1 millisecond time step size are compromises chosen after balancing accuracy and efficiency. Therefore, in one embodiment of this application, when building the two-dimensional lookup table, the beam incidence angle step size and propagation time step size are adaptively determined based on the gradient distribution of the sound velocity profile at the current measurement location: the time step size is decreased within the propagation time range corresponding to depths with larger sound velocity gradients; the time step size is increased within the propagation time range corresponding to depths with smaller sound velocity gradients; and the angle step size is decreased in regions where the beam incidence angle is close to the critical angle.

[0039] In the process of constructing the lookup table in step S3, the choice of step size is a key issue. While a fixed step size is simple to implement, it cannot balance accuracy and efficiency. This embodiment employs an adaptive step size strategy to optimize the grid layout of the lookup table. For the time step, the system first calculates the absolute value of the sound velocity gradient based on the sound velocity profile. Distribution with depth. In depth regions with large sound velocity gradients (e.g., mezzanines), >0.1 s -¹), due to the dramatic curvature of the sound ray and the rapid change in the absorption coefficient, more frequent time sampling is required. Specifically, the system divides the propagation time axis into several intervals, each corresponding to a certain depth range, with a time step size within that interval. It is inversely proportional to the absolute value of the average sound speed gradient in that interval: ,in, As a reference step size (in this embodiment, it can be...) (Set to 2ms) This represents the global average sound speed gradient. This represents the absolute value of the average sound velocity gradient within this interval. Thus, the time step can be reduced to 0.5 ms in the strata region and increased to 5 ms in the isothermal region. For the angular step size, the system focuses on the region near the critical angle. The critical angle is the maximum angle of incidence at which sound rays just avoid total internal reflection, and it depends on the ratio of the surface sound velocity to the bottom sound velocity. Near the critical angle, the sound ray path is extremely sensitive to the angle of incidence; even small angle changes can lead to significant differences in propagation path and absorption loss. Therefore, the system reduces the angular step size to 0.2° within ±5° of the critical angle, while maintaining a step size of 0.5° in other regions. This adaptive step size strategy increases the total number of grid points in the lookup table by only about 30%, but improves the interpolation accuracy in key regions by about 50%, achieving a good balance between accuracy and efficiency.

[0040] During long-term autonomous operation of the unmanned platform, the temperature and salinity structure of the water body may change over time, such as tidal-induced stratification and frontal displacement. These changes can cause deviations between the absorption coefficient profile obtained in step S1 based on the global salinity model and the actual value, thus affecting the accuracy of the lookup table. To address this issue, this application introduces a lookup table update mechanism based on measured sound velocity profiles. Specifically, after acquiring new sound velocity profile data, the measured sound velocity profile is compared with the theoretical sound velocity profile obtained through global salinity model inversion. If the deviation at the same depth exceeds a preset threshold, the absorption coefficient profile and accurate absorption loss value are recalculated based on the measured sound velocity profile, and the two-dimensional lookup table is updated accordingly. Specifically, the unmanned platform acquires new sound velocity profile data periodically or as needed during operation (through a sound velocity profiler onboard the platform or a drop-off probe). Each time a new measured sound velocity profile is acquired... Subsequently, the system simultaneously uses the method in step S1 (based on global salinity model inversion) to calculate a theoretical sound speed profile. The system then compares the sound speed values ​​of the two at the same depth point and calculates the deviation at each depth point. If the maximum deviation at all depth points Exceeding the preset threshold (In this embodiment, it can be set to 2 m / s). If the current global salinity model inversion results are no longer accurately reflecting the actual hydrological conditions, an update is required. At this point, the system abandons the inversion results based on the global salinity model and directly uses the measured sound velocity profile. As input, steps S1 (skipping the salinity and temperature inversion steps and directly calculating the absorption coefficient profile using the measured sound velocity profile, or re-inverting the temperature and salinity profile based on the measured sound velocity profile), S2, and S3 are re-executed to generate a new two-dimensional lookup table, which replaces the old table for use. It should be noted that the preset threshold mentioned in the above-mentioned lookup table update mechanism based on the measured sound velocity profile (i.e., "if the deviation between the two at the same depth exceeds the preset threshold") can be set according to the hydrological characteristics of the operating sea area: in sea areas where the sound velocity profile changes drastically with time and space (e.g., estuaries, frontal zones), a smaller value (e.g., 1 m / s) can be set to improve update sensitivity; in hydrologically stable sea areas such as the deep sea, a larger value (e.g., 3 m / s) can be set to reduce unnecessary updates. In this embodiment, 2 m / s is chosen as the preferred value that balances sensitivity and computational efficiency.

