Underwater Propeller Control Method and System Based on Water Environment Analysis

Abstract

The present invention discloses a method and system for controlling an underwater thruster based on water environment analysis. The method includes: collecting image data and laser point cloud data of the target water body to construct a three-dimensional environmental model; obtaining the initial position information and target position of the underwater thruster, and generating a working path based on environmental model analysis; collecting Doppler sonar data along the path, analyzing water flow information, and generating correction control instructions; and analyzing the quality of instruction transmission to determine the optimal transmission scheme. By integrating multi-source environmental perception data, the present invention achieves dynamic modeling and analysis of complex water environments, improving the path planning accuracy and autonomous control capabilities of the underwater thruster, and is suitable for precise navigation and dynamic obstacle avoidance scenarios in complex underwater operation tasks.

Description

Technical Field

[0001] The present invention relates to the field of intelligent underwater navigation and control technology, and in particular to an underwater thruster control method and system based on water environment analysis. Background Art

[0002] With the increasing development of marine resources and underwater exploration missions, underwater thrusters, as key actuators for underwater vehicles, have been widely used in fields such as ocean mapping, environmental monitoring, and underwater search and rescue. Traditional underwater thruster control methods, most of which rely on preset paths or simplified water models for path planning and control, are unable to cope with the complex and changing underwater environment. Especially when environmental information is incomplete or water flow is highly disturbed, path deviation and control failure are prone to occur, seriously affecting the efficiency and safety of mission execution.

[0003] Existing control methods, such as sonar or inertial navigation systems, assist with path adjustment. However, these methods often lack the ability to model and analyze the water environment in three dimensions in real time, making it difficult to accurately identify obstacles, terrain changes, or abnormal water flow along the path. Furthermore, due to limited underwater communication conditions, the transmission quality of correction commands cannot be guaranteed, resulting in delayed or even interrupted control responses, further reducing the system's stability and intelligence.

[0004] In addition, existing underwater propulsion control systems are mostly based on single sensor data, lack the fusion processing of multi-source information such as images and point clouds, and cannot achieve comprehensive perception and modeling of environmental characteristics; at the same time, the control strategy is relatively rough in terms of water flow interference identification and adaptive control, and lacks in-depth analysis and processing mechanisms for water flow direction, speed and its time series changes, which limits the system's application capabilities in complex water environments.

[0005] Therefore, there is an urgent need for an underwater thruster control method that can integrate multi-source perception data, has real-time environmental modeling capabilities, and can perform intelligent correction control based on dynamic water flow information. At the same time, it can dynamically adjust the control instruction transmission scheme in combination with the communication link status to improve the intelligence level, robustness and task execution efficiency of the control system. Summary of the Invention

[0006] In order to solve at least one of the above technical problems, the present invention proposes an underwater thruster control method and system based on water environment analysis.

[0007] A first aspect of the present invention provides an underwater thruster control method based on water environment analysis, comprising:

[0008] Acquire image data and laser point cloud data of the target water environment, and construct a three-dimensional environment model of the target water environment based on the image data and laser point cloud data;

[0009] Acquiring initial position information and target position information of the underwater propeller, performing a water environment analysis between the initial position and the target position based on the three-dimensional environment model, and determining a working path of the underwater propeller;

[0010] Acquiring Doppler sonar data of the working path, determining water flow information of the working path according to the Doppler sonar data, and determining a deviation correction control instruction of the underwater propeller according to the water flow information;

[0011] Performing a transmission quality analysis of the deviation correction control instruction on the working path, and determining a transmission scheme for the deviation correction control instruction according to the transmission quality.

[0012] In this solution, the image data and laser point cloud data of the target water environment are obtained, and a three-dimensional environment model of the target water environment is constructed based on the image data and laser point cloud data, specifically:

[0013] Continuous frame image data of the target water environment is collected by a binocular vision camera, and laser point cloud data of the target water environment is obtained synchronously based on a laser scanning device. When the time stamp difference between the image data and the laser point cloud data exceeds a preset time synchronization threshold, the image data and the laser point cloud data are time-aligned using a nearest neighbor matching method based on the timestamp;

[0014] Extracting image feature points of the image data based on a feature point detection algorithm, and performing curvature analysis on the laser point cloud data to determine laser feature points of the laser point cloud data;

[0015] Perform spatial registration on the time-aligned image data and the laser point cloud data, and calculate the matching degree of the spatial projection of the image feature points in the laser point cloud data. When the matching degree of the two in the overlapping area is lower than the preset matching threshold, the area is marked as a low-confidence area, and the image feature points in the low-confidence area are projected to the corresponding laser feature points until the matching degree of the image feature points in the spatial projection of the laser point cloud data is no lower than the preset matching threshold.

[0016] 3D mesh reconstruction is performed on areas where the matching degree meets the standards. Feature points of obstacle outlines and terrain elevation in the water environment are extracted. When the number of consecutive feature points is lower than the minimum number of topological connections required for path planning, the area is marked as a feature point sparse area. The feature point sparse area is scanned again by laser, and 3D features are supplemented based on the second scan.

[0017] Finally, a three-dimensional environment model including the spatial coordinates of obstacles and terrain gradients is generated.

[0018] In this solution, the initial position information and the target position information of the underwater propeller are obtained, and the water environment between the initial position and the target position is analyzed based on the three-dimensional environment model to determine the working path of the underwater propeller. Specifically,

[0019] Based on the spatial coordinates of obstacles and terrain elevation feature points in the three-dimensional environment model, an initial straight line path is generated between the initial position and the target position using the A* algorithm, the initial straight line path is projected into the three-dimensional environment model, and a set of obstacle coordinates and a set of terrain elevation points that the path projection passes through are extracted;

[0020] Calculating the intersection area of the initial straight path projection and the obstacle coordinate set; when the obstacle density in the intersection area exceeds a preset density threshold, marking the area as a high-risk area; calculating the gap width between adjacent obstacles in the high-risk area; when the gap width is less than the minimum pass width of the underwater propeller, upgrading the high-risk area to an impassable area, and recording the boundary coordinates of the impassable area;

[0021] Based on the boundary coordinates of the impassable area, the initial straight path is divided into several segments to be bypassed, with the intersection of the initial straight path and the impassable area as a segmentation point. For each segment to be bypassed, a ring search space is generated by expanding a preset safety distance outward along the boundary of the impassable area. Within the ring search space, a Dijkstra algorithm is used to calculate several local paths that bypass the impassable area. The local paths are connected to the undivided endpoints of the initial straight path to generate multiple candidate feasible paths.

