Adaptive robust submarine topography tracking method and system for multi-beam sounding

Abstract

The invention provides an adaptive robust submarine topography tracking method and system for multi-beam sounding, and the method comprises the steps: S1, predicting the depth of a next to-be-detected beam based on the historical successfully detected beam information, and setting a main search gate and an auxiliary search gate; s2, performing primary detection in the main search gate, if the primary detection succeeds, marking a current beam as a normal point, and executing a step S4, and if the primary detection fails, executing a step S3; s3, performing secondary detection in the auxiliary search gate, if the secondary detection succeeds, marking the current beam as a recovery point, and executing the step S4, and if the secondary detection fails, marking the current beam as an invalid point, keeping the current historical queue unchanged, and executing the step S4; and S4, outputting the depth value and the state mark of the current wave beam, judging whether a to-be-processed wave beam exists or not, and if so, repeating the steps S1 to S4. According to the method, by introducing a secondary failure protection detection and intelligent recovery mechanism, the violent fluctuation of the submarine topography can be effectively handled, and the optimal balance of efficiency and robustness is realized.

Description

Technical Field

[0001] This invention relates to the fields of underwater acoustic engineering and marine surveying technology, specifically to an adaptive robust seabed topography tracking method and system for multibeam bathymetry. Background Technology

[0002] Seafloor topographic mapping is fundamental to modern marine exploration, channel dredging, subsea pipeline laying, underwater target detection, and marine scientific research. Acoustic bathymetry, especially multibeam sonar systems, is the primary means of acquiring high-precision, high-efficiency seafloor topographic data. Its core principle is to calculate the water depth at the measured point by calculating the time it takes for sound waves to travel from the transducer to the seabed and back, combined with the sound velocity profile.

[0003] Among numerous bathymetry signal processing algorithms, amplitude-based detection methods, especially the Weighted Mean Time (WMT) method, have been widely used due to their simple principle, moderate computational cost, and strong noise resistance. The core of the WMT method lies in using the envelope energy of the echo signal as a weight to calculate the weighted average arrival time of the echo. This time point is considered to be the most likely to represent the true location of the seabed, making it more stable than simple frontal detection or peak detection methods.

[0004] To improve data processing efficiency, the industry widely adopts the "tracking gate" technique. This technique is based on the reasonable assumption that "seabed topography is continuous between adjacent measurement points." When processing continuous beams or pings, the system uses the depth value of the previously successfully detected point to pre-set a finite-width depth search window for subsequent detection points. This method concentrates computational resources on the area most likely to produce echoes, greatly reducing the invalid search range and significantly improving data processing speed, enabling it to meet the requirements of real-time mapping.

[0005] However, existing WMT bathymetry techniques based on simple tracking gates exhibit significant limitations when faced with complex and varied seabed topography, mainly in the following aspects: (1) Tracking loss in steep slopes and cliffs: In areas with abrupt changes in topography, such as seabed ridges, steep slopes, trenches, or cliffs, the depth difference between adjacent beams may far exceed the width of the preset tracking gate. When the depth value of the previous beam cannot effectively predict the depth of the next beam, the actual echo signal will fall outside the tracking window. The system cannot detect a valid signal within the preset narrow window, thus determining it as a detection failure, resulting in voids or breaks in the topographic data at this point, which seriously affects the integrity and accuracy of the seabed topography.

[0006] (2) The contradiction between efficiency and robustness: To address the aforementioned "tracking loss" problem, a conservative strategy is to set an extremely wide search threshold to cover any possible terrain changes. However, this detection method includes a large amount of invalid water data, significantly increasing the computational burden, reducing processing efficiency, and defeating the original purpose of the tracking gate technique. Conversely, if a narrow tracking gate is used to maintain high efficiency, it will frequently fail in complex terrain areas, sacrificing the robustness and reliability of the measurement. Existing technologies struggle to achieve a good balance between the two.