[0041] This bias-triggered update mechanism has the following advantages: First, it only triggers updates when there are significant environmental changes, avoiding unnecessary computational overhead; second, it utilizes the high precision of measured sound velocity profiles to compensate for potential biases in local sea areas within global salinity models; third, the updated lookup table can promptly reflect the latest hydrological conditions, ensuring the continuous accuracy of time-varying gain compensation. In the lake measurements of this embodiment, due to the large daily variation in lake water temperature, updates are typically triggered twice a day, once in the morning and once in the afternoon. Each update takes approximately 2 seconds (including lookup table reconstruction), without affecting the normal depth sounding operation schedule.

[0042] In the aforementioned two-dimensional lookup table update mechanism, when the deviation between the measured sound velocity profile and the theoretical sound velocity profile exceeds a threshold, the entire lookup table needs to be recalculated. However, in some cases, the change in the sound velocity profile may be limited to a specific depth range (for example, only the surface water temperature changes, while the deep water remains unchanged). In this case, recalculating all grid points would result in unnecessary computational waste. To address this issue, another embodiment of this application introduces a local update strategy. That is, when updating the two-dimensional lookup table, only the grid points within the beam incidence angle range and one-way propagation time range affected by the change in the sound velocity profile are recalculated, while the remaining grid points remain unchanged. Specifically, in the aforementioned two-dimensional lookup table update mechanism, the system first compares the differences between the old and new sound velocity profiles to identify the depth range where the sound velocity change exceeds a certain threshold (i.e., the "affected depth range"). Then, the system analyzes the one-way propagation time range and beam incidence angle range corresponding to these depth ranges. The one-way propagation time range can be obtained through ray tracing inversion: for a given depth range, the time required for the sound wave to propagate from the surface to the top and bottom of that range is calculated, thereby determining the affected time range. The beam incidence angle range is determined based on geometric relationships: only beams passing through the affected depth range are affected, and their incidence angle range depends on the horizontal position and ray curvature of that depth range. After determining the affected time and angle ranges, the system only re-executes steps S1 and S2 for grid points within that range, while grid points outside the range remain unchanged. This local update strategy significantly reduces the amount of update computation. In a typical update scenario of this embodiment, the measured sound velocity profile only changed significantly within the surface 0-20 meters (heating by about 2°C), corresponding to a propagation time range of 0-27 milliseconds and a beam incidence angle range of 0-45°. The affected grid points account for approximately 15% of all grid points, therefore the update computation is only 15% of that of full reconstruction, reducing the update time from 2 seconds to 0.3 seconds, with almost no impact on the continuity of the sounding operation.

[0043] It should be noted that the periodic update and the deviation-triggered update in the above embodiments can work together, that is, the update is triggered based on which one arrives first. If the deviation update has not been triggered when the periodic update timer arrives, the periodic update is performed and the timer is reset; if the deviation-triggered update occurs first, the periodic timer is reset simultaneously after the update is performed, so as to avoid wasting computing resources due to the interval between the two updates being too short.

[0044] Step S4: Based on the two-dimensional lookup table, during the actual depth sounding process, according to the current beam's incident angle and one-way propagation time, a lookup or interpolation is performed in the two-dimensional lookup table to obtain the absorption loss correction value, which is then applied to the time-varying gain compensation of the echo signal.