[0022] The total length, number of turns, and number of obstacle avoidances of each candidate feasible path are calculated, a path score is performed based on the total length, number of turns, and number of obstacle avoidances, a path is selected based on the path score, and an operating path of the underwater thruster is determined.

[0023] In this solution, the acquiring of Doppler sonar data of the working path and the determining of water flow information of the working path according to the Doppler sonar data are specifically as follows:

[0024] The Doppler sonar array carried by the underwater propeller transmits sound wave signals along the extension direction of the working path and receives reflected signals to obtain Doppler sonar data;

[0025] Calculating the signal-to-noise ratio of the Doppler sonar data, marking areas where the signal-to-noise ratio is lower than a preset threshold as data interference areas, obtaining water turbidity information in the data interference areas, and adjusting the acoustic wave emission energy of the multispectral sonar array according to the water turbidity information until the signal-to-noise ratio of the Doppler sonar data is no lower than the preset threshold;

[0026] Calculating the sound wave frequency offset of Doppler sonar data, and determining real-time water flow information in the axial direction of the working path according to the sound wave frequency offset, wherein the water flow information includes water flow velocity and water flow direction;

[0027] The water flow information is used to construct a water flow information distribution map of the working path with time series changes, and the flow velocity change gradient in the continuous time window is extracted based on the water flow information distribution map. When the flow velocity mutation amplitude of the adjacent time window exceeds the preset mutation threshold, the current time period is marked as the abnormal water flow information acquisition period;

[0028] The Doppler sonar data during the abnormal period of water flow information acquisition are resampled, and the high-frequency noise interference is eliminated through sliding average filtering. The water flow information distribution map is corrected based on the resampled Doppler sonar data.

[0029] In this solution, the correction control instruction of the underwater propeller is determined according to the water flow information, specifically:

[0030] Based on the water flow information distribution map of the working path, extract the real-time water flow speed and direction at the current position of the underwater propeller, calculate the heading deviation angle between the real-time water flow direction and the working path, and when the heading deviation angle exceeds a preset angle threshold, calibrate the current water flow as a strong interference water flow, and generate a lateral deviation rate based on the product of the heading deviation angle and the real-time water flow speed;

[0031] determining a lateral deviation risk level based on a ratio of the lateral deviation rate to a maximum lateral deviation resistance capability of the underwater propulsor, and generating a first corrective torque based on a reverse vector component of the lateral deviation rate when the lateral deviation risk level exceeds a preset risk threshold;

[0032] The axial component of the real-time water flow velocity along the working path is simultaneously calculated. When the composite speed of the axial velocity component and the current propulsion speed of the propeller is lower than a preset minimum speed threshold, the current water flow is calibrated as a reverse strong resistance water flow. The output power of the main propulsion motor of the propeller is increased according to the speed amplitude of the reverse strong resistance water flow to generate a second corrective thrust.

[0033] Performing vector superposition of the first correcting torque and the second correcting thrust to generate an initial correcting control instruction;

[0034] Acquire real-time attitude sensor data from the underwater thruster, calculate the predicted heading angle and predicted position coordinate changes of the underwater thruster after the initial correction control command is applied, and when the predicted heading angle does not match the minimum curvature radius of the remaining path segment of the working path, adjust the weight coefficient of the first correction torque according to the curvature change gradient of the remaining path segment until the predicted heading angle meets the path tracking accuracy requirements;

[0035] When the lateral deviation between the predicted position coordinates and the working path continues to increase, a third compensation torque is generated based on the integral term of the lateral deviation, and the third compensation torque is superimposed on the initial correction control instruction to form a final correction control instruction.

[0036] In this solution, the transmission quality analysis of the correction control instruction on the working path is performed, and a transmission scheme of the correction control instruction is determined according to the transmission quality, specifically:

[0037] The communication link parameters between the underwater thruster and the surface control terminal are collected in real time, including signal strength, bit error rate, and transmission delay. The signal strength is compared with a preset strength threshold. When the signal strength is lower than the strength threshold, the current communication quality index is calculated based on the weighted sum of the bit error rate and delay.

[0038] The transmission time periods are classified according to the communication quality index. When the communication quality index is higher than the preset quality threshold, the current time period is marked as a Class A transmission time period, and the correction control instruction is transmitted immediately at the original frequency.

[0039] When the communication quality index is lower than the preset quality threshold, the current time period is marked as a Class II transmission time period, the signal strength fluctuation rate of the Class II transmission time period is obtained, and the number of redundant transmissions of the correction control instruction is calculated based on the fluctuation rate;

[0040] If the signal strength fluctuation rate during the second transmission time period exceeds a preset fluctuation threshold, the communication link is determined to be at risk of intermittent interruption. The motion state parameters of the underwater thruster in the current working path are extracted. The position change of the thruster within a preset time period in the future is predicted based on the motion state parameters. The number of redundant transmissions is adjusted according to the position change, and a compressed encoded version of the correction control instruction is generated.

[0041] When sending compressed coded instructions to the underwater thruster within the second-class transmission time period, the reception status of the instruction confirmation feedback signal is synchronously monitored. When the number of consecutive failures to receive the confirmation feedback signal exceeds a preset threshold, the current communication link is determined to be in an unreliable transmission state, the real-time attitude data and the remaining path tracking error of the underwater thruster are obtained, and the autonomous correction compensation amount is calculated based on the attitude data and the tracking error. The autonomous correction compensation amount is then embedded in the correction control instruction of the next cycle to form an incremental correction instruction.

[0042] When multiple consecutive preset transmission time periods are calibrated as Class II transmission time periods and the cumulative value of the autonomous correction compensation amount exceeds the maximum correction capability of the thruster, it is determined that the current communication quality cannot guarantee the path tracking accuracy, the underwater thruster is urgently stopped, and the transmission plan of the correction control instruction is obtained.

[0043] A second aspect of the present invention further provides an underwater propeller control system based on water environment analysis, the system comprising: a memory and a processor, wherein the memory includes an underwater propeller control method program based on water environment analysis, and when the underwater propeller control method program based on water environment analysis is executed by the processor, the following steps are implemented:

[0044] Acquire image data and laser point cloud data of the target water environment, and construct a three-dimensional environment model of the target water environment based on the image data and laser point cloud data;

[0045] Acquiring initial position information and target position information of the underwater propeller, performing a water environment analysis between the initial position and the target position based on the three-dimensional environment model, and determining a working path of the underwater propeller;

[0046] Acquiring Doppler sonar data of the working path, determining water flow information of the working path according to the Doppler sonar data, and determining a deviation correction control instruction of the underwater propeller according to the water flow information;

[0047] Performing a transmission quality analysis of the deviation correction control instruction on the working path, and determining a transmission scheme for the deviation correction control instruction according to the transmission quality.

