Unmanned ship underwater obstacle avoidance system based on distributed wide-beam sonar

Abstract

The invention discloses an unmanned ship underwater obstacle avoidance system based on distributed wide-beam sonar, and the system collects original environment information through a wide-beam sonar group, carries out the dynamic noise reduction processing through adaptive filtering, and obtains environment detection information. And acquiring the motion state of the unmanned ship through a navigation compensation module, and performing motion error compensation on the environment detection information to obtain environment correction information. And constructing an ellipsoid space safety area based on the environment correction information, and calculating a safety navigation boundary in real time to obtain safety navigation information. The obstacle avoidance decision module generates a multi-level obstacle avoidance strategy according to the real-time navigation state and the safe navigation information, and the execution control module converts the strategy into an obstacle avoidance instruction and dynamically adjusts the output frequency to adapt to different navigation working conditions. The system effectively solves the problems that a traditional obstacle avoidance system is insufficient in detection precision and lags behind in response in a complex underwater environment, all-directional and high-precision control over underwater obstacle avoidance of the unmanned surface vehicle is achieved, and the real-time performance and environmental adaptability of the obstacle avoidance system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned boats, and particularly to an underwater obstacle avoidance system for unmanned boats based on distributed wide-beam sonar. Background Art

[0002] With the increasingly wide application of unmanned boats in the field of underwater operations, the navigation safety challenges they face in complex underwater environments are becoming more and more severe. The autonomous obstacle avoidance system is the core technology to ensure navigation safety, and sonar detection has become the mainstream solution due to its all-weather working characteristics. Traditional systems mostly adopt centralized transceiver-integrated sonar, and realize environmental perception through mechanical rotation or phased array methods. However, in actual operations, affected by the characteristics of complex underwater environments and the physical limitations of equipment, the current technical bottlenecks include: in dynamic sea conditions, the violent swing of the hull causes the sonar beam pointing to be misaligned, and the superposition of the shallow water reverberation effect (such as the effective detection distance attenuation of 70% at a water depth of 5 meters) results in distorted obstacle feature extraction; the inherent electromagnetic coupling problem of the transceiver-integrated structure forms a near-field blind area (typical value 0.35 - 1 meter), which, together with the response delay of the mechanical scanning mechanism (≥0.5 seconds), restricts the fast obstacle avoidance ability; when multiple sonars work together, frequency band conflicts (such as the detection distance decreasing by 40% due to 137kHz interference) and the inherent delay of centralized processing (the full-link cycle exceeds 500ms) make it difficult to meet the real-time obstacle avoidance requirements during high-speed navigation. These factors together cause the obstacle misjudgment rate of existing systems in complex waters to generally exceed 12%, seriously restricting the operation reliability of unmanned boats in typical scenarios such as reef areas and shoals. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to propose an underwater obstacle avoidance system for unmanned boats based on distributed wide-beam sonar, so as to solve the problems of insufficient sonar detection accuracy and real-time obstacle avoidance ability in complex underwater environments.

[0004] To achieve the above technical objectives, the technical solution adopted by the present invention is as follows: An unmanned boat underwater obstacle avoidance system based on a distributed wide beam sonar, comprising a sonar detection module, a navigation compensation module, a safety domain construction module, an obstacle avoidance decision-making module, and an execution control module. The sonar detection module includes a wide beam sonar group, and each wide beam sonar group includes a wide beam transmitter and a wide beam receiver. The sonar detection module is used to collect original environmental information and perform dynamic noise reduction processing on the original environmental information by using adaptive filtering to obtain environmental detection information; the navigation compensation module is used to obtain the motion state of the unmanned boat and perform motion error compensation on the environmental detection information to obtain environmental correction information; the safety domain construction module is used to construct an ellipsoidal space safety area of the current unmanned boat according to the environmental correction information and calculate the safety navigation boundary corresponding to the ellipsoidal space safety area in real time to obtain safety navigation information; the obstacle avoidance decision-making module is used to obtain the real-time navigation state and safety navigation information of the current unmanned boat and generate a multi-level obstacle avoidance strategy with priorities; the execution control module is used to convert the multi-level obstacle avoidance strategy into an obstacle avoidance instruction, and the obstacle avoidance instruction includes a heading control instruction and a speed control instruction, and dynamically adjusts the output frequency of the obstacle avoidance instruction to adapt to different navigation conditions.

[0005] In some embodiments, the sonar detection module is used to collect original environmental information and perform dynamic noise reduction processing on the original environmental information by using adaptive filtering to obtain environmental detection information, including: Control the wide beam transmitter to emit acoustic wave signals. The wide beam transmitter uses linear frequency modulation, and the transmission power of the wide beam transmitter is generated according to the reverberation intensity monitored in real time; Receive underwater echo signals through the wide beam receiver, and perform signal amplification processing, band-pass filtering processing, and analog-to-digital conversion processing on the underwater echo signals to obtain original environmental information; Perform adaptive filtering processing on the original environmental information, including: Perform primary filtering on the original environmental information by using the normalized least mean square algorithm to obtain the first filtering information; Perform secondary filtering on the first filtering information by using a convolutional neural network model to obtain the second filtering information; Perform time delay estimation on the second filtering information by using a pulse compression algorithm, correct the earliest echo arrival time, and obtain the third filtering information; Calculate the distance to the obstacle according to the sound speed and the third filtering information, and generate environmental detection information including the position of the obstacle, the distance to the obstacle, and the intensity of the obstacle; And, in the case of communication anomalies, temporarily store the environmental detection information in the local cache, and record the data acquisition timestamp and the signal quality index.

[0006] In some embodiments, the convolutional neural network model is obtained through the following steps: Build an initial network model and input sample data for training. The sample data includes sample echo signals. The initial network model includes a feature extraction layer and a feature classification layer. The feature extraction layer is used to perform multi-scale feature extraction on the sample echo signals to obtain sample target echoes and sample noise components. The feature classification layer is used to distinguish the sample target echoes from the sample noise components; Repeat training the initial network model until the classification accuracy reaches a preset threshold to obtain a convolutional neural network model.