[0007] (3) Lack of effective failure recovery mechanism: Most traditional algorithms have a passive and simplistic approach to handling "tracking loss". For example, some algorithms simply mark the point as invalid and continue to track using earlier historical data, which may cause the error to propagate in subsequent beams. Some algorithms may attempt to search again over a larger area, but they lack intelligent guidance strategies and often blindly search the entire water map, resulting in low success rate and high computational cost. More importantly, once recovery is successful, the lack of a "reset" mechanism for the tracking state makes it easy to continue using incorrect terrain trends, leading to subsequent tracking failures.

[0008] (4) Over-reliance on prior knowledge: Some improved algorithms attempt to dynamically adjust the tracking gate size by introducing a pre-set maximum slope value or relying on historical nautical chart data with low accuracy. However, the actual seabed topography is highly variable, and the preset parameters are difficult to apply universally. They lack the ability to adaptively adjust based on real-time detection results and perform poorly in unknown sea areas or areas with abnormal topography.

[0009] The patent with publication number CN106526601A discloses a multibeam bathymetry signal processing method and device. This patent uses Kalman filtering for terrain tracking, but does not mention the specific algorithms used by the primary and secondary detection units. The automation level of this method is incomplete, and it lacks quality assessment and graded utilization of the detection results.

[0010] In summary, a prominent contradiction exists in existing technologies: while simple tracking gate techniques are highly efficient, they suffer from poor reliability in areas with complex topography; and while wide-threshold or full-body map searches can guarantee a certain detection rate, they come at the cost of efficiency. Therefore, there is an urgent need in this field for a novel bathymetry algorithm that can significantly improve the detection success rate and robustness in complex seabed topography while maintaining high processing speed, possessing strong "topography tracking" and "failure self-recovery" capabilities. Summary of the Invention

[0011] To address the shortcomings of existing technologies, the purpose of this invention is to provide an adaptive robust seabed topography tracking method and system for multibeam bathymetry.

[0012] An adaptive robust seabed topography tracking method for multibeam bathymetry, provided by the present invention, includes the following steps: Step S1: Based on the beam information of historically successfully detected beams, predict the depth of the next beam to be detected, and set a main search gate and a secondary search gate, wherein the range of the secondary search gate is much wider than that of the main search gate; Step S2: Perform a first-level detection within the main search gate. If the first-level detection is successful, mark the current beam as a normal point and update the coordinates and depth information of the normal point to the historical queue. Then proceed to step S4. If the first-level detection fails, proceed to step S3. Step S3: Perform secondary detection within the secondary search gate. If the secondary detection is successful, mark the current beam as a recovery point, reset the historical queue, and add the recovery point as initial data. Then proceed to step S4. If the secondary detection fails, mark the current beam as an invalid point, keep the current historical queue unchanged, and proceed to step S4. Step S4: Output the depth value and status flag of the current beam, and determine whether there are still beams to be processed. If so, repeat steps S1 to S4 until all beams have been processed.

[0013] Preferably, based on historically successfully detected beam information, the depth of the next beam to be detected is predicted, and a main search gate and a secondary search gate are set, including: maintaining a list of the most recently detected beams. A queue of successfully detected beam depth values ​​and their corresponding lateral coordinates, where the queue length is... Dynamic adjustments are made based on terrain complexity; a polynomial fitting algorithm is used to fit the depth-location relationship in the historical data queue, and the polynomial coefficients obtained from the fitting are used as the basis for the results. Calculate the position of the next beam. Predicted depth value :

[0014] To predict depth values Set up a main search gate at the center. ,in,

[0015]

[0016] Among them, the threshold half-width It is a configurable parameter whose value is much smaller than the total water depth; Set up a secondary search gate ,in, It is the minimum detection depth. That is the maximum detection depth.

[0017] Preferably, a first-level detection is performed within the main search gate. If the first-level detection is successful, the current beam is marked as a normal point, and the coordinates and depth information of the normal point are updated to the historical queue. Step S4 is then executed. If the first-level detection fails, step S3 is executed, including: within the main search gate... Within this time window, a weighted average time algorithm is used for depth detection. The energy-weighted average time of the echo signal within that time window is calculated and converted into a depth value. Simultaneously, the signal-to-noise ratio of the detection result is calculated. If the depth value lie in And the signal-to-noise ratio Higher than the preset first threshold If the detection is successful, the first-level detection is successful; otherwise, the first-level detection fails. The beams that are successfully detected are marked as normal points, and the coordinates and depth information of the normal points are added to the tail of the historical queue. If the queue is full, the data at the head of the queue is removed synchronously.