[0048] The present invention discloses a method and system for controlling an underwater thruster based on water environment analysis. The method includes: collecting image data and laser point cloud data of the target water body to construct a three-dimensional environmental model; obtaining the initial position information and target position of the underwater thruster, and generating a working path based on environmental model analysis; collecting Doppler sonar data along the path, analyzing water flow information, and generating correction control instructions; and analyzing the quality of instruction transmission to determine the optimal transmission scheme. By integrating multi-source environmental perception data, the present invention achieves dynamic modeling and analysis of complex water environments, improving the path planning accuracy and autonomous control capabilities of the underwater thruster, and is suitable for precise navigation and dynamic obstacle avoidance scenarios in complex underwater operation tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flow chart of an underwater thruster control method based on water environment analysis according to the present invention is shown;

[0050] Figure 2 A flow chart showing the construction of a three-dimensional environmental model of a target water environment according to the present invention is shown;

[0051] Figure 3 A flow chart showing the present invention for determining the working path of an underwater propeller;

[0052] Figure 4 A block diagram of an underwater thruster control system based on water environment analysis of the present invention is shown. DETAILED DESCRIPTION

[0053] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0055] Figure 1 A flow chart of an underwater thruster control method based on water environment analysis of the present invention is shown.

[0056] like Figure 1 As shown, the first aspect of the present invention provides an underwater thruster control method based on water environment analysis, comprising:

[0057] S102, acquiring image data and laser point cloud data of a target water environment, and constructing a three-dimensional environment model of the target water environment based on the image data and laser point cloud data;

[0058] S104, obtaining initial position information and target position information of the underwater propeller, performing a water environment analysis between the initial position and the target position based on the three-dimensional environment model, and determining a working path of the underwater propeller;

[0059] S106, acquiring Doppler sonar data of the working path, determining water flow information of the working path according to the Doppler sonar data, and determining a deviation correction control instruction for the underwater propeller according to the water flow information;

[0060] S108 , analyzing the transmission quality of the correction control instruction on the working path, and determining a transmission scheme for the correction control instruction according to the transmission quality.

[0061] It should be noted that underwater propulsion systems are susceptible to dynamic currents, hidden obstacles, and unstable communications in complex water environments. Traditional methods, due to perception limitations and delayed response, can easily cause yaw or collisions. This solution improves navigation reliability through multi-source data fusion and closed-loop control. First, binocular vision and laser point cloud data are integrated to construct a 3D environmental model. Spatial-temporal alignment and feature matching techniques are used to restore obstacle distribution and terrain gradients, enhancing environmental perception integrity. Secondly, obstacle density field analysis and terrain accessibility assessment are combined to generate an optimal path. High-risk areas are dynamically segmented and local detour strategies are optimized to balance path efficiency and safety. Next, Doppler sonar is used to analyze real-time current information, and dynamic noise suppression and feedforward-feedback composite control are used to generate correction commands to accurately offset current interference. Finally, transmission modes are intelligently switched based on communication quality grading. Command reliability is ensured through compression coding and redundant retransmissions. Autonomous correction logic is embedded to address communication interruptions. This completes a closed-loop control process from environmental modeling and path planning to dynamic correction, significantly improving navigation accuracy and anti-interference capabilities in complex waters.

[0062] Figure 2 The flowchart of the present invention for constructing a three-dimensional environmental model of a target water environment is shown.

[0063] According to an embodiment of the present invention, the acquiring of image data and laser point cloud data of the target water environment, and constructing a three-dimensional environment model of the target water environment based on the image data and laser point cloud data, specifically comprises:

[0064] S202, collecting continuous frame image data of the target water environment through a binocular vision camera, and synchronously obtaining laser point cloud data of the target water environment based on a laser scanning device, and when the time stamp difference between the image data and the laser point cloud data exceeds a preset time synchronization threshold, performing time alignment processing on the image data and the laser point cloud data based on a nearest neighbor matching method of the timestamp;

[0065] S204, extracting image feature points of the image data based on a feature point detection algorithm, and performing curvature analysis on the laser point cloud data to determine laser feature points of the laser point cloud data;

[0066] S206, spatially registering the time-aligned image data with the laser point cloud data, calculating the matching degree of the spatial projection of the image feature points in the laser point cloud data, and when the matching degree of the two in the overlapping area is lower than a preset matching threshold, marking the area as a low-confidence area, and projecting the image feature points in the low-confidence area to the corresponding laser feature points until the matching degree of the image feature points in the spatial projection of the laser point cloud data is no lower than the preset matching threshold;

[0067] S208: Reconstructing a 3D mesh for the area where the matching degree meets the standard, extracting obstacle outline feature points and terrain elevation feature points in the water environment, and marking the area as a sparse feature point area when the number of consecutive feature points is lower than the minimum topological connection number required for path planning. A laser secondary scan is performed on the sparse feature point area, and 3D features are supplemented based on the secondary scan.

[0068] S210, finally generating a three-dimensional environment model including the spatial coordinates of obstacles and the terrain undulation gradient.

[0069] It should be noted that in the process of 3D modeling, the lack of synchronization of sensor data and insufficient feature matching accuracy lead to model distortion, which is specifically manifested in spatial dislocation caused by timestamp deviation, blurred obstacle outlines caused by weak correlation between image and point cloud features, and missing terrain information in feature-sparse areas, resulting in inaccurate 3D modeling. Therefore, the timestamp nearest neighbor matching technology is used to eliminate the timing deviation between binocular vision and laser scanning, and control the time alignment error of cross-sensor data to millisecond level, solving the problem of model splicing dislocation caused by differences in device sampling frequency. Secondly, through image feature point detection and The collaborative processing of point cloud curvature analysis accurately extracts underwater obstacle edge features and terrain mutation points, enhancing the geometric consistency of heterogeneous data spatial registration. Field measurements have shown that this can reduce model distortion caused by feature mismatching. For low-confidence areas, dynamic projection mapping is used to forcibly align image features to laser feature points, filling local data gaps caused by turbid water or insufficient illumination to ensure the complete reproduction of obstacle contours. A laser secondary scanning strategy is triggered for areas with sparse feature points, actively completing terrain elevation data to improve model topological connectivity and avoid planning interruptions caused by missing key path nodes. The resulting three-dimensional environment model has sub-meter obstacle positioning accuracy and centimeter-level terrain gradient resolution, providing a reliable navigation reference for underwater thrusters. Compared with traditional methods, it reduces the error rate of obstacle misjudgment in path planning and significantly reduces the frequency of emergency obstacle avoidance actions caused by model blind spots. The feature point detection algorithm includes SIFT, SURF or ORB two-dimensional feature extraction algorithms for image data, and ISS, Harris3D or curvature feature detection three-dimensional point cloud feature extraction algorithms for laser point cloud data; the image feature points include two-dimensional image features with significant gradients such as corner points, edge points, spots, etc.; laser feature points include corner points, edge points, plane intersection points, and curvature extreme points; the minimum topological connection number refers to the minimum feature point density requirement to ensure the feasibility of path planning in underwater three-dimensional environment modeling, which is a preset value for managers.