[0045] After the lookup table is pre-built, absorption loss compensation during actual depth sounding becomes extremely efficient. For all beams corresponding to each transmitted pulse, the system performs the following operations sequentially: First, for the current beam, obtain its incident angle θ and one-way propagation time t. The incident angle θ is directly given by the beamforming algorithm and is a known quantity; the one-way propagation time t is calculated by combining the bottom detection time (or water column sampling time) of the beam with the sound velocity profile.

[0046] Secondly, determine whether the current beam one-way propagation time t is within the time range of the lookup table. Inside. like If so, a lookup table is used for querying or interpolation. Specifically, the system determines the value based on the current beam incidence angle. and one-way transmission time Locate its four nearest grid points in the lookup table. , , as well as Then, the bilinear interpolation method is used to calculate the absorption loss correction value corresponding to the current beam. The formula for bilinear interpolation is: in, , , , , and These represent the absorption loss values ​​for four grid points. Bilinear interpolation can fully utilize the information from neighboring grid points, obtaining smoother and more accurate results than nearest-neighbor interpolation.

[0047] Step S5: If the one-way propagation time of the current beam exceeds the maximum range of the two-dimensional lookup table, then the absorption coefficient corresponding to the last node of the two-dimensional lookup table is used to extrapolate the calculation according to the straight-line propagation model.

[0048] If the one-way propagation time t of the current beam exceeds the maximum range of the lookup table (i.e., If the propagation time is too long, the absorption loss value cannot be directly obtained from the lookup table. This situation typically occurs in extremely deep waters or under abnormal propagation conditions. In this case, the system uses an extrapolation method: it uses the absorption coefficient corresponding to the last row node of the two-dimensional lookup table (i.e., the row with the longest propagation time). Extrapolation calculations are performed using a linear propagation model. Specifically, the extrapolated absorption loss value is calculated as follows: The calculation is as follows: Where R is the one-way propagation distance of the current beam, which is determined by the propagation time t and the surface sound speed. The product is approximately obtained as (R = t × t). The multiplication factor is 2 because sound waves need to travel back and forth (transmit and receive). The absorption coefficient is taken from the last node of the lookup table, corresponding to the angle closest to the current beam incidence angle. This extrapolation method is based on the following physical assumptions: within the depth range beyond the lookup table, the absorption coefficient approximately remains the value at the last node of the lookup table, and the ray bending effect is no longer significant, allowing for approximation as linear propagation. Although this is an approximation, the extrapolation error is within an acceptable range because the distance beyond the range is usually small, and the absorption coefficient tends to stabilize with increasing depth.

[0049] Finally, the obtained absorption loss correction value (or This is applied to time-varying gain compensation of the beam echo signal. Specifically, the time-varying gain amplifier adjusts its gain curve according to the absorption loss correction value, ensuring that the amplitude of the amplified echo signal is independent of distance, thereby recovering the true background reflection intensity and backscattering characteristics. In this embodiment, the time-varying gain control circuit receives a gain control word from the digital signal processor. This control word is calculated from the absorption loss correction value, with a control accuracy of 0.1 dB and an update period synchronized with the beam transmission period (approximately 10 milliseconds).

[0050] Figure 3 This table illustrates the absorption loss lookup table constructed under the conditions of this embodiment (sound frequency 26.5kHz). The horizontal axis represents the one-way propagation time of the signal (in milliseconds), and the vertical axis represents the beam incidence angle (in degrees). The color intensity represents the absorption loss value (in dB). The table clearly shows that, for the same propagation time, beams with larger incidence angles have greater absorption losses than those with smaller incidence angles. This is because beams with larger incidence angles have longer propagation paths and experience more absorption in shallow water. At the same incidence angle, the longer the propagation time (corresponding to greater distance), the greater the absorption loss, and the growth trend is approximately linear. This lookup table visually demonstrates the two-dimensional variation of absorption loss with incidence angle and propagation time, providing a data foundation for subsequent online queries.