[0018] Preferably, secondary detection is performed within the secondary search gate. If the secondary detection is successful, the current beam is marked as a recovery point, the historical queue is reset, and the recovery point is added as initial data. Step S4 is then executed. If the secondary detection fails, the current beam is marked as an invalid point, the current historical queue remains unchanged, and step S4 is executed, including: performing secondary detection on the beams that failed detection, wherein the secondary detection is performed within the secondary search gate. The weighted average time algorithm is applied to perform a full-range search to obtain the depth value. and signal-to-noise ratio If the depth value lie in And the signal-to-noise ratio Higher than the preset second threshold If the detection is successful, the secondary detection is successful; otherwise, the secondary detection fails. For beams that are successfully detected, mark them as recovery points, clear all data in the historical queue, and add the recovery points as initial data. For beams that fail to be detected, mark them as invalid points, stop depth estimation for those points, keep the current historical data queue unchanged, and continue processing the next beam.

[0019] Preferably, in the main search gate Within this process, a weighted average time algorithm is used for depth detection, including: extracting data from the time series of the echo signal. Signal segment within the interval ,Pick The top 10% are recorded as ,in, , Speed ​​of sound; Calculate the signal segment envelope Set dynamic threshold Determine that the envelope exceeds continuous region The largest continuous region is denoted as In the largest continuous region Within, the weighted average time is calculated using the square of the envelope as the weight. :

[0020] Calculate depth And the signal-to-noise ratio of the detection result. .

[0021] This invention also provides an adaptive robust seabed topography tracking system for multibeam bathymetry. This system can be implemented by executing the steps of the adaptive robust seabed topography tracking method for multibeam bathymetry. That is, those skilled in the art can understand the adaptive robust seabed topography tracking method for multibeam bathymetry as a preferred embodiment of the adaptive robust seabed topography tracking system for multibeam bathymetry. The system includes: The prediction module predicts the depth of the next beam to be detected based on the beam information of historically successfully detected beams, and sets a main search gate and a secondary search gate, wherein the range of the secondary search gate is much wider than that of the main search gate; The first-level detection module performs first-level detection within the main search gate. If the first-level detection is successful, the current beam is marked as a normal point, and the coordinates and depth information of the normal point are updated to the historical queue, triggering the control module to work. If the first-level detection fails, the second-level detection module is triggered to work. The secondary detection module performs secondary detection within the secondary search gate. If the secondary detection is successful, the current beam is marked as a recovery point, the historical queue is reset, and the recovery point is added as initial data to it, triggering the control module to work. If the secondary detection fails, the current beam is marked as an invalid point, the current historical queue remains unchanged, and the control module is triggered to work. The control module outputs the current beam's depth value and status flag, determines whether there are still beams to be processed, and if so, triggers the prediction module, the first-level control module, the second-level control module, and the control module to work in sequence until all beams have been processed.

[0022] Compared with the prior art, the present invention has the following beneficial effects: (1) By introducing a secondary failure protection detection and intelligent recovery mechanism, the present invention can effectively cope with the dramatic fluctuations of the seabed topography. Even if the tracking gate is temporarily lost, it can recapture the real seabed signal through full-range search and quickly correct the tracking trajectory through historical reset, which greatly reduces the data loss rate under complex terrain.

[0023] (2) In flat or gently changing terrain areas, the algorithm runs in the prediction-based narrow gate main detection mode most of the time, with low computational cost and maintains efficiency similar to traditional tracking gate technology. Only when necessary (at abrupt terrain changes) will the computationally intensive secondary detection be started, achieving the best balance between efficiency and robustness.

[0024] (3) The algorithm does not rely on fixed prior terrain parameters. The prediction model and historical queue length can be dynamically adjusted according to the real-time detection results, and can automatically adapt to unknown sea areas and complex and ever-changing seabed environments.