[0070] Figure 3 A flow chart of determining the working path of an underwater thruster according to the present invention is shown.

[0071] According to an embodiment of the present invention, the initial position information and the target position information of the underwater propeller are obtained, and the water environment between the initial position and the target position is analyzed based on the three-dimensional environment model to determine the working path of the underwater propeller. Specifically,

[0072] S302: Based on the spatial coordinates of obstacles and the terrain elevation feature points in the three-dimensional environment model, an initial straight line path is generated between the initial position and the target position using an A* algorithm, the initial straight line path is projected into the three-dimensional environment model, and a set of obstacle coordinates and a set of terrain elevation points that the path projection passes through is extracted;

[0073] S304: Calculate the intersection area of the initial straight path projection and the obstacle coordinate set. When the obstacle density in the intersection area exceeds a preset density threshold, mark the area as a high-risk area. Calculate the gap width between adjacent obstacles in the high-risk area. When the gap width is less than the minimum pass width of the underwater propeller, upgrade the high-risk area to an impassable area, and record the boundary coordinates of the impassable area.

[0074] S306: Based on the boundary coordinates of the impassable area, the initial straight path is divided into a plurality of path segments to be bypassed, with the intersection of the initial straight path and the impassable area as a segmentation point. For each path segment to be bypassed, an annular search space is generated by expanding outward from the boundary of the impassable area by a preset safety distance. Within the annular search space, a Dijkstra algorithm is used to calculate a plurality of local paths that bypass the impassable area. The local paths are connected to the undivided endpoints of the initial straight path to generate a plurality of candidate feasible paths.

[0075] S308, calculating the total length, number of turns, and number of obstacle avoidances of each candidate feasible path, scoring the path based on the total length, number of turns, and number of obstacle avoidances, selecting a path based on the path score, and determining the working path of the underwater thruster.

[0076] It should be noted that the initial straight line path is generated based on the obstacle coordinates and terrain elevation data of the three-dimensional environmental model, and the high-risk area is dynamically identified through the density field analysis of the path projection and obstacle coordinates. The impassable area is accurately calibrated in combination with the mechanical parameters of the propeller body (such as the minimum pass width), so as to avoid the problem that the traditional global path planning algorithm falls into the local optimum or generates a mechanically inaccessible path in the dense obstacle area; secondly, the initial path is divided by the boundary coordinates of the impassable area and a circular search space is constructed. The Dijkstra algorithm is used to generate a multi-branch local detour path in the restricted area, which effectively solves the path interruption caused by environmental mutations in the traditional A* algorithm. The cracking problem is solved, and a safe distance expansion strategy is preset to ensure that a physical obstacle avoidance margin is retained between the detour path and the obstacle. Finally, based on a multi-dimensional scoring model of path length, number of turns and number of avoidances, a global optimal path that takes into account navigation efficiency (shortest path priority), motion smoothness (minimum number of turns) and safety (minimum number of obstacle approaches) is screened out from the candidate paths. Compared with the single-index path selection method, this method can reduce the energy loss of the propeller caused by frequent turns, and reduce the risk of control instability caused by the path being close to obstacles in narrow channels. This method can stably generate a continuous and navigable path in complex underwater terrain, and the path smoothness meets the propeller dynamic constraints.

[0077] According to an embodiment of the present invention, acquiring Doppler sonar data of the working path and determining water flow information of the working path according to the Doppler sonar data specifically includes:

[0078] The Doppler sonar array carried by the underwater propeller transmits sound wave signals along the extension direction of the working path and receives reflected signals to obtain Doppler sonar data;

[0079] Calculating the signal-to-noise ratio of the Doppler sonar data, marking areas where the signal-to-noise ratio is lower than a preset threshold as data interference areas, obtaining water turbidity information in the data interference areas, and adjusting the acoustic wave emission energy of the multispectral sonar array according to the water turbidity information until the signal-to-noise ratio of the Doppler sonar data is no lower than the preset threshold;

[0080] Calculating the sound wave frequency offset of Doppler sonar data, and determining real-time water flow information in the axial direction of the working path according to the sound wave frequency offset, wherein the water flow information includes water flow velocity and water flow direction;

[0081] The water flow information is used to construct a water flow information distribution map of the working path with time series changes, and the flow velocity change gradient in the continuous time window is extracted based on the water flow information distribution map. When the flow velocity mutation amplitude of the adjacent time window exceeds the preset mutation threshold, the current time period is marked as the abnormal water flow information acquisition period;

[0082] The Doppler sonar data during the abnormal period of water flow information acquisition are resampled, and the high-frequency noise interference is eliminated through sliding average filtering. The water flow information distribution map is corrected based on the resampled Doppler sonar data.

[0083] It should be noted that underwater thrusters often cause sonar signal attenuation and high-frequency noise interference in water flow monitoring, which leads to distortion of flow rate measurement. In addition, the delay in detecting sudden changes in dynamic water flow can easily cause the correction control to fail. Highly robust current perception is achieved through the adaptive acquisition and dynamic correction mechanism of Doppler sonar data. First, based on real-time signal-to-noise ratio detection and water turbidity correlation analysis, the multispectral sonar emission energy is dynamically adjusted to ensure the stability of sound wave penetration intensity and echo quality in waters with different transmittances. Second, the sound wave frequency shift is jointly analyzed in the time and frequency domains to accurately calculate the axial flow velocity vector and construct a time series current distribution map. Sliding time window gradient detection technology is combined to capture sudden changes in flow velocity. The data during abnormal periods triggers directional resampling and sliding average filtering of the sonar array, effectively eliminating high-frequency noise interference caused by instantaneous occlusion, so that the corrected water flow information distribution map can still maintain the flow velocity direction solution accuracy in turbulent areas. Finally, a water flow perception data chain that takes into account both real-time and anti-interference is formed, providing dynamic water flow field information for subsequent correction control. Compared with traditional solutions, the number of correction errors caused by data quality fluctuations is reduced, and the track tracking lag under sudden water disturbances is significantly reduced.