[0051] Through the coordination of steps S3 to S5, this application achieves an innovative architecture of "offline precise calculation + online rapid lookup": Step S3 completes the complex bending ray integration calculation offline in advance and solidifies it in the form of a lookup table; Steps S4 and S5 only require constant-time lookup or interpolation operations in the online stage to obtain high-precision absorption loss correction values. This architecture perfectly balances the contradiction between high precision and real-time performance, enabling unmanned platforms to achieve adaptive time-varying gain control with limited computing power.

[0052] During long-term autonomous operation of an unmanned platform, even if environmental changes have not triggered a deviation threshold, a slow drift trend may still occur (such as seasonal water temperature changes). To prevent this slow drift from accumulating to a negligible level, the system employs a periodic update strategy. Specifically, during unmanned platform operation, the system reacquires the sound velocity profile data at the current measurement location every preset time interval or travel distance, and repeats steps S1 to S3 to dynamically update the two-dimensional lookup table. Specifically, every preset time interval (30 minutes in this embodiment) or travel distance (5 kilometers in this embodiment), regardless of whether a deviation threshold is triggered, the system actively reacquires the sound velocity profile data at the current measurement location and repeats steps S1 to S3 to dynamically update the two-dimensional lookup table. The time interval and travel distance can be adjusted according to the severity of hydrological changes in the operating area: in areas with drastic hydrological changes, such as estuaries, this can be shortened to 10 minutes or 1 kilometer; in areas with stable hydrology, such as the ocean, it can be extended to 1 hour or 10 kilometers. This periodic update mechanism ensures that the lookup table always reflects the latest hydrological conditions, providing a reliable guarantee for long-term autonomous operation.

[0053] While periodic update strategies are simple and reliable, they have two potential problems: first, the update time is fixed and cannot adapt to the rate of environmental change; second, unnecessary updates may occur when the hydrological environment has not changed significantly, wasting computational resources. To address these issues, this application introduces a dynamic triggering mechanism based on changes in the absorption coefficient profile as a supplement or replacement for periodic updates. Specifically, the triggering condition for reacquiring sound velocity profile data in the above embodiments can be: when the root mean square error of the two most recently acquired absorption coefficient profiles at the same depth exceeds a preset threshold, an update is immediately triggered. In detail, the system executes step S1 each time and obtains a new absorption coefficient profile. Then, compare it with the absorption coefficient profile used in the previous update. Compare the two and calculate the root mean square error (RMSE) at the same depth point: Where N represents the number of depth sampling points. If the RMSE value exceeds the preset threshold RMSE_thresh (set to 0.5 dB / km in this embodiment), the absorption coefficient profile is considered to have changed significantly, requiring an immediate update, regardless of whether the periodic update timing has arrived. This dynamic triggering mechanism based on changes in the absorption coefficient profile can respond promptly to sudden environmental changes (e.g., frontal passage, tidal-induced stratification changes, etc.) while avoiding unnecessary updates when the environment is stable. In the lake field measurements of this embodiment, this mechanism successfully captured the change in absorption coefficient caused by the rapid rise in surface water temperature (approximately 0.5℃ / hour) due to afternoon solar radiation, triggering the update before the periodic update timing arrived, ensuring the continuity of compensation accuracy.

[0054] In practical multibeam echo sonar systems, echo signal attenuation comes not only from water absorption but also from the geometric spread of the sound wave. Spread loss, caused by energy dispersion due to the increasing propagation distance of the sound wavefront, together with absorption loss, determines the total attenuation of the echo signal. Therefore, complete time-varying gain compensation needs to consider both absorption loss and spread loss simultaneously. Figure 1 The example method could also include: adding the absorption loss correction value to the extended loss correction value calculated based on the spherical extended model to generate a comprehensive time-varying gain value, which is then simultaneously applied to the gain control of the echo signal. Specifically, the system obtains the absorption loss correction value... Then, the correction value for the expanded loss is further calculated. Extended loss correction value The calculation uses a spherical extension model, which is applicable to the propagation of a point sound source in a free field. Its formula is: Where R is the one-way propagation distance of the sound wave, in meters. 20log 10 (R) represents the sound intensity attenuation caused by spherical expansion, in dB. In actual calculations, R is calculated by combining the one-way propagation time t of the current beam with the sound velocity profile c(z), usually by multiplying the average sound velocity along the propagation path by the propagation time.