[0025] (4) The unique “recovery point” historical reset mechanism ensures that after successful recovery, the erroneous prediction trend will not be continued, effectively curbing the chain reaction caused by local failure and ensuring the accuracy of subsequent detection. Attached Figure Description

[0026] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 The flowchart illustrates an adaptive robust seabed topography tracking method for multibeam bathymetry provided by this invention.

[0027] Figure 2 This is a graph showing the data processing results of Example 1.

[0028] Figure 3 This is a graph comparing the detection success rates of algorithms A and B in Example 2.

[0029] Figure 4 The image shows example images of the detection results of algorithms A and B in Example 2. Detailed Implementation

[0030] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0031] Figure 1 The flowchart of an adaptive robust seabed topography tracking method for multibeam bathymetry according to the present invention is shown in the figure, including the following steps: Step 1, System initialization, set the following algorithm parameters: Minimum detection depth; Maximum detection depth; Threshold half width; : Maximum length of the historical queue; : Main detection signal-to-noise ratio threshold; The signal-to-noise ratio threshold for secondary detection is slightly lower than... ; Initialize an empty historical data queue to store the location and depth information of successfully detected points. .

[0032] Step 2: Obtain the first valid detection point.

[0033] Step 2.1, for the first beam or the initial beam after tracking interruption, a conservative strategy is adopted: Full-body WMT testing was conducted within the area to obtain reliable initial depth. .

[0034] Step 2.2, take the detection point data obtained in Step 2.1. Add to historical data queue .

[0035] Step 3: Predict the depth of the next beam.

[0036] Step 3.1, if If the number of valid data points is greater than or equal to 2, then trend prediction is performed. Specifically, from... Take the most recent one from The horizontal coordinate vector of each point and depth vector First-order linear fitting was performed using the least squares method to obtain the polynomial coefficients. Calculate the position of the next beam Prediction depth: .

[0037] Step 3.2, if If there are not enough midpoints, use the last depth value in the queue as the midpoint. .

[0038] Step 4: Set the main search gate (narrow gate).

[0039] Calculate the main search gate boundary and convert the depth gate into a time gate. ,in , The speed of sound.

[0040]

[0041]

[0042] It is important to note the threshold width. It can be dynamically adjusted according to factors such as terrain complexity and signal-to-noise ratio, and its value is much smaller than the total water depth.

[0043] Step 5: Perform WMT detection (Level 1 detection) inside the main search gate.

[0044] Step 5.1: Extract from the time series of the echo signal. Signal segment within the interval ,Pick The top 10% are recorded as ; Step 5.2, calculate the signal segment. ,Pick The top 10% are recorded as ; Step 5.3, Set dynamic threshold Determine that the envelope exceeds continuous region The largest continuous region is denoted as ; Step 5.4, in the largest continuous region Within, the weighted average time is calculated using the square of the envelope as the weight. :

[0045] Step 5.5, Calculate depth And the signal-to-noise ratio of the detection result. .

[0046] Step 6: Determine if the main detection was successful.

[0047] judge exist Inside and Is the condition met? If it is met, proceed to step 10; if it is not met, proceed to step 7.

[0048] Step 7: Perform WMT (secondary inspection) inside the wide gate.

[0049] Step 7.1, set the wide search gate to And convert it into the corresponding time gate. .

[0050] Step 7.2, at this wide time gate Within this process, repeat the WMT detection procedure from step 5 to obtain the depth value. and signal-to-noise ratio .

[0051] Step 8: Determine whether the secondary detection was successful.

[0052] judge exist Inside and Is the condition met? If met, proceed to step 9; if not, proceed to step 11.

[0053] Step 9: Mark as a recovery point and reset the history queue.

[0054] Step 9.1: Detect the current beam. Set as Mark the beam as .

[0055] Step 9.2, Clear , will the current point Add it to the queue as a new starting point and proceed to step 12.

[0056] Through the above steps, the unique "recovery point" historical reset mechanism ensures that erroneous prediction trends will not be continued after successful recovery, effectively curbing the chain reaction caused by local failures and guaranteeing the accuracy of subsequent detection.