[0007] In some embodiments, the navigation compensation module is used to obtain the motion state of the unmanned boat and perform motion error compensation on the environmental detection information to obtain environmental correction information, including: Collect the real-time attitude data of the unmanned boat through a MEMS inertial measurement unit. The real-time attitude data includes roll angle, pitch angle, and heading angle; Obtain the position information and speed information of the unmanned boat through a dual-frequency GNSS receiver, and construct a tightly coupled navigation solution model with the position information, speed information, and real-time attitude data; Perform data fusion on the tightly coupled navigation solution model based on an improved robust UKF algorithm to calculate the motion compensation parameters of the unmanned boat in the world coordinate system. The motion compensation parameters include position offset and attitude correction; Perform coordinate transformation compensation on the environmental detection information according to the motion compensation parameters, including: Establish the conversion relationship between the hull coordinate system and the world coordinate system; Convert the obstacle distance in the environmental detection information from the hull coordinate system to the world coordinate system; Compensate for the sonar measurement error caused by the hull movement; Output the environmental correction information including the accurate position of the obstacle in the world coordinate system; And, when the GNSS signal is lost, switch to the pure inertial navigation mode and record the navigation status flag bit.

[0008] In some embodiments, the safety domain construction module is used to construct an ellipsoidal space safety area of the current unmanned boat according to the environmental correction information, including: Obtain the obstacle positions in the environmental correction information one by one, and extract the azimuth angle and distance information of the obstacles relative to the unmanned boat; Take the first actual position of the wide beam transmitter and the second actual position of the wide beam receiver at the current moment as the two foci, and calculate the ellipsoid parameters based on the acoustic wave propagation time difference, including: Take the shortest propagation distance corresponding to the original environmental information of the earliest echo as the major axis of the ellipsoid; Determine the focal length of the ellipsoid according to the distance between the transceiver transducers, and calculate the minor axis of the ellipsoid according to the real-time draft depth of the unmanned boat; Obtain the ellipsoid parameters based on the major axis, minor axis, and focal length of the ellipsoid; Construct a dynamic ellipsoid safety model based on the ellipsoid parameters in the world coordinate system, including: Establish an ellipsoid equation with the transceiver as the two foci; Update the position of the ellipsoid space according to the real-time navigation data; Visualize the safe area of the ellipsoid space.

[0009] In some embodiments, calculate the safe navigation boundary corresponding to the safe area of the ellipsoid space in real time to obtain safe navigation information, including: Calculate the spatial relationship between the safe area of the ellipsoid space and the hull contour of the unmanned boat to generate an initial safe navigation boundary; Output the initial safe area dataset of the safe area of the ellipsoid space according to the initial safe navigation boundary. The initial safe area dataset includes an initial boundary coordinate set and an initial nearest obstacle warning level; Based on the initial boundary coordinate set, establish a three-dimensional spatial grid map, which includes multiple grid cells; Mark the grid cells as safe areas, warning areas, or dangerous areas one by one according to the initial nearest obstacle warning level; Obtain the dynamic parameters of the unmanned boat, and calculate the navigable area according to the dynamic parameters. The dynamic parameters include the current speed, steering angle, roll angle, and pitch angle of the unmanned boat, including: Generate the minimum turning radius according to the current speed and steering angle of the unmanned boat; Generate the influence coefficient of the current minimum turning radius on the actual operable space according to the roll angle and pitch angle; Generate the navigable area of the three-dimensional spatial grid map according to the minimum turning radius, influence coefficient, and preset safety margin threshold; Perform multi-level boundary analysis in the navigable area, including: Calculate the minimum safety distance from the ellipsoid surface to the nearest obstacle to obtain the primary boundary; Predict the track envelope within a preset short time interval in the future according to the primary boundary and the current speed to generate a secondary boundary; Generate a dynamic safety corridor according to the primary boundary and the secondary boundary to obtain the safe navigation boundary; And, when it is detected that the width of the dynamic safety corridor is less than a preset multiple of the hull width, automatically trigger the speed reduction mode and re-plan the safe navigation boundary; Generate safe navigation information according to the safe navigation boundary. The safe navigation information includes the optimal safe heading suggestion, the maximum allowable speed in each direction, the emergency braking distance parameter, and the multi-level warning status identifier.

[0010] In some embodiments, the obstacle avoidance decision-making module is used to obtain the real-time navigation state and safe navigation information of the current unmanned boat, and generate a multi-level obstacle avoidance strategy with priorities, including: Construct an obstacle avoidance decision tree based on risk assessment. The obstacle avoidance decision tree includes a first-level response decision level, a second-level response decision level, and a third-level response decision level. The first-level response decision level is configured to immediately trigger an emergency braking instruction when the detected collision time is less than 3 seconds. The second-level response decision level is configured to generate a heading correction strategy when the width of the dynamic safety corridor is less than 2 times the boat width. The third-level response decision level is configured to select an obstacle avoidance route based on the principle of optimal energy consumption when there are multiple feasible paths; Input the real-time navigation state and safe navigation information of the current unmanned boat into the obstacle avoidance decision tree to obtain an output result, and perform dynamic strategy optimization on the output result, including: Calculate the feasibility index of each obstacle avoidance plan in the output result in real time. The feasibility index includes a path smoothness score, an energy consumption cost score, and a task delay cost score; Use a fuzzy logic algorithm to weight and fuse the feasibility indices to obtain a comprehensive score for each obstacle avoidance plan; Generate the priorities of multiple obstacle avoidance plans according to the comprehensive score, and generate a multi-level obstacle avoidance strategy package.

[0011] In some embodiments, the execution control module is used to convert the multi-level obstacle avoidance strategy into obstacle avoidance instructions. The obstacle avoidance instructions include a heading control instruction and a speed control instruction, including: Convert the path planning parameters of each obstacle avoidance path in the multi-level obstacle avoidance strategy into a heading control instruction, including: Extract the heading angle sequence of the obstacle avoidance path; Use a trajectory smoothing algorithm to generate a continuous steering instruction according to the heading angle sequence; Calculate the actual rudder angle control amount according to the rudder effect characteristics to obtain the heading control instruction; Convert the speed planning parameters of each obstacle avoidance path in the multi-level obstacle avoidance strategy into a speed control instruction, including: Analyze the recommended speed under the safety distance constraint and generate an acceleration curve that conforms to the propulsion characteristics; Calculate the final thrust control parameter according to the acceleration curve to obtain the speed control instruction; Generate an obstacle avoidance instruction according to the heading control instruction and the speed control instruction, and output it through a multi-mode output interface. The multi-mode output interface is configured as: Adopt a periodic instruction output mode in the normal cruise mode; Enable an event-triggered instruction output mode in the emergency obstacle avoidance mode; Switch to a degraded instruction output mode when the system is abnormal; And, before output, perform safety verification processing on the obstacle avoidance instruction, including: Verify the valid range of the instruction parameters of the obstacle avoidance instruction; Limit the instruction change rate of the obstacle avoidance instruction; Check the coordination between multiple instructions of the obstacle avoidance instruction.