[0084] According to an embodiment of the present invention, determining the deviation correction control instruction of the underwater propeller according to the water flow information is specifically:

[0085] Based on the water flow information distribution map of the working path, extract the real-time water flow speed and direction at the current position of the underwater propeller, calculate the heading deviation angle between the real-time water flow direction and the working path, and when the heading deviation angle exceeds a preset angle threshold, calibrate the current water flow as a strong interference water flow, and generate a lateral deviation rate based on the product of the heading deviation angle and the real-time water flow speed;

[0086] determining a lateral deviation risk level based on a ratio of the lateral deviation rate to a maximum lateral deviation resistance capability of the underwater propulsor, and generating a first corrective torque based on a reverse vector component of the lateral deviation rate when the lateral deviation risk level exceeds a preset risk threshold;

[0087] It should be noted that when the underwater propeller is cruising along the working path, the effect of the water flow on the underwater propeller will cause the cruising position of the underwater propeller to deviate from the working path. In the scenarios of low flow rate and high deviation angle or high flow rate and low deviation angle, it is easy to produce torque over-compensation or under-compensation, causing the propeller to oscillate and yaw or correct the deviation lag. The static calibration mode of the lateral anti-drift capability does not take into account the real-time motion state of the propeller (such as steering angular velocity, propulsion power load), resulting in a mismatch between the torque output and the body dynamic characteristics. Under extreme working conditions, it may induce propulsion motor overload or rudder effect saturation. And, precise torque control is achieved through vector analysis of water disturbance and dynamic risk assessment: the deviation angle and lateral deviation rate of the water flow direction and path heading are solved in real time, the deviation risk level is dynamically calculated in combination with the current motion parameters of the thruster, and a proportionally adjustable correction torque is generated according to the reverse vector component. This not only avoids the adaptability defects of the single threshold response mechanism in complex flow fields, but also ensures the physical executableness of the control instructions through the dynamic adaptation of the torque output and the anti-drift capability of the body, thereby maintaining the heading stability in a strong interference environment while reducing the overshoot loss of the actuator.

[0088] The axial component of the real-time water flow velocity along the working path is simultaneously calculated. When the composite speed of the axial velocity component and the current propulsion speed of the propeller is lower than a preset minimum speed threshold, the current water flow is calibrated as a reverse strong resistance water flow. The output power of the main propulsion motor of the propeller is increased according to the speed amplitude of the reverse strong resistance water flow to generate a second corrective thrust.

[0089] Performing vector superposition of the first correcting torque and the second correcting thrust to generate an initial correcting control instruction;

[0090] It should be noted that the composite speed refers to the vector sum of the underwater propeller's own propulsion speed and the axial component of the water flow, that is, the actual movement speed after the two are superimposed; the initial correction control instruction first decomposes the lateral anti-offset torque and axial anti-drag thrust into the propeller coordinate system, and uses the direction cosine matrix to convert it to the global coordinate system to eliminate the directional coupling error; the weight coefficients of the lateral and axial components are dynamically allocated based on the real-time water flow interference intensity. For example, in a strong lateral flow scenario, the lateral torque is preferentially amplified to offset the offset, and when the reverse resistance is dominant, the axial thrust is focused on maintaining the speed; combined with the physical limits of the propeller actuator, the composite instruction is amplitude saturated and rate smoothed to ensure that the output is not overloaded and the direction is stable; finally, the correction effect of the instruction on the heading deviation is verified by feedforward prediction. If the residual error exceeds the threshold, the weight coefficient is fine-tuned based on the feedback integral term to form an initial instruction that takes into account both anti-interference and motion stability, so that the control instruction meets the dynamic constraints and can accurately offset the complex water flow disturbance.

[0091] Acquire real-time attitude sensor data from the underwater thruster, calculate the predicted heading angle and predicted position coordinate changes of the underwater thruster after the initial correction control command is applied, and when the predicted heading angle does not match the minimum curvature radius of the remaining path segment of the working path, adjust the weight coefficient of the first correction torque according to the curvature change gradient of the remaining path segment until the predicted heading angle meets the path tracking accuracy requirements;

[0092] When the lateral deviation between the predicted position coordinates and the working path continues to increase, a third compensation torque is generated based on the integral term of the lateral deviation, and the third compensation torque is superimposed on the initial correction control instruction to form a final correction control instruction.

[0093] It should be noted that if the turning radius corresponding to the predicted heading angle is greater than the minimum curvature radius of the path, the thruster will not be able to adhere to the curved path due to steering hysteresis, resulting in the accumulation of outer offset during path tracking; the first correcting torque weight coefficient is adjusted by calculating the theoretical heading angle deviation between the predicted heading angle and the target curvature of the remaining segment of the path, and a proportional-integral control model is constructed based on the deviation to convert the curvature error into a weight adjustment amount (e.g., for every unit increase in the deviation, the weight is increased by a specific proportional coefficient); at the same time, the proportional coefficient is dynamically corrected according to the matching degree between the current steering rate of the thruster and the rate of change of the path curvature to ensure that the weight adjustment is adapted to the steering capability of the body; finally, the weight coefficient is updated through real-time iteration and the fit between the corrected heading angle and the path curvature is verified until the heading tracking error converges to within the preset threshold range, thereby achieving adaptive matching between the correcting torque and the path geometric constraints.

[0094] According to an embodiment of the present invention, the performing of transmission quality analysis of the correction control instruction on the working path and determining a transmission scheme for the correction control instruction based on the transmission quality are specifically as follows:

[0095] The communication link parameters between the underwater thruster and the surface control terminal are collected in real time, including signal strength, bit error rate, and transmission delay. The signal strength is compared with a preset strength threshold. When the signal strength is lower than the strength threshold, the current communication quality index is calculated based on the weighted sum of the bit error rate and delay.

[0096] The transmission time periods are classified according to the communication quality index. When the communication quality index is higher than the preset quality threshold, the current time period is marked as a Class A transmission time period, and the correction control instruction is transmitted immediately at the original frequency.