[0057] Step 10: Mark as normal point and update the history queue.

[0058] Step 10.1: Detect the current beam. Set as Mark the beam as .

[0059] Step 10.2, set the current point join in If the queue length exceeds If so, remove the oldest point and proceed to step 12.

[0060] Step 11: Mark as invalid point.

[0061] Invalidate the depth result of the current beam and mark the beam as... ,Keep constant.

[0062] Step 12: Output the results and process the next beam.

[0063] Output the current beam's depth value, status flag, and relevant quality information. Determine if there are any subsequent beams to process. If so, return to step 3; otherwise, end the process.

[0064] This invention also provides an adaptive robust seabed topography tracking system for multibeam bathymetry. This system can be implemented by executing the steps of the adaptive robust seabed topography tracking method for multibeam bathymetry. That is, those skilled in the art can understand the adaptive robust seabed topography tracking method for multibeam bathymetry as a preferred embodiment of the adaptive robust seabed topography tracking system for multibeam bathymetry. The system includes: The prediction module predicts the depth of the next beam to be detected based on the beam information of historically successfully detected beams, and sets a main search gate and a secondary search gate, wherein the range of the secondary search gate is much wider than that of the main search gate; The first-level detection module performs first-level detection within the main search gate. If the first-level detection is successful, the current beam is marked as a normal point, and the coordinates and depth information of the normal point are updated to the historical queue, triggering the control module to work. If the first-level detection fails, the second-level detection module is triggered to work. The secondary detection module performs secondary detection within the secondary search gate. If the secondary detection is successful, the current beam is marked as a recovery point, the historical queue is reset, and the recovery point is added as initial data to it, triggering the control module to work. If the secondary detection fails, the current beam is marked as an invalid point, the current historical queue remains unchanged, and the control module is triggered to work. The control module outputs the current beam's depth value and status flag, determines whether there are still beams to be processed, and if so, triggers the prediction module, the first-level control module, the second-level control module, and the control module to work in sequence until all beams have been processed.

[0065] To verify the effectiveness of the algorithm of this invention, two sets of implementation tests were conducted on simulation data.

[0066] Example 1: Single ping test.

[0067] A complex profile including gentle slopes, platforms, steep cliffs, and deep ravines is simulated, and the algorithm of this invention (Algorithm B) is compared with the traditional fixed-width tracking gate algorithm (Algorithm A). The data processing results are as follows: Figure 2 As shown, the results indicate that Algorithm A suffers from significant data loss at steep cliffs and deep ravines. Algorithm B, on the other hand, successfully restored the detection function at steep cliffs through a two-stage detection process, with only a few invalid points at the abrupt terrain changes themselves, thus achieving overall continuity in the detection.

[0068] Example 2: Multi-ping test.

[0069] Simulations were performed on complex terrain profiles, comparing the traditional fixed-width tracking gate algorithm (Algorithm A) with the algorithm of this invention (Algorithm B). The data processing results are as follows: Figures 3-4 As shown, the results indicate that the traditional algorithm has poor detection robustness and a low success rate in certain ping tests. By adopting the algorithm of this invention, both detection robustness and success rate are significantly improved.

[0070] Based on the combined results of Examples 1 and 2, it can be concluded that the detection performance and success rate of the algorithm of the present invention are significantly improved compared with the traditional fixed-width tracking gate algorithm.

[0071] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0072] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. An adaptive robust seabed topography tracking method for multibeam bathymetry, characterized in that, include: Step S1: Based on the beam information of historically successfully detected beams, predict the depth of the next beam to be detected, and set a main search gate and a secondary search gate, wherein the range of the secondary search gate is much wider than that of the main search gate; Step S2: Perform a first-level detection within the main search gate. If the first-level detection is successful, mark the current beam as a normal point and update the coordinates and depth information of the normal point to the historical queue. Then proceed to step S4. If the first-level detection fails, proceed to step S3. Step S3: Perform secondary detection within the secondary search gate. If the secondary detection is successful, mark the current beam as a recovery point, reset the historical queue, and add the recovery point as initial data. Then proceed to step S4. If the secondary detection fails, mark the current beam as an invalid point, keep the current historical queue unchanged, and proceed to step S4. Step S4: Output the depth value and status flag of the current beam, and determine whether there are still beams to be processed. If so, repeat steps S1 to S4 until all beams have been processed.