[0012] In some embodiments, dynamically adjust the output frequency of the obstacle avoidance instruction to adapt to different navigation conditions, including: Determine the reference output frequency according to the current navigation condition, including: Obtain the real-time navigation state parameters of the unmanned boat; Identify the type of working condition currently located according to the real-time navigation state parameters; Query the working condition-frequency mapping table to determine the reference output frequency; Gradually increase the reference output frequency when the speed in the real-time navigation state parameters increases, and record the frequency adjustment log information; During the adjustment process of the reference output frequency, perform smooth transition processing, including: Set the frequency switching transition interval, and adopt a progressive adjustment algorithm to maintain the continuity of the obstacle avoidance instruction during output.

[0013] In some embodiments, the underwater obstacle avoidance system of the unmanned boat based on the distributed wide-beam sonar further includes a reverberation suppression module and a zero-distance blind area compensation module. The reverberation suppression module is used to establish a seabed-water surface multipath reflection channel model, use an adaptive notch filter to suppress reverberation in a specific frequency band, eliminate false echoes, and adjust the signal pulse width of the transmitted signal to optimize the range resolution; the zero-distance blind area compensation module is used to establish an error compensation model based on the hull vibration, fuse inertial data to predict the position of proximal obstacles, and use a time gating algorithm to isolate self-interference signals.

[0014] Adopting the above technical solution, compared with the prior art, the present invention has the following beneficial effects: The present invention provides an underwater obstacle avoidance system for an unmanned boat based on a distributed wide-beam sonar. The original environmental information is collected by a wide-beam sonar group, and adaptive filtering is used for dynamic noise reduction processing to obtain environmental detection information. The motion state of the unmanned boat is obtained through the navigation compensation module, and motion error compensation is performed on the environmental detection information to obtain environmental correction information. An ellipsoidal space safety region is constructed based on the environmental correction information, and the safe navigation boundary is calculated in real time to obtain safe navigation information. The obstacle avoidance decision-making module generates a multi-level obstacle avoidance strategy according to the real-time navigation state and the safe navigation information. The execution control module converts the strategy into an obstacle avoidance instruction and dynamically adjusts the output frequency to adapt to different navigation conditions. This system effectively solves the problems of insufficient detection accuracy and response lag of traditional obstacle avoidance systems in complex underwater environments, realizes all-round and high-precision control of underwater obstacle avoidance of unmanned boats, and significantly improves the real-time performance and environmental adaptability of the obstacle avoidance system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0016] Figure 1 It is a layout schematic diagram of the distributed wide-beam sonar on the unmanned boat described in the specific implementation manner; Figure 2 It is a schematic diagram of the ellipsoidal space safety area described in the specific implementation manner; Figure 3 It is a schematic diagram of the relationship between the beam angle of the wide-beam sonar and the unmanned boat described in the specific implementation manner; Figure 4 It is a schematic diagram of the dynamic activity space of the unmanned boat and the ellipsoidal space safety area described in the specific implementation manner; Figure 5 It is a schematic diagram for calculating the distribution distance of the wide-beam transmitters described in the specific implementation manner.

[0017] The reference numerals in the drawings are as follows: 1. Unmanned boat; 2. Wide-beam sonar group; 21. Wide-beam transmitter; 22. Wide-beam receiver; 3. Ellipsoidal space safety area; 4. Dynamic activity space; a. Beam angle of the wide-beam transmitter; b. Beam angle of the wide-beam receiver. SPECIFIC IMPLEMENTATION MANNER

[0018] The following will further describe the present invention in detail in conjunction with the drawings and embodiments. It should be specifically noted that the following embodiments are only used to illustrate the present invention, but do not limit the scope of the present invention. Similarly, the following embodiments are only some embodiments of the present invention rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0019] Please refer to Figures 1 to 5, this embodiment provides an underwater obstacle avoidance system for an unmanned boat based on a distributed wide beam sonar, which includes a sonar detection module, a navigation compensation module, a safety domain construction module, an obstacle avoidance decision module, and an execution control module. The sonar detection module includes a wide beam sonar group 2, and each wide beam sonar group 2 includes a wide beam transmitter 21 and a wide beam receiver 22. The sonar detection module is used to collect the original environmental information and perform dynamic noise reduction processing on the original environmental information by using adaptive filtering to obtain environmental detection information; the navigation compensation module is used to obtain the motion state of the unmanned boat 1 and perform motion error compensation on the environmental detection information to obtain environmental correction information; the safety domain construction module is used to construct an ellipsoidal space safety area 3 of the current unmanned boat 1 according to the environmental correction information and calculate the corresponding safe navigation boundary of the ellipsoidal space safety area 3 in real time to obtain safe navigation information; the obstacle avoidance decision module is used to obtain the real-time navigation state and safe navigation information of the current unmanned boat 1 and generate a multi-level obstacle avoidance strategy with priorities; the execution control module is used to convert the multi-level obstacle avoidance strategy into an obstacle avoidance instruction, and the obstacle avoidance instruction includes a heading control instruction and a speed control instruction, and dynamically adjusts the output frequency of the obstacle avoidance instruction to adapt to different navigation conditions.

[0020] In this embodiment, the wide beam sonar group 2 adopts a separated design of sonar detection and transmission, and is composed of a wide beam transmitter 21 (piezoelectric / magnetostrictive / electromagnetic type) and a wide beam receiver 22 that are physically isolated and have a spacing . Preferably, both the beam angle a of the wide beam transmitter and the beam angle b of the wide beam receiver are 150 degrees and meet the condition of not directly irradiating the bottom of the boat. The wide beam transmitter 21 and the wide beam receiver 22 are installed at the bottom of the ship's side according to the layout rules, and the independent transmission circuit is connected to the signal processing circuit including an amplification / filter circuit through a waterproof cable. Among them, is the maximum detection distance of the ultrasonic radar, is the maximum safe area distance to be measured, that is, the distance from the target to the unmanned boat 1, takes a value of 0.4 - 0.6.