[0055] Then, the system will absorb the loss correction value. With extended loss correction value Add them together to generate the combined time-varying gain value. : The integrated time-varying gain value It is also applied to gain control of the echo signal. In specific implementation, the digital signal processor... The value generates a gain control word, which is sent to the preamplifier or variable gain amplifier to adjust its amplification factor for the echo signal. Because... It incorporates the effects of both absorption and spread losses; therefore, the amplitude of the compensated echo signal should theoretically be independent of distance, reflecting only the reflection and backscattering characteristics of the seabed sediment. This comprehensive compensation method makes echo signals at different distances comparable, providing high-quality input data for subsequent bottom tracking, sediment classification, and water column image generation.

[0056] In the above embodiments of this application, the comprehensive time-varying gain value Although accurate compensation for absorption loss and spread loss is already included, due to the discretization nature of the lookup table, the differences between adjacent beams... The gain value may exhibit slight fluctuations. Although these fluctuations are small (typically on the order of 0.1~0.2 dB), they can manifest as horizontal stripes or granular noise in the backscattered image, affecting the visual quality of the image and subsequent quantitative analysis. To address this issue, this embodiment applies a moving average filter to the integrated time-varying gain value to eliminate gain fluctuations caused by lookup table discretization, and then applies the filtered gain value to the gain control of the echo signal. The basic principle of moving average filtering is to take the arithmetic mean of the integrated time-varying gain values ​​of the current beam and several beams before and after it, using this as the final gain control value for the current beam. Specifically, for the k-th beam, its filtered gain value... for: Where M is the length of the sliding window, and an odd number is chosen to ensure symmetry. In this embodiment, M is 5, meaning the gain values ​​of the current beam and the two beams to its left and right are averaged. The choice of sliding window length requires a trade-off between smoothing effect and signal fidelity: a window that is too short will result in poor smoothing, while a window that is too long may smooth out the actual changes in seabed reflection intensity. M=5 can effectively eliminate gain jumps while better preserving gain fluctuations caused by changes in seabed sediment. For beams at the edge of a sector (e.g., the first and last beams), since the number of beams on both sides is insufficient, the window length can be appropriately reduced (e.g., M=3) or the original gain value can be used directly.

[0057] After moving average filtering, the gain variation between adjacent beams becomes smooth and continuous, and the lateral stripes in the backscattered image are effectively suppressed. In the lake field test of this application embodiment, the comparison of backscattered images before and after filtering shows that the texture of the filtered image is more natural and uniform, and the stability and success rate of bottom tracking are significantly improved. This filtering method is computationally simple and has strong real-time performance, making it very suitable for implementation on embedded platforms.

[0058] From the above appendix Figure 1The example of the time-varying gain adaptive control method for multibeam echo sounders demonstrates several advantages. First, by acquiring a continuously varying absorption coefficient profile with depth, it overcomes the limitations of traditional methods that use fixed absorption coefficients. This allows the estimation of absorption loss to accurately reflect the current water absorption characteristics of the sea area. Integrating the absorption coefficient along the actual curved acoustic ray path avoids path length errors caused by treating acoustic rays as straight-line propagation, thus obtaining a more accurate beam signal absorption loss value. Second, pre-generating a two-dimensional lookup table indexed by beam incident angle and propagation time allows for offline completion of complex curved acoustic ray integration calculations, reducing the online computational burden and meeting the real-time processing requirements of unmanned platforms. Third, by quickly querying or interpolating absorption loss correction values ​​during actual echo sounding and applying them to time-varying gain compensation, a balance between high precision and high efficiency is achieved, effectively improving the quality of echo sounding data from multibeam echo sounders in different hydrological environments. In summary, the technical solution of this application can adaptively provide accurate time-varying gain compensation based on the actual water temperature and salinity distribution and acoustic ray bending effect, enhancing the environmental adaptability and measurement reliability of multibeam echo sounders.