2. The adaptive robust seabed topography tracking method for multibeam bathymetry according to claim 1, characterized in that, The process of predicting the depth of the next beam to be detected based on historically successfully detected beam information and setting a main search gate and a secondary search gate includes: Maintain a list containing the most recent A queue of successfully detected beam depth values ​​and their corresponding lateral coordinates, where the queue length is... Dynamically adjusted based on terrain complexity; A polynomial fitting algorithm is used to fit the depth-position relationship in the historical data queue, and the polynomial coefficients obtained from the fitting are used to... Calculate the position of the next beam. Predicted depth value : To predict depth values Set up a main search gate at the center. ,in, Among them, the threshold half-width It is a configurable parameter whose value is much smaller than the total water depth; Set up a secondary search gate ,in, It is the minimum detection depth. That is the maximum detection depth.

3. The adaptive robust seabed topography tracking method for multibeam bathymetry according to claim 1, characterized in that, The first-level detection is performed within the main search gate. If the first-level detection is successful, the current beam is marked as a normal point, and the coordinates and depth information of the normal point are updated to the historical queue. Step S4 is then executed. If the first-level detection fails, step S3 is executed, including: The main search gate Within this time window, a weighted average time algorithm is used for depth detection. The energy-weighted average time of the echo signal within that time window is calculated and converted into a depth value. Simultaneously, the signal-to-noise ratio of the detection result is calculated. ; If the depth value lie in And the signal-to-noise ratio Higher than the preset first threshold If the result is positive, the first-level detection is successful; otherwise, the first-level detection fails. The beams that are successfully detected are marked as normal points, and the coordinates and depth information of the normal points are added to the tail of the historical queue. If the queue is full, the data at the head of the queue is removed simultaneously.

4. The adaptive robust seabed topography tracking method for multibeam bathymetry according to claim 1, characterized in that, The process involves performing a secondary detection within the secondary search gate. If the secondary detection is successful, the current beam is marked as a recovery point, the historical queue is reset, and the recovery point is added as initial data. Step S4 is then executed. If the secondary detection fails, the current beam is marked as an invalid point, the current historical queue remains unchanged, and step S4 is executed, including: For beams that fail detection, a secondary detection is performed, wherein the secondary detection is performed at the sub-search gate. The weighted average time algorithm is applied to perform a full-range search to obtain the depth value. and signal-to-noise ratio ; If the depth value lie in And the signal-to-noise ratio Higher than the preset second threshold If the result is positive, the secondary detection is successful; otherwise, the secondary detection fails. The successfully detected beam is marked as a recovery point, all data in the historical queue is cleared, and the recovery point is added to it as initial data; If a beam fails to detect a point, mark it as invalid and stop depth estimation for that point. Keep the current historical data queue unchanged and continue processing the next beam.

5. The adaptive robust seabed topography tracking method for multibeam bathymetry according to claim 3, characterized in that, The main search gate Within this area, a weighted average time algorithm is used for depth detection, including: Extracting from the time series of the echo signal Signal segment within the interval ,Pick The top 10% are recorded as ,in, , Speed ​​of sound; Calculate the signal segment envelope ; Set dynamic threshold Determine the envelope exceeds continuous region The largest continuous region is denoted as ; In the largest continuous region Within, the weighted average time is calculated using the square of the envelope as the weight. : Calculate depth and the signal-to-noise ratio of the detection result. .