[0021] Specifically, please refer to Figure 5 , where C and D in the figure are the positions of the wide beam sonar group 2, E is the target, and F is the intersection point of the closest distance from the target E to the unmanned boat. Preferably, one wide beam receiver 22 and one wide beam transmitter 21 are arranged at the bow and stern of the bottom of the ship's side on both sides of the unmanned boat 1 respectively to achieve separated reception and transmission, and meet the requirements for the spacing between the receiving and transmitting transducers: when the maximum safe area distance is EF and the maximum detection distance of the ultrasonic radar is R, the square of the spacing CD between the receiving and transmitting transducers is less than When a single wide-beam transmitter 21 and a single wide-beam receiver 22 are combined, they must be placed separately; when there are multiple wide-beam receivers 22 or wide-beam transmitters 21, the wide-beam receivers 22 are placed separately from each other, and the wide-beam transmitters 21 are placed separately from each other, and the wide-beam transmitter 21 and the wide-beam receiver 22 can be placed together. The edge emission lines (the lines where the energy decays to 1 / 2 of the maximum) of the beam angle a of the wide-beam transmitter and the beam angle b of the wide-beam receiver should not directly irradiate the bottom of the ship to generate reflected echoes to avoid detection errors. The beam angle should be as large as possible, and the covered solid angle is determined according to the bottom structure of the ship.

[0022] When collecting the original environmental information, an asynchronous same-frequency or synchronous different-frequency modulation signal transmission strategy is adopted. After the receiver detects the reflected sound wave, adaptive filtering based on the NLMS algorithm is used for dynamic noise reduction. The filter characteristics are automatically adjusted according to the real-time reverberation intensity to retain the effective obstacle echoes, and finally the environmental detection information with improved signal-to-noise ratio is output.

[0023] The navigation compensation module obtains the motion state (including roll angle, heading angle and acceleration) of the unmanned boat 1 through a tightly coupled system constructed by the MEMS-IMU and the dual-frequency GNSS receiver, and uses an improved robust UKF algorithm based on SVD for data fusion. The specific implementation of the motion error compensation is as follows: first, the sonar coordinate system is converted to the geodetic coordinate system, and then the beam pointing offset caused by the hull swing is eliminated through the kinematic model, so that the ranging error of the environmental correction information is controlled within 0.3 meters.

[0024] The ellipsoidal space safety region 3 takes the real-time positions of the transmitter and the receiver as the foci, and constructs the shortest propagation distance of the sound wave as the major axis of the ellipsoid in the world coordinate system. The minor axis radius is adaptively adjusted according to the current ship speed. The safe navigation boundary is determined by calculating the minimum distance between the ellipsoid surface and the obstacle point cloud, and the obstacle avoidance strategy is triggered when the boundary threshold is breached.

[0025] During the generation process of the multi-level obstacle avoidance strategy, the priority is divided according to three-dimensional data of the obstacle distance, relative speed and threat level. The heading control instruction is quantified by the deflection angle of the rudder, and the speed control instruction is converted into the percentage of the propulsion motor speed. The output frequencies of both are dynamically adjusted according to the water depth conditions (10 Hz in shallow water areas / 5 Hz in open waters).

[0026] The connection between the execution control module and the power system adopts a hardware-triggered synchronization method to ensure that the time synchronization accuracy is less than 0.1 millisecond, so that the system maintains an overall response delay within 200 ms at a ship speed of 15 knots. All modules are cooperatively controlled through a processing unit with an industrial-grade MCU as the core, and finally the real-time matching between the dynamic activity space 4 of the unmanned boat 1 and the ellipsoidal space safety region 3 is achieved.

[0027] In this embodiment, through the non-directional wide-angle detection characteristics of the distributed wide-beam sonar group 2 and combined with the dynamic noise reduction processing of adaptive filtering, the reliability of environmental detection information in complex underwater environments is effectively improved. The navigation compensation module compensates for motion errors based on the motion state of the unmanned boat 1 to ensure the spatial accuracy of the environmental correction information. The safety domain construction module realizes the precise definition of the safe navigation boundary through the dynamic calculation of the ellipsoidal space safety region 3. The multi-level obstacle avoidance strategies generated by the obstacle avoidance decision module cooperate with the heading control command and the speed control command of the execution control module, enabling the system to adaptively adjust the output frequency according to different navigation conditions, and finally realizing the all-round and high-precision control of the underwater obstacle avoidance of the unmanned boat 1. Each module works in cooperation, while keeping the hardware structure simple, significantly improving the real-time performance and environmental adaptability of the obstacle avoidance system.

[0028] In some embodiments, the sonar detection module is used to collect the original environmental information and perform dynamic noise reduction processing on the original environmental information using adaptive filtering to obtain the environmental detection information, including: Controlling the wide-beam transmitter to emit acoustic wave signals. The wide-beam transmitter uses linear frequency modulation, and the transmission power of the wide-beam transmitter is generated according to the reverberation intensity monitored in real time; Receiving the underwater echo signal through the wide-beam receiver, and performing signal amplification processing, band-pass filtering processing, and analog-to-digital conversion processing on the underwater echo signal to obtain the original environmental information; Performing adaptive filtering processing on the original environmental information, including: Performing primary filtering on the original environmental information using the normalized least mean square algorithm to obtain the first filtered information; Performing secondary filtering on the first filtered information using a convolutional neural network model to obtain the second filtered information; Performing time delay estimation on the second filtered information using the pulse compression algorithm, correcting the earliest echo arrival time, and obtaining the third filtered information; Calculating the obstacle distance according to the sound speed and the third filtered information, and generating environmental detection information including the obstacle position, the obstacle distance, and the obstacle intensity; And, in the case of communication anomalies, temporarily storing the environmental detection information in the local cache, and recording the data acquisition timestamp and the signal quality index.

[0029] In this embodiment, the wide-beam transmitter 21 uses a linear frequency modulation method, and its transmission power is dynamically adjusted based on the real-time reverberation intensity feedback. When the reverberation intensity exceeds the preset threshold, the transmission power is automatically reduced to avoid signal saturation, and vice versa, the power is increased to ensure the detection distance.

[0030] Preferably, the wide-beam receiver 22 achieves echo optimization through a three-stage signal processing chain: for signal amplification processing, a programmable gain amplifier is used, and its gain coefficient is inversely proportional to the received signal strength; the cut-off frequency of the band-pass filtering processing is set according to the frequency band characteristics of the transmitted signal; for analog-to-digital conversion processing, oversampling technology is used to improve the signal-to-noise ratio.