[0059] Please see Figure 4 As shown, in one embodiment, a time-varying gain adaptive control device for a multibeam bathymetry sonar is provided. This device may include a first acquisition module 401, a first calculation module 402, a table creation module 403, a second acquisition module 404, and a second calculation module 405, as detailed below: The first acquisition module 401 is used to obtain the corresponding salinity profile based on the sound velocity profile data and environmental parameters at the current measurement location, and to obtain the temperature profile based on the empirical formula of sound velocity, and to calculate the absorption coefficient profile that changes continuously with depth. The first calculation module 402 is used to divide the water body into layers along the depth based on the absorption coefficient profile that changes continuously with depth and the sound ray tracking results. The sound velocity and absorption coefficient within the layer are regarded as changing at equal gradients. The absorption coefficient is integrated along the actual curved sound ray path of each layer, and the absorption loss of all layers is accumulated to obtain the accurate absorption loss of the beam signal caused by water absorption. The table creation module 403 is used to create a two-dimensional lookup table based on the absorption coefficient profile that changes continuously with depth and the integral model of the curved sound path. During the system initialization or route planning stage, the table is pre-indexed with the beam incident angle and the signal one-way propagation time to store the accurate absorption loss values ​​under the corresponding beam incident angle and one-way propagation time. The second acquisition module 404 is used to obtain the absorption loss correction value by querying or interpolating in the two-dimensional lookup table based on the current incident angle and one-way propagation time of the beam during the actual depth sounding process, and then applying it to the time-varying gain compensation of the echo signal. The second calculation module 405 is used to extrapolate the calculation based on the linear propagation model using the absorption coefficient corresponding to the last row node of the two-dimensional lookup table if the one-way propagation time of the current beam exceeds the maximum range of the two-dimensional lookup table.

[0060] From the above Figure 4 As can be seen from the example of the time-varying gain adaptive control device for multibeam echo sounders, on the one hand, by acquiring an absorption coefficient profile that continuously varies with depth, the limitations of traditional methods using fixed absorption coefficients are overcome, enabling the estimation of absorption loss to truly reflect the water absorption characteristics of the current sea area. Furthermore, integrating the absorption coefficient along the actual curved sound ray path avoids path length errors caused by treating the sound ray as a straight line, thus obtaining a more accurate beam signal absorption loss value. On the other hand, a two-dimensional lookup table indexed by beam incident angle and propagation time is pre-generated, allowing complex curved sound ray integration calculations to be completed offline, reducing the online computational burden and meeting the real-time processing requirements of unmanned platforms. Thirdly, by quickly querying or interpolating absorption loss correction values ​​during actual depth sounding and applying them to time-varying gain compensation, a balance between high precision and high efficiency is achieved, effectively improving the depth sounding data quality of multibeam echo sounders in different hydrological environments. In summary, the technical solution of this application can adaptively provide accurate time-varying gain compensation based on the actual water temperature and salinity distribution and sound ray bending effect, enhancing the environmental adaptability and measurement reliability of multibeam echo sounders.

[0061] In one embodiment, a computer device is provided, the internal structure of which can be shown as follows: Figure 5 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used for communication with external clients via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a time-varying gain adaptive control method for a multi-beam bathysound sonar.