6. An adaptive robust seabed topography tracking system for multibeam bathymetry, characterized in that, include: The prediction module predicts the depth of the next beam to be detected based on the beam information of historically successfully detected beams, and sets a main search gate and a secondary search gate, wherein the range of the secondary search gate is much wider than that of the main search gate; The first-level detection module performs first-level detection within the main search gate. If the first-level detection is successful, the current beam is marked as a normal point, and the coordinates and depth information of the normal point are updated to the historical queue, triggering the control module to work. If the first-level detection fails, the second-level detection module is triggered to work. The secondary detection module performs secondary detection within the secondary search gate. If the secondary detection is successful, the current beam is marked as a recovery point, the historical queue is reset, and the recovery point is added as initial data to it, triggering the control module to work. If the secondary detection fails, the current beam is marked as an invalid point, the current historical queue remains unchanged, and the control module is triggered to work. The control module outputs the current beam's depth value and status flag, determines whether there are still beams to be processed, and if so, triggers the prediction module, first-level detection module, second-level detection module, and control module to work in sequence until all beams have been processed.

7. An adaptive robust seabed topography tracking system for multibeam bathymetry according to claim 6, characterized in that, The process of predicting the depth of the next beam to be detected based on historically successfully detected beam information and setting a main search gate and a secondary search gate includes: Maintain a list containing the most recent A queue of successfully detected beam depth values ​​and their corresponding lateral coordinates, where the queue length is... Dynamically adjusted based on terrain complexity; A polynomial fitting algorithm is used to fit the depth-position relationship in the historical data queue, and the polynomial coefficients obtained from the fitting are used to... Calculate the position of the next beam. Predicted depth value : To predict depth values Set up a main search gate at the center. ,in, Among them, the threshold half-width It is a configurable parameter whose value is much smaller than the total water depth; Set up a secondary search gate ,in, It is the minimum detection depth. That is the maximum detection depth.

8. An adaptive robust seabed topography tracking system for multibeam bathymetry according to claim 6, characterized in that, The process involves performing a first-level detection within the main search gate. If the first-level detection is successful, the current beam is marked as a normal point, and the coordinates and depth information of the normal point are updated to the historical queue, triggering the control module to operate. If the first-level detection fails, the second-level detection module is triggered, including: The main search gate Within this time window, a weighted average time algorithm is used for depth detection. The energy-weighted average time of the echo signal within that time window is calculated and converted into a depth value. Simultaneously, the signal-to-noise ratio of the detection result is calculated. ; If the depth value lie in And the signal-to-noise ratio Higher than the preset first threshold If the result is positive, the first-level detection is successful; otherwise, the first-level detection fails. The beams that are successfully detected are marked as normal points, and the coordinates and depth information of the normal points are added to the tail of the historical queue. If the queue is full, the data at the head of the queue is removed simultaneously.

9. An adaptive robust seabed topography tracking system for multibeam bathymetry according to claim 6, characterized in that, The process involves performing a secondary detection within the secondary search gate. If the secondary detection is successful, the current beam is marked as a recovery point, the historical queue is reset, and the recovery point is added as initial data, triggering the control module to operate. If the secondary detection fails, the current beam is marked as an invalid point, the current historical queue remains unchanged, and the control module is triggered to operate, including: For beams that fail detection, a secondary detection is performed, wherein the secondary detection is performed at the sub-search gate. The weighted average time algorithm is applied to perform a full-range search to obtain the depth value. and signal-to-noise ratio ; If the depth value lie in And the signal-to-noise ratio Higher than the preset second threshold If the result is positive, the secondary detection is successful; otherwise, the secondary detection fails. The successfully detected beam is marked as a recovery point, all data in the historical queue is cleared, and the recovery point is added to it as initial data; If a beam fails to detect a point, mark it as invalid and stop depth estimation for that point. Keep the current historical data queue unchanged and continue processing the next beam.

10. An adaptive robust seabed topography tracking system for multibeam bathymetry according to claim 8, characterized in that, The main search gate Within this area, a weighted average time algorithm is used for depth detection, including: Extracting from the time series of the echo signal Signal segment within the interval ,Pick The top 10% are recorded as ,in, , Speed ​​of sound; Calculate the signal segment envelope ; Set dynamic threshold Determine the envelope exceeds continuous region The largest continuous region is denoted as ; In the largest continuous region Within, the weighted average time is calculated using the square of the envelope as the weight. : Calculate depth And the signal-to-noise ratio of the detection result. .

Images