[0031] The adaptive filtering processing of the original environmental information adopts a hierarchical progressive architecture. Among them, the normalized least mean square algorithm suppresses the steady-state noise through variable step parameters, and its convergence speed is adaptively matched with the signal dynamic range; the convolutional neural network model adopts a three-layer convolutional structure, and uses the weight matrix trained offline to extract and filter the features of non-stationary interference; the pulse compression algorithm improves the time-delay estimation accuracy through matched filtering processing, and its compression ratio is proportional to the time-bandwidth product of the transmitted signal. The finally obtained third filtered information calculates the distance of the obstacle through the sound speed-time delay conversion model, where the sound speed value is corrected in real time according to the water temperature and salinity.

[0032] The environmental detection information includes the position, distance and intensity of the obstacle. Preferably, the position of the obstacle is analyzed by the beamforming algorithm, and the intensity of the obstacle is processed by logarithmic compression. The data caching mechanism during communication anomalies adopts a double-buffer design. The data acquisition timestamp is generated by hardware clock synchronization. The signal quality indicators include parameters such as dynamic range and harmonic distortion, providing a reliability criterion for subsequent data fusion.

[0033] In this embodiment, through the dynamic power adjustment of the wide-beam transmitter 21 and the three-stage signal processing chain of the wide-beam receiver 22, it effectively adapts to different reverberation environments; adopts adaptive filtering processing including the normalized least mean square algorithm, convolutional neural network model and pulse compression algorithm to achieve hierarchical suppression of steady-state and non-steady-state noise; accurately analyzes obstacle information through the sound speed-time delay conversion model and beamforming algorithm; the double-buffer design and signal quality indicator recording during communication anomalies ensure data reliability. This embodiment significantly improves the accuracy and robustness of underwater environment detection.

[0034] In some embodiments, the convolutional neural network model is obtained through the following steps: Construct an initial network model and input sample data for training. The sample data includes sample echo signals. The initial network model includes a feature extraction layer and a feature classification layer. The feature extraction layer is used to perform multi-scale feature extraction on the sample echo signals to obtain sample target echoes and sample noise components. The feature classification layer is used to distinguish the sample target echoes from the sample noise components; Repeat training the initial network model until the classification accuracy reaches a preset threshold to obtain the convolutional neural network model.

[0035] In this embodiment, the sample echo signal is the original acoustic signal collected by the wide-beam receiver 22, which contains target reflections and various interferences, and is obtained by combining underwater field measurements and simulation modeling. The feature extraction layer realizes multi-scale analysis through a set of convolution kernels of different sizes. Among them, large-scale convolution kernels are used to extract the macroscopic waveform features of the sample target echo, and small-scale convolution kernels are used to capture the local perturbation patterns of the sample noise components. The feature classification layer adopts a fully connected network structure and outputs the binary discrimination result of the sample target echo and the sample noise components through the sigmoid activation function.

[0036] Preferably, the preset threshold is set according to the misjudgment tolerance of the actual application scenario, and its determination process is evaluated through the recall-precision curve of the validation set. When the balance point of the two reaches the engineering requirements, the training is terminated. Preferably, the repeated training adopts the mini-batch gradient descent method, and the classification accuracy of the validation set is calculated after each iteration. When the accuracy fluctuations of three consecutive training cycles are less than the convergence criterion, the model optimization is completed. The finally obtained convolutional neural network model has its parameters frozen and embedded into the signal processing circuit to perform real-time identification and filtering of noise components in the secondary filtering stage.

[0037] In this embodiment, through the hierarchical analysis of the sample echo signal by the multi-scale feature extraction layer and the precise distinction between the sample target echo and the sample noise components by the feature classification layer, the trained convolutional neural network model has the ability of adaptive noise suppression. The iterative training mechanism controlled by the preset threshold ensures the convergence quality of the model, and the effectively improved classification accuracy finally guarantees the recognition reliability of the underwater target echo signal.

[0038] In some embodiments, the navigation compensation module is used to obtain the motion state of the unmanned boat and perform motion error compensation on the environmental detection information to obtain environmental correction information, including: Collect the real-time attitude data of the unmanned boat through the MEMS inertial measurement unit. The real-time attitude data includes roll angle, pitch angle, and heading angle; Obtain the position information and speed information of the unmanned boat through the dual-frequency GNSS receiver, and construct a tightly coupled navigation solution model with the position information, speed information, and real-time attitude data; Based on the improved robust UKF algorithm, perform data fusion on the tightly coupled navigation solution model, and calculate the motion compensation parameters of the unmanned boat in the world coordinate system. The motion compensation parameters include position offset and attitude correction; Perform coordinate transformation compensation on the environmental detection information according to the motion compensation parameters, including: Establish the conversion relationship between the hull coordinate system and the world coordinate system; Convert the obstacle distance in the environmental detection information from the hull coordinate system to the world coordinate system; Compensate for the sonar measurement error caused by the hull movement; Output environmental correction information containing the accurate positions of obstacles in the world coordinate system; In addition, when the GNSS signal is lost, switch to the pure inertial navigation mode and record the navigation status flag bits.

[0039] In this embodiment, preferably, the MEMS inertial measurement unit adopts an industrial-grade microelectromechanical system (such as the SBG Apogee series), which measures the real-time motion state of the unmanned boat 1 through the combination of a three-axis fiber optic gyroscope and a quartz accelerometer. The roll angle, pitch angle, and heading angle respectively represent the change amounts of the hull's lateral tilt, longitudinal tilt, and azimuth pointing. Their measurement data is output after temperature compensation and zero-bias calibration.

[0040] The dual-frequency GNSS receiver achieves centimeter-level accuracy by receiving L1 / L2 band satellite signals and supporting RTK differential positioning. The position information and velocity information it outputs and the inertial measurement data form complementary observables. Preferably, the tightly coupled navigation solution model processes the GNSS pseudorange / Doppler observations and inertial raw data through a deep combination method. Among them, the position information is used to constrain the cumulative error of inertial navigation, and the velocity information is used to assist in judging the carrier phase cycle slip.

[0041] The improved robust UKF algorithm introduces an adaptive factor based on singular value decomposition (SVD) to adjust the observation weight matrix in the standard unscented Kalman filter framework, automatically reducing its impact on state estimation when GNSS has single, continuous, or mixed observation anomalies. The position offset reflects the deviation between the actual motion trajectory of the unmanned boat 1 and the ideal track, and the attitude correction amount includes the rotation error of the hull coordinate system caused by wind and wave disturbances.