[0097] When the communication quality index is lower than the preset quality threshold, the current time period is marked as a Class II transmission time period, the signal strength fluctuation rate of the Class II transmission time period is obtained, and the number of redundant transmissions of the correction control instruction is calculated based on the fluctuation rate;

[0098] If the signal strength fluctuation rate during the second transmission time period exceeds a preset fluctuation threshold, the communication link is determined to be at risk of intermittent interruption. The motion state parameters of the underwater thruster in the current working path are extracted. The position change of the thruster within a preset time period in the future is predicted based on the motion state parameters. The number of redundant transmissions is adjusted according to the position change, and a compressed encoded version of the correction control instruction is generated.

[0099] When sending compressed coded instructions to the underwater thruster within the second-class transmission time period, the reception status of the instruction confirmation feedback signal is synchronously monitored. When the number of consecutive failures to receive the confirmation feedback signal exceeds a preset threshold, the current communication link is determined to be in an unreliable transmission state, the real-time attitude data and the remaining path tracking error of the underwater thruster are obtained, and the autonomous correction compensation amount is calculated based on the attitude data and the tracking error. The autonomous correction compensation amount is then embedded in the correction control instruction of the next cycle to form an incremental correction instruction.

[0100] When multiple consecutive preset transmission time periods are calibrated as Class II transmission time periods and the cumulative value of the autonomous correction compensation amount exceeds the maximum correction capability of the thruster, it is determined that the current communication quality cannot guarantee the path tracking accuracy, the underwater thruster is urgently stopped, and the transmission plan of the correction control instruction is obtained.

[0101] It should be noted that the communication level is divided in real time based on the dynamic weighted evaluation of signal strength, bit error rate and delay. In high-quality channel periods, full-frequency command instant transmission is used to ensure control timeliness. In low-quality channel periods, redundant retransmission and compression coding adaptation (such as Huffman coding to reduce data volume) are used to improve the command reception success rate in weak signal environments. At the same time, the risk of communication interruption is predicted based on the signal fluctuation rate, and incremental correction commands are pre-generated to solve the problem of control command loss or delay caused by signal attenuation in traditional underwater communications. Secondly, a multi-cycle feedback monitoring mechanism is used to identify the continuous communication degradation state, trigger the autonomous correction compensation logic of the thruster body, and use the local closed-loop control of real-time attitude data and path tracking error to generate incremental commands, maintain basic heading stability during communication interruption, and avoid the accumulation of path deviation caused by command interruption. Finally, a dual fuse mechanism is set for communication quality deterioration and correction capability exceeding the limit. When the autonomous compensation amount exceeds the mechanical limit of the thruster, an emergency shutdown is performed to prevent collision or equipment damage caused by loss of control. This solution deeply integrates communication link status into the control command generation and transmission strategy, achieving full reliability enhancement from command encoding optimization, redundant anti-packet loss, to autonomous fault tolerance. Compared with traditional fixed-rate transmission solutions, this reduces path tracking error exceeding limits due to communication failures and significantly extends the continuous operation time of the thruster in complex hydrological environments. The number of redundant transmissions refers to the number of times the correction control command is repeated; the autonomous correction compensation includes the heading correction torque, lateral position compensation thrust, and attitude stabilization control.

[0102] According to an embodiment of the present invention, the further embodiment includes:

[0103] Real-time monitoring of the turbidity change rate of the target water environment. When the turbidity change rate exceeds a preset change threshold, the underwater propeller displacement from the current moment to the completion of the laser secondary scan is calculated based on the propeller motion trajectory prediction model to generate a displacement compensation vector.

[0104] Pre-correcting the spatial positions of the image feature points in the low-confidence area according to the displacement compensation vector, synchronously extracting the confidence weights of the curvature feature points in the current laser point cloud data, and weightedly fusing the pre-corrected image feature points with the high-confidence curvature feature points to generate a mixed feature point set that is resistant to turbidity interference;

[0105] Based on the real-time roll and pitch angle data fed back by the thruster attitude sensor, the mixed feature point set is compensated for fluid disturbances, the feature point projection distortion caused by thruster turbulence is eliminated, and a stabilized three-dimensional feature topology structure is output;

[0106] The coordinates of obstacles in the 3D environment model are updated according to the stabilized 3D feature topology structure.

[0107] It should be noted that in dynamic turbid water environments, the spatiotemporal registration of laser and visual feature points faces severe challenges due to sudden changes in water transmittance. Sediment turbulence causes a time lag between the secondary scan of the low-confidence region and the propeller movement, leading to delayed 3D model updates and a disconnect between obstacle avoidance commands. The combined effects of propeller turbulence and water flow disturbances further amplify feature point projection distortion, causing obstacle coordinate drift and navigation risks. Therefore, a dynamic compensation mechanism is triggered by real-time turbidity change monitoring. This problem of feature point misalignment is addressed through displacement prediction and multi-source data fusion. Projection distortion is eliminated by combining pose feedback and fluid disturbance compensation, resulting in a stabilized 3D feature topology. Pre-obstacle avoidance commands are generated through real-time obstacle coordinate updates and collision risk simulation, enabling path optimization and energy consumption control in turbulent areas. This complete system of multi-dimensional variable collaborative perception and dynamic model closed-loop updates overcomes the limitations of single data sources or static models, providing highly robust real-time perception and autonomous obstacle avoidance capabilities for complex underwater environments.

[0108] Figure 4 A block diagram of an underwater thruster control system based on water environment analysis of the present invention is shown.

[0109] A second aspect of the present invention further provides an underwater propeller control system 4 based on water environment analysis, the system comprising: a memory 41 and a processor 42, wherein the memory includes an underwater propeller control method program based on water environment analysis, and when the underwater propeller control method program based on water environment analysis is executed by the processor, the following steps are implemented:

[0110] Acquire image data and laser point cloud data of the target water environment, and construct a three-dimensional environment model of the target water environment based on the image data and laser point cloud data;

[0111] Acquiring initial position information and target position information of the underwater propeller, performing a water environment analysis between the initial position and the target position based on the three-dimensional environment model, and determining a working path of the underwater propeller;

[0112] Acquiring Doppler sonar data of the working path, determining water flow information of the working path according to the Doppler sonar data, and determining a deviation correction control instruction of the underwater propeller according to the water flow information;

[0113] Performing a transmission quality analysis of the deviation correction control instruction on the working path, and determining a transmission scheme for the deviation correction control instruction according to the transmission quality.