[0062] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, performs the following steps: Step S1: Based on the sound velocity profile data and environmental parameters at the current measurement location, obtain the corresponding salinity profile, and invert the temperature profile based on the empirical formula for sound velocity to calculate the absorption coefficient profile that changes continuously with depth. Step S2: Based on the absorption coefficient profile, the water body is divided into layers along the depth by combining the sound ray tracking results. The sound velocity and absorption coefficient within the layer are regarded as equal gradient changes. The absorption coefficient is integrated along the actual curved sound ray path of each layer, and the absorption loss of all layers is accumulated to obtain the accurate absorption loss of the beam signal caused by water absorption. Step S3: Based on the absorption coefficient profile obtained in step S1 and the curved acoustic path integral model used in step S2, during the system initialization or route planning stage, a two-dimensional lookup table is established in advance using the beam incident angle and signal one-way propagation time as indexes to store the accurate absorption loss values ​​under the corresponding beam incident angle and one-way propagation time. Step S4: Based on the two-dimensional lookup table, during the actual depth sounding process, according to the current beam incident angle and one-way propagation time, the two-dimensional lookup table is queried or interpolated to obtain the absorption loss correction value and apply it to the time-varying gain compensation of the echo signal. Step S5: If the one-way propagation time of the current beam exceeds the maximum range of the two-dimensional lookup table, then the absorption coefficient corresponding to the last node of the two-dimensional lookup table is used to extrapolate the calculation according to the straight-line propagation model.

[0063] The aforementioned computer program adaptively provides accurate time-varying gain compensation, enhancing the environmental adaptability and measurement reliability of multibeam echo sounders.

[0064] In one embodiment, a computer-readable storage medium is provided that stores a computer program, which, when executed by a processor, performs the following steps: Step S1: Based on the sound velocity profile data and environmental parameters at the current measurement location, obtain the corresponding salinity profile, and invert the temperature profile based on the empirical formula for sound velocity to calculate the absorption coefficient profile that changes continuously with depth. Step S2: Based on the absorption coefficient profile, the water body is divided into layers along the depth by combining the sound ray tracking results. The sound velocity and absorption coefficient within the layer are regarded as equal gradient changes. The absorption coefficient is integrated along the actual curved sound ray path of each layer, and the absorption loss of all layers is accumulated to obtain the accurate absorption loss of the beam signal caused by water absorption. Step S3: Based on the absorption coefficient profile obtained in step S1 and the curved acoustic path integral model used in step S2, during the system initialization or route planning stage, a two-dimensional lookup table is established in advance using the beam incident angle and signal one-way propagation time as indexes to store the accurate absorption loss values ​​under the corresponding beam incident angle and one-way propagation time. Step S4: Based on the two-dimensional lookup table, during the actual depth sounding process, according to the current beam incident angle and one-way propagation time, the two-dimensional lookup table is queried or interpolated to obtain the absorption loss correction value and apply it to the time-varying gain compensation of the echo signal. Step S5: If the one-way propagation time of the current beam exceeds the maximum range of the two-dimensional lookup table, then the absorption coefficient corresponding to the last node of the two-dimensional lookup table is used to extrapolate the calculation according to the straight-line propagation model.

[0065] The steps described above, implemented by the computer program when executed by the processor, adaptively provide accurate time-varying gain compensation, enhancing the environmental adaptability and measurement reliability of multibeam echo sounders.

[0066] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server side and client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0067] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0068] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0069] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A time-varying gain adaptive control method for a multibeam echo sounder, characterized in that, The method includes: Based on the sound velocity profile data and environmental parameters at the current measurement location, the corresponding salinity profile is obtained, and the temperature profile is obtained by inversion based on the empirical formula for sound velocity. The absorption coefficient profile that changes continuously with depth is then calculated. Based on the absorption coefficient profile, the water body is layered along the depth by combining the sound ray tracking results. The sound velocity and absorption coefficient within each layer are regarded as equal gradient changes. The absorption coefficient is integrated along the actual curved sound ray path of each layer, and the absorption loss of all layers is accumulated to obtain the accurate absorption loss of the beam signal caused by water absorption. Based on the absorption coefficient profile and the curved sound path integral model, during the system initialization or route planning stage, a two-dimensional lookup table is pre-established using the beam incident angle and signal one-way propagation time as indexes to store the accurate absorption loss values ​​under the corresponding beam incident angle and one-way propagation time. Based on the two-dimensional lookup table, during the actual depth sounding process, the absorption loss correction value is obtained by querying or interpolating the two-dimensional lookup table according to the current beam's incident angle and one-way propagation time, and then applied to the time-varying gain compensation of the echo signal. If the one-way propagation time of the current beam exceeds the maximum range of the two-dimensional lookup table, then the absorption coefficient corresponding to the last node of the two-dimensional lookup table is used to extrapolate the calculation according to the straight-line propagation model.