[0042] Preferably, the conversion relationship between the hull coordinate system and the world coordinate system is established through a homogeneous transformation matrix, and its rotation parameters are dynamically updated by real-time attitude data. The conversion result is used to construct the dynamic activity space 4 of the unmanned boat 1.

[0043] The sonar measurement error compensation corrects the echo arrival time through kinematic inverse solution for the beam pointing offset and time delay measurement distortion caused by the hull sway.

[0044] The pure inertial navigation mode relies on the MEMS inertial measurement unit to independently calculate during the GNSS failure period, and its duration is controlled by setting a threshold according to the accuracy of the inertial sensor. The navigation status flag bits are used to identify system working mode switching events and include status parameters such as GNSS availability and inertial navigation error level.

[0045] In this embodiment, through the tightly coupled navigation solution model of the MEMS inertial measurement unit and the dual-frequency GNSS receiver, combined with the data fusion processing of the improved robust UKF algorithm, it is possible to effectively compensate for the measurement errors caused by the motion state of the unmanned boat 1. This solution realizes the accurate conversion of the environmental detection information from the hull coordinate system to the world coordinate system. By calculating the position offset and attitude correction amount, the accuracy of the environmental correction information of the obstacle position is significantly improved. When the GNSS signal is lost, the system can automatically switch to the pure inertial navigation mode and record the navigation status flag bit, ensuring the continuous reliability of the navigation compensation module. This embodiment effectively solves the problem of the interference of the hull motion on sonar measurement and provides an accurate environmental correction information basis for the underwater obstacle avoidance of the unmanned boat 1.

[0046] In some embodiments, the safety domain construction module is used to construct an ellipsoidal space safety region of the current unmanned boat according to the environmental correction information, including: Obtain the obstacle positions in the environmental correction information one by one, and extract the azimuth angle and distance information of the obstacles relative to the unmanned boat; Taking the first actual position of the wide-beam transmitter and the second actual position of the wide-beam receiver at the current moment as the two foci, calculate the ellipsoid parameters based on the time difference of acoustic wave propagation, including: Taking the shortest propagation distance corresponding to the original environmental information of the earliest echo as the major axis of the ellipsoid; Determine the focal length of the ellipsoid according to the distance between the transceiver transducers, and calculate the minor axis of the ellipsoid according to the real-time draft depth of the unmanned boat; Obtain the ellipsoid parameters according to the major axis of the ellipsoid, the minor axis of the ellipsoid and the focal length of the ellipsoid; Construct a dynamic ellipsoidal safety model in the world coordinate system according to the ellipsoid parameters, including: Establish an ellipsoid equation with the transceiver transducers as the two foci; Update the position of the ellipsoidal space according to the real-time navigation data; Visualize the ellipsoidal space safety region.

[0047] In this embodiment, the environmental correction information refers to the obstacle space position data processed by the navigation compensation module, which contains the accurate coordinates of the obstacles in the world coordinate system. The azimuth angle and distance information of the obstacles relative to the unmanned boat 1 are obtained by calculating the arrival time and beam pointing angle of the sonar echo signal. Among them, the azimuth angle represents the included angle relationship between the obstacle and the bow direction of the unmanned boat 1, and the distance information reflects the straight-line interval between the obstacle and the hull.

[0048] The first actual position of the wide beam transmitter 21 and the second actual position of the wide beam receiver 22 refer to the transducer spatial coordinates after hull motion compensation, which are updated in real time by fusing GNSS positioning data and the attitude data of the inertial measurement unit. The time difference of acoustic wave propagation refers to the time interval between the transmitted signal and the earliest echo signal, and this parameter is used to calculate the shortest propagation path of the acoustic wave in the water medium.

[0049] The determination of the major axis of the ellipsoid is based on the acoustic wave propagation distance corresponding to the original environmental information of the earliest echo, and this distance is equal to half of the major axis length. The distance between the transmitting and receiving transducers is used as the determination reference for the focal length of the ellipsoid, and its value is calculated through the spatial geometric relationship of the transducer installation positions.

[0050] The real-time draft depth of the unmanned boat 1 is obtained through a pressure sensor or a draft line detection device, and this parameter is used to constrain the lower limit of the minor axis of the ellipsoid to ensure that the safe area does not include the hull's own structure.

[0051] In the construction process of the dynamic ellipsoid safety model, first, an ellipsoid equation based on two foci is established according to the principles of analytic geometry, and then the continuous update of the model's spatial position is driven by real-time navigation data. Preferably, the visualization of the ellipsoid spatial safety region 3 is realized through three-dimensional graphics rendering technology, and its display range is dynamically adjusted according to the ellipsoid parameters, providing an intuitive spatial safety situation perception for the operator. In this embodiment, the dynamic ellipsoid model accurately represents the safe water area range around the unmanned boat 1, providing a reliable spatial constraint basis for obstacle avoidance decision-making.

[0052] In this embodiment, the first actual position of the wide beam transmitter 21 and the second actual position of the wide beam receiver 22 are used as two foci to construct the ellipsoid spatial safety region 3. Based on the time difference of acoustic wave propagation, the ellipsoid parameters are accurately calculated to achieve the precise positioning of underwater obstacles. The real-time draft depth of the unmanned boat 1 is used to constrain the minor axis of the ellipsoid to ensure that the safe area does not include the hull's own structure. The spatial position of the ellipsoid is dynamically updated through real-time navigation data, and an ellipsoid equation with the transmitting and receiving transducers as two foci is established to form a reliable spatial constraint basis. The visualization of the ellipsoid spatial safety region 3 intuitively displays the safe water area range, providing effective support for obstacle avoidance decision-making.