[0114] The present invention discloses a method and system for controlling an underwater thruster based on water environment analysis. The method includes: collecting image data and laser point cloud data of the target water body to construct a three-dimensional environmental model; obtaining the initial position information and target position of the underwater thruster, and generating a working path based on environmental model analysis; collecting Doppler sonar data along the path, analyzing water flow information, and generating correction control instructions; and analyzing the quality of instruction transmission to determine the optimal transmission scheme. By integrating multi-source environmental perception data, the present invention achieves dynamic modeling and analysis of complex water environments, improving the path planning accuracy and autonomous control capabilities of the underwater thruster, and is suitable for precise navigation and dynamic obstacle avoidance scenarios in complex underwater operation tasks.

[0115] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0116] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0117] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0118] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0119] Alternatively, if the integrated units described above are implemented as software modules and sold or used as standalone products, they can also be stored on a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product, stored on a storage medium, includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute all or part of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as removable storage devices, ROM, RAM, magnetic disks, or optical disks.

[0120] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for controlling an underwater propeller based on water environment analysis, characterized in that: The following steps are involved: Acquire image data and laser point cloud data of the target water environment, and construct a three-dimensional environment model of the target water environment based on the image data and laser point cloud data; Acquiring initial position information and target position information of the underwater propeller, performing a water environment analysis between the initial position and the target position based on the three-dimensional environment model, and determining a working path of the underwater propeller; Acquiring Doppler sonar data of the working path, determining water flow information of the working path according to the Doppler sonar data, and constructing a water flow information distribution map of the working path that changes in time series according to the water flow information; The correction control instruction of the underwater propeller is determined according to the water flow information, specifically: Based on the water flow information distribution map of the working path, the real-time water flow speed and direction at the current position of the underwater propeller are extracted, and the heading deviation angle between the real-time water flow direction and the working path is calculated. When the heading deviation angle exceeds a preset angle threshold, the current water flow is calibrated as a strong interference water flow, and a lateral deviation rate is generated according to the product of the heading deviation angle and the real-time water flow speed; determining a lateral deviation risk level based on a ratio of the lateral deviation rate to a maximum lateral deviation resistance capability of the underwater propulsor, and generating a first corrective torque based on a reverse vector component of the lateral deviation rate when the lateral deviation risk level exceeds a preset risk threshold; The axial component of the real-time water flow velocity along the working path is simultaneously calculated. When the composite speed of the axial velocity component and the current propulsion speed of the propeller is lower than a preset minimum speed threshold, the current water flow is calibrated as a reverse strong resistance water flow. The output power of the main propulsion motor of the propeller is increased according to the speed amplitude of the reverse strong resistance water flow to generate a second corrective thrust. Performing vector superposition of the first correcting torque and the second correcting thrust to generate an initial correcting control instruction; Acquire real-time attitude sensor data from the underwater thruster, calculate the predicted heading angle and predicted position coordinate changes of the underwater thruster after the initial correction control command is applied, and when the predicted heading angle does not match the minimum curvature radius of the remaining path segment of the working path, adjust the weight coefficient of the first correction torque according to the curvature change gradient of the remaining path segment until the predicted heading angle meets the path tracking accuracy requirements; When the lateral deviation between the predicted position coordinates and the working path continues to increase, a third compensation torque is generated based on the integral term of the lateral deviation, and the third compensation torque is added to the initial correction control instruction to form a final correction control instruction; Performing a transmission quality analysis of the deviation correction control instruction on the working path, and determining a transmission scheme for the deviation correction control instruction according to the transmission quality.

2. The underwater propeller control method based on water environment analysis according to claim 1 is characterized in that: The step of acquiring image data and laser point cloud data of the target water environment and constructing a three-dimensional environment model of the target water environment based on the image data and laser point cloud data is as follows: Continuous frame image data of the target water environment is collected by a binocular vision camera, and laser point cloud data of the target water environment is obtained synchronously based on a laser scanning device. When the time stamp difference between the image data and the laser point cloud data exceeds a preset time synchronization threshold, the image data and the laser point cloud data are time-aligned using a nearest neighbor matching method based on the timestamp; Extracting image feature points of the image data based on a feature point detection algorithm, and performing curvature analysis on the laser point cloud data to determine laser feature points of the laser point cloud data; Perform spatial registration on the time-aligned image data and the laser point cloud data, and calculate the matching degree of the spatial projection of the image feature points in the laser point cloud data. When the matching degree of the two in the overlapping area is lower than the preset matching threshold, the area is marked as a low-confidence area, and the image feature points in the low-confidence area are projected to the corresponding laser feature points until the matching degree of the image feature points in the spatial projection of the laser point cloud data is no lower than the preset matching threshold. 3D mesh reconstruction is performed on areas where the matching degree meets the standards. Feature points of obstacle outlines and terrain elevation in the water environment are extracted. When the number of consecutive feature points is lower than the minimum number of topological connections required for path planning, the area is marked as a feature point sparse area. The feature point sparse area is scanned again by laser, and 3D features are supplemented based on the second scan. Finally, a three-dimensional environment model including the spatial coordinates of obstacles and terrain gradients is generated.

3. The underwater propulsion control method based on water environment analysis according to claim 1 is characterized in that: The process of obtaining the initial position information and the target position information of the underwater propeller, analyzing the water environment between the initial position and the target position according to the three-dimensional environment model, and determining the working path of the underwater propeller is as follows: Based on the spatial coordinates of obstacles and terrain elevation feature points in the three-dimensional environment model, an initial straight line path is generated between the initial position and the target position using the A* algorithm, the initial straight line path is projected into the three-dimensional environment model, and a set of obstacle coordinates and a set of terrain elevation points that the path projection passes through are extracted; Calculating the intersection area of the initial straight path projection and the obstacle coordinate set; when the obstacle density in the intersection area exceeds a preset density threshold, marking the area as a high-risk area; calculating the gap width between adjacent obstacles in the high-risk area; when the gap width is less than the minimum pass width of the underwater propeller, upgrading the high-risk area to an impassable area, and recording the boundary coordinates of the impassable area; Based on the boundary coordinates of the impassable area, the initial straight path is divided into several segments to be bypassed, with the intersection of the initial straight path and the impassable area as a segmentation point. For each segment to be bypassed, a ring search space is generated by expanding a preset safety distance outward along the boundary of the impassable area. Within the ring search space, a Dijkstra algorithm is used to calculate several local paths that bypass the impassable area. The local paths are connected to the undivided endpoints of the initial straight path to generate multiple candidate feasible paths. The total length, number of turns, and number of obstacle avoidances of each candidate feasible path are calculated, a path score is performed based on the total length, number of turns, and number of obstacle avoidances, a path is selected based on the path score, and an operating path of the underwater thruster is determined.