2. The time-varying gain adaptive control method for multibeam echo sounder according to claim 1, characterized in that, When establishing the two-dimensional lookup table, the beam incidence angle step size and propagation time step size are adaptively determined based on the gradient distribution of the sound velocity profile at the current measurement location: the time step size is decreased within the propagation time range corresponding to the depth range with a large sound velocity gradient; the time step size is increased within the propagation time range corresponding to the depth range with a small sound velocity gradient; and the angle step size is decreased in the region where the beam incidence angle is close to the critical angle.

3. The time-varying gain adaptive control method for multibeam echo sounder according to claim 1, characterized in that, The method further includes: After acquiring new sound speed profile data, the measured sound speed profile is compared with the theoretical sound speed profile obtained by inversion through the global salinity model. If the deviation between the two at the same depth exceeds a preset threshold, the absorption coefficient profile and the accurate absorption loss value are recalculated based on the measured sound velocity profile, and the two-dimensional lookup table is updated accordingly.

4. The time-varying gain adaptive control method for multibeam echo sounder according to claim 3, characterized in that, When updating the two-dimensional lookup table, only the grid points in the beam incident angle interval and the one-way propagation time interval affected by the change in sound speed profile are recalculated, while the remaining grid points remain unchanged.

5. The time-varying gain adaptive control method for multibeam echo sounder according to claim 1, characterized in that, The method further includes: The absorption loss correction value is added to the extended loss correction value calculated based on the spherical extended model to generate a comprehensive time-varying gain value, which is then applied to the gain control of the echo signal.

6. The time-varying gain adaptive control method for multibeam echo sounder according to claim 5, characterized in that, The method further includes: performing a moving average filter on the integrated time-varying gain value to reduce gain jumps caused by lookup table discretization, and applying the filtered gain value to the gain control of the echo signal.

7. The time-varying gain adaptive control method for multibeam echo sounder according to claim 1, characterized in that, When interpolating in the two-dimensional lookup table, a bilinear interpolation method is used, and interpolation calculations are performed between four neighboring grid points in the two-dimensional lookup table based on the current beam incidence angle and one-way propagation time.

8. A time-varying gain adaptive control device for a multibeam echo sounder, characterized in that, The device includes: The first acquisition module is used to obtain the corresponding salinity profile based on the sound velocity profile data and environmental parameters at the current measurement location, and to obtain the temperature profile based on the empirical formula of sound velocity, and to calculate the absorption coefficient profile that changes continuously with depth. The first calculation module is used to divide the water body into layers along the depth based on the absorption coefficient profile and the sound ray tracking results, treat the sound velocity and absorption coefficient within the layer as equal gradient changes, integrate the absorption coefficient along the actual curved sound ray path of each layer, and accumulate the absorption loss of all layers to obtain the accurate absorption loss of the beam signal caused by water absorption. The table building module is used to build a two-dimensional lookup table based on the absorption coefficient profile and the curved sound path integral model, in advance during the system initialization or route planning stage, using the beam incident angle and signal one-way propagation time as indexes to store the accurate absorption loss values ​​under the corresponding beam incident angle and one-way propagation time. The second acquisition module is used to, based on the two-dimensional lookup table, in the actual depth sounding process, perform a query or interpolation in the two-dimensional lookup table according to the current beam's incident angle and one-way propagation time to obtain the absorption loss correction value and apply it to the time-varying gain compensation of the echo signal. The second calculation module is used to extrapolate the calculation using the absorption coefficient corresponding to the last row node of the two-dimensional lookup table according to the straight-line propagation model if the one-way propagation time of the current beam exceeds the maximum range of the two-dimensional lookup table.

9. A computer device, the computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the time-varying gain adaptive control method for the multibeam echo sounder as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the time-varying gain adaptive control method for the multibeam echo sounder as described in any one of claims 1 to 7.

Images