[0053] In some embodiments, the safe navigation boundary corresponding to the ellipsoid spatial safety region is calculated in real time to obtain safe navigation information, including: Calculating the spatial relationship between the ellipsoid spatial safety region and the hull contour of the unmanned boat to generate an initial safe navigation boundary; According to the initial safe navigation boundary, an initial safe area data set of the ellipsoid spatial safety region is output, and the initial safe area data set includes an initial boundary coordinate set and an initial nearest obstacle warning level; Based on the initial set of boundary coordinates, a three-dimensional spatial grid map is established, and the three-dimensional spatial grid map includes multiple grid cells; Mark each grid cell as a safe area, a warning area, or a dangerous area one by one according to the initial nearest obstacle warning level; Obtain the dynamic parameters of the unmanned boat, and calculate the navigable area according to the dynamic parameters. The dynamic parameters include the current speed, steering angle, roll angle, and pitch angle of the unmanned boat, including: Generate the minimum turning radius according to the current speed and steering angle of the unmanned boat; Generate the influence coefficient of the current minimum turning radius on the actual operable space according to the roll angle and pitch angle; Generate the navigable area of the three-dimensional spatial grid map according to the minimum turning radius, the influence coefficient, and the preset safety margin threshold; Perform multi-level boundary analysis in the navigable area, including: Calculate the minimum safety distance from the ellipsoid surface to the nearest obstacle to obtain the primary boundary; Predict the track envelope within a preset short time interval in the future according to the primary boundary and the current speed to generate the secondary boundary; Generate a dynamic safety corridor according to the primary boundary and the secondary boundary to obtain the safe navigation boundary; And when it is detected that the width of the dynamic safety corridor is less than a preset multiple of the hull width, automatically trigger the speed reduction mode and re-plan the safe navigation boundary; Generate safe navigation information according to the safe navigation boundary. The safe navigation information includes the optimal safe heading suggestion, the maximum allowable speed in each direction, the emergency braking distance parameter, and the multi-level warning status identifier.

[0054] In this embodiment, the initial safe navigation boundary refers to the initial safe range generated by calculating the spatial geometric relationship between the ellipsoidal space safe area 3 and the hull contour of the unmanned boat 1, and this boundary is used to define the basic space range within which the unmanned boat 1 can navigate safely.

[0055] The initial safe area dataset contains the geometric information composed of the initial set of boundary coordinates and the risk assessment information composed of the initial nearest obstacle warning level. Among them, the initial set of boundary coordinates is obtained through spatial geometric calculation, and the initial nearest obstacle warning level is comprehensively determined by the earliest arrival time and intensity of the sonar echo signal, and is specifically divided into three levels: safe, warning, and dangerous.

[0056] The construction of the three-dimensional spatial grid map is based on the world coordinate system. Each grid cell corresponds to the actual water area volume, and its size is determined according to the sonar detection accuracy and the size of the unmanned boat 1.

[0057] The dynamic parameters of the unmanned boat 1 are key parameters reflecting the motion characteristics of the unmanned boat 1, including the current speed, steering angle, roll angle, and pitch angle. These parameters can be collected in real time by on-board sensors and used to calculate the actual operable space.

[0058] The minimum turning radius is calculated through the kinematic relationship between the current speed and the steering angle of the unmanned boat 1, representing the lower limit of the maneuverability of the hull under specific motion states. The influence coefficient of the current minimum turning radius on the actual operable space can be obtained by calculating the dynamic model of the unmanned boat 1, which takes into account the non-linear effects of the roll angle and the pitch angle on the maneuvering performance.

[0059] The navigable area is an actual feasible area comprehensively delimited in a three-dimensional space grid map based on the minimum turning radius, the influence coefficient, and a preset safety margin threshold. This area takes into account the hull dynamics limitations and environmental constraints. Preferably, the preset safety margin threshold is set according to the braking performance and operation response delay of the unmanned boat 1 to ensure that there is enough buffer space in the calculation of the navigable area.

[0060] The prediction of the track envelope in the multi-level boundary analysis adopts a kinematic-based trajectory deduction algorithm, which fully considers the inertial characteristics and hydrodynamic effects of the unmanned boat 1.

[0061] The primary boundary is obtained by calculating the minimum safety distance from the surface of the ellipsoid to the nearest obstacle, reflecting the static safety margin at the current moment. The secondary boundary predicts the track envelope in the future short-time interval based on the primary boundary and the current speed, and is used to evaluate the dynamic navigation risk. The dynamic safety corridor generated by fusing the primary boundary and the secondary boundary has a judgment criterion proportional to the hull width. When the corridor width is insufficient, the speed reduction mode is triggered, and this mode realizes controllable deceleration by gradually reducing the propulsion power.

[0062] The safe navigation information is the basis for the final output of the navigation decision, including key parameters such as the optimal safe heading recommendation. These parameters are generated by comprehensively analyzing the spatial characteristics of the dynamic safety corridor and the hull motion constraints.

[0063] This embodiment provides precise navigation decision support for the unmanned boat 1 through multi-level safety boundary calculation and dynamic risk assessment. Specifically, through the collaborative construction of the initial safe navigation boundary and the three-dimensional space grid map, the navigable area is accurately calculated in combination with the dynamic parameters of the unmanned boat 1. The multi-level boundary analysis is used to generate a dynamic safety corridor, and when it is detected that the corridor width is insufficient, the speed reduction mode is automatically triggered. The finally output safe navigation information provides a comprehensive decision basis for the unmanned boat 1 that takes into account both the static safety margin and the dynamic navigation risk.

[0064] In some embodiments, the obstacle avoidance decision module is used to obtain the real-time navigation state and safe navigation information of the current unmanned boat, and generate a multi-level obstacle avoidance strategy with priorities, including: Build an obstacle avoidance decision tree based on risk assessment. The obstacle avoidance decision tree includes a first-level response decision level, a second-level response decision level, and a third-level response decision level. The first-level response decision level is configured to immediately trigger an emergency braking instruction when the detected collision time is less than 3 seconds. The second-level response decision level is configured to generate a course correction strategy when the width of the dynamic safety corridor is less than 2 times the ship width. The third-level response decision level is configured to select an obstacle avoidance route based on the principle of optimal energy consumption when there are multiple feasible paths. Input the real-time navigation state and safe navigation information of the current unmanned boat into the obstacle avoidance decision tree to obtain an output result, and perform dynamic strategy optimization on the output result, including: Calculate the feasibility index of each obstacle avoidance plan in the output result in real time. The feasibility index includes a path smoothness score, an energy consumption cost score, and a task delay cost score. Use a fuzzy logic algorithm to weight and fuse the feasibility indexes to obtain the comprehensive score of each obstacle avoidance plan. Generate the priority of multiple obstacle avoidance plans according to the comprehensive score, and generate a multi-level obstacle avoidance strategy package. Adaptively adjust the multi-level obstacle avoidance strategy package according to the control characteristics of the unmanned boat, including: Add lateral displacement compensation for flat-bottom boat types. Optimize the steering rate parameters for V-shaped hulls. Dynamically adjust the safety threshold considering the sea state level. When a new obstacle appears during the execution of the strategy, start the strategy replanning mechanism and retain the historical decision log.