4. The underwater propeller control method based on water environment analysis according to claim 1, characterized in that: The acquiring of Doppler sonar data of the working path and determining the water flow information of the working path according to the Doppler sonar data are specifically as follows: The Doppler sonar array carried by the underwater propeller transmits sound wave signals along the extension direction of the working path and receives reflected signals to obtain Doppler sonar data; Calculating the signal-to-noise ratio of the Doppler sonar data, marking areas where the signal-to-noise ratio is lower than a preset threshold as data interference areas, obtaining water turbidity information in the data interference areas, and adjusting the acoustic wave emission energy of the multispectral sonar array according to the water turbidity information until the signal-to-noise ratio of the Doppler sonar data is no lower than the preset threshold; Calculating the sound wave frequency offset of Doppler sonar data, and determining real-time water flow information in the axial direction of the working path according to the sound wave frequency offset, wherein the water flow information includes water flow velocity and water flow direction; The water flow information is used to construct a water flow information distribution map of the working path with time series changes, and the flow velocity change gradient in the continuous time window is extracted based on the water flow information distribution map. When the flow velocity mutation amplitude of the adjacent time window exceeds the preset mutation threshold, the current time period is marked as the abnormal water flow information acquisition period; The Doppler sonar data during the abnormal period of water flow information acquisition are resampled, and the high-frequency noise interference is eliminated through sliding average filtering. The water flow information distribution map is corrected based on the resampled Doppler sonar data.

5. The underwater propulsion control method based on water environment analysis according to claim 1 is characterized in that: The performing of transmission quality analysis of the correction control instruction on the working path and determining a transmission scheme of the correction control instruction according to the transmission quality is specifically as follows: The communication link parameters between the underwater thruster and the surface control terminal are collected in real time, including signal strength, bit error rate, and transmission delay. The signal strength is compared with a preset strength threshold. When the signal strength is lower than the strength threshold, the current communication quality index is calculated based on the weighted sum of the bit error rate and delay. The transmission time periods are classified according to the communication quality index. When the communication quality index is higher than the preset quality threshold, the current time period is marked as a Class A transmission time period, and the correction control instruction is transmitted immediately at the original frequency. When the communication quality index is lower than the preset quality threshold, the current time period is marked as a Class II transmission time period, the signal strength fluctuation rate of the Class II transmission time period is obtained, and the number of redundant transmissions of the correction control instruction is calculated based on the fluctuation rate; If the signal strength fluctuation rate during the second transmission time period exceeds a preset fluctuation threshold, the communication link is determined to be at risk of intermittent interruption. The motion state parameters of the underwater thruster in the current working path are extracted. The position change of the thruster within a preset time period in the future is predicted based on the motion state parameters. The number of redundant transmissions is adjusted according to the position change, and a compressed encoded version of the correction control instruction is generated. When sending compressed coded instructions to the underwater thruster within the second-class transmission time period, the reception status of the instruction confirmation feedback signal is synchronously monitored. When the number of consecutive failures to receive the confirmation feedback signal exceeds a preset threshold, the current communication link is determined to be in an unreliable transmission state, the real-time attitude data and the remaining path tracking error of the underwater thruster are obtained, and the autonomous correction compensation amount is calculated based on the attitude data and the tracking error. The autonomous correction compensation amount is then embedded in the correction control instruction of the next cycle to form an incremental correction instruction. When multiple consecutive preset transmission time periods are calibrated as Class II transmission time periods and the cumulative value of the autonomous correction compensation amount exceeds the maximum correction capability of the thruster, it is determined that the current communication quality cannot guarantee the path tracking accuracy, the underwater thruster is urgently stopped, and the transmission plan of the correction control instruction is obtained.

6. An underwater propulsion system control system based on water environment analysis, characterized in that: The underwater propeller control system based on water environment analysis includes a storage and a processor. The storage includes an underwater propeller control method program based on water environment analysis. When the underwater propeller control method program based on water environment analysis is executed by the processor, the following steps are implemented: Acquire image data and laser point cloud data of the target water environment, and construct a three-dimensional environment model of the target water environment based on the image data and laser point cloud data; Acquiring initial position information and target position information of the underwater propeller, performing a water environment analysis between the initial position and the target position based on the three-dimensional environment model, and determining a working path of the underwater propeller; Acquiring Doppler sonar data of the working path, determining water flow information of the working path according to the Doppler sonar data, and constructing a water flow information distribution map of the working path that changes in time series according to the water flow information; The correction control instruction of the underwater propeller is determined according to the water flow information, specifically: Based on the water flow information distribution map of the working path, the real-time water flow speed and direction at the current position of the underwater propeller are extracted, and the heading deviation angle between the real-time water flow direction and the working path is calculated. When the heading deviation angle exceeds a preset angle threshold, the current water flow is calibrated as a strong interference water flow, and a lateral deviation rate is generated according to the product of the heading deviation angle and the real-time water flow speed; determining a lateral deviation risk level based on a ratio of the lateral deviation rate to a maximum lateral deviation resistance capability of the underwater propulsor, and generating a first corrective torque based on a reverse vector component of the lateral deviation rate when the lateral deviation risk level exceeds a preset risk threshold; The axial component of the real-time water flow velocity along the working path is simultaneously calculated. When the composite speed of the axial velocity component and the current propulsion speed of the propeller is lower than a preset minimum speed threshold, the current water flow is calibrated as a reverse strong resistance water flow. The output power of the main propulsion motor of the propeller is increased according to the speed amplitude of the reverse strong resistance water flow to generate a second corrective thrust. Performing vector superposition of the first correcting torque and the second correcting thrust to generate an initial correcting control instruction; Acquire real-time attitude sensor data from the underwater thruster, calculate the predicted heading angle and predicted position coordinate changes of the underwater thruster after the initial correction control command is applied, and when the predicted heading angle does not match the minimum curvature radius of the remaining path segment of the working path, adjust the weight coefficient of the first correction torque according to the curvature change gradient of the remaining path segment until the predicted heading angle meets the path tracking accuracy requirements; When the lateral deviation between the predicted position coordinates and the working path continues to increase, a third compensation torque is generated based on the integral term of the lateral deviation, and the third compensation torque is added to the initial correction control instruction to form a final correction control instruction; Performing a transmission quality analysis of the deviation correction control instruction on the working path, and determining a transmission scheme for the deviation correction control instruction according to the transmission quality.

Images