[0065] In this embodiment, the obstacle avoidance decision module is the core control unit for comprehensively processing the navigation state and safety information of the unmanned boat 1, and it realizes intelligent obstacle avoidance through a hierarchical decision-making mechanism. The obstacle avoidance decision tree based on risk assessment adopts a three-level response architecture. Among them, the first-level response decision level realizes emergency avoidance through real-time collision time determination, the second-level response decision level implements course correction based on the width of the dynamic safety corridor, and the third-level response decision level performs path optimization based on the principle of optimal energy consumption.

[0066] The real-time navigation state includes the current motion parameters and environmental perception data of the unmanned boat 1. These data and the safe navigation information together constitute the decision input. The feasibility index is obtained by quantitatively evaluating the key performance indicators of the obstacle avoidance plan. Among them, the path smoothness score reflects the continuity of the track curvature, the energy consumption cost score calculates the power consumption of the propulsion system, and the task delay cost score evaluates the travel time loss caused by obstacle avoidance. The fuzzy logic algorithm weighted fusion processes the non-linear relationship of each score through the membership function, and finally outputs the comprehensive score as the basis for plan optimization.

[0067] The generation of the multi-level obstacle avoidance strategy package takes into account the differences in the handling characteristics of different ship types. Among them, the lateral displacement compensation is designed for the cross-drift characteristics of flat-bottomed ship types, and the steering rate parameter optimization adapts to the turning performance of V-shaped hulls. The safety threshold dynamic adjustment mechanism updates the decision boundary in combination with real-time sea condition data, and the strategy replanning mechanism ensures rapid response when encountering new obstacles. At the same time, the historical decision log provides data support for subsequent strategy optimization.

[0068] In this embodiment, through the combination of a multi-level decision-making architecture and a dynamic optimization mechanism, the intelligent generation and real-time adjustment of obstacle avoidance strategies in complex environments are realized. A multi-level response is achieved by constructing an obstacle avoidance decision tree based on risk assessment, and a multi-level obstacle avoidance strategy package with optimized priorities is generated through the fusion of fuzzy logic with the feasibility index. Adaptive adjustments are made according to the characteristics of different ship types to ensure the real-time and reliability of obstacle avoidance decisions in complex environments.

[0069] In some embodiments, the execution control module is used to convert the multi-level obstacle avoidance strategy into obstacle avoidance instructions. The obstacle avoidance instructions include a heading control instruction and a speed control instruction, including: Converting the path planning parameters of each obstacle avoidance path in the multi-level obstacle avoidance strategy into a heading control instruction, including: Extracting the heading angle sequence of the obstacle avoidance path; Using a trajectory smoothing algorithm to generate a continuous steering instruction according to the heading angle sequence; Calculating the actual rudder angle control amount according to the rudder effect characteristics to obtain the heading control instruction; Converting the speed planning parameters of each obstacle avoidance path in the multi-level obstacle avoidance strategy into a speed control instruction, including: Analyzing the recommended speed under the safety distance constraint to generate an acceleration curve that conforms to the propulsion characteristics; Calculating the final thrust control parameter according to the acceleration curve to obtain the speed control instruction; Generating obstacle avoidance instructions according to the heading control instruction and the speed control instruction, and outputting them through a multi-mode output interface. The multi-mode output interface is configured as: Adopting a periodic instruction output mode in the normal cruising mode; Enabling an event-triggered instruction output mode in the emergency obstacle avoidance mode; Switching to a degraded instruction output mode when the system is abnormal; And, performing safety verification processing on the obstacle avoidance instructions before output, including: Verifying the valid range of the instruction parameters of the obstacle avoidance instructions; Limiting the instruction change rate of the obstacle avoidance instructions; Checking the coordination between multiple instructions of the obstacle avoidance instructions.

[0070] In this embodiment, the execution control module realizes precise navigation control by connecting to the power control system of the unmanned boat 1 through a control interface. The path planning parameters in the multi-level obstacle avoidance strategy are converted into instructions through the extraction of the heading angle sequence, where the heading angle sequence represents the set of key turning nodes of the obstacle avoidance path. The trajectory smoothing algorithm is used to eliminate sudden changes in the heading and generate continuous turning instructions that conform to the ship's maneuvering characteristics, and its smoothness is determined according to the maneuvering performance of the unmanned boat 1.

[0071] The conversion process of the speed planning parameters determines the basic speed requirement through the safety distance constraint, and this constraint condition is dynamically adjusted based on the real-time calculation results of the ellipsoidal space safety region 3. The recommended speed is generated according to the current obstacle distribution situation, and its calculation process comprehensively considers the propulsion characteristics and braking performance of the unmanned boat 1. The working mode switching condition of the multi-mode output interface is directly related to the navigation state of the unmanned boat 1, and the response time of the event-triggered instruction output mode needs to meet the real-time requirements of emergency obstacle avoidance.

[0072] The safety verification process ensures the reliability of the instructions through multi-dimensional verification. Among them, the effective range verification of the instruction parameters is set according to the physical limits of the power system of the unmanned boat 1, and the instruction change rate limit parameter is determined according to the dynamic response characteristics of the actuator.

[0073] This embodiment realizes the high-fidelity conversion from the decision-making strategy to the execution action through the refined instruction conversion and multiple safety verification mechanisms. By accurately converting the multi-level obstacle avoidance strategy into heading control instructions and speed control instructions, combining the trajectory smoothing algorithm and the calculation of thrust control parameters, and adopting the multi-mode output interface and safety verification process, it ensures the accurate generation and reliable execution of the obstacle avoidance instructions.

[0074] In some embodiments, the output frequency of the obstacle avoidance instructions is dynamically adjusted to adapt to different navigation conditions, including: Determine the reference output frequency according to the current navigation condition, including: Obtain the real-time navigation state parameters of the unmanned boat; Identify the type of working condition currently in according to the real-time navigation state parameters; Query the working condition-frequency mapping table to determine the reference output frequency; Gradually increase the reference output frequency when the speed in the real-time navigation state parameters increases, and record the frequency adjustment log information; During the adjustment process of the reference output frequency, perform smooth transition processing, including: Set the frequency switching transition interval and adopt a progressive adjustment algorithm to maintain the continuity of the obstacle avoidance instructions during the output process.

[0075] In this embodiment, dynamically adjusting the output frequency of the obstacle avoidance instruction refers to the key mechanism for adaptively adjusting the update rate of the control instruction according to the actual navigation conditions of the unmanned boat 1. The current navigation conditions are identified by analyzing real-time navigation state parameters, which include but are not limited to key indicators characterizing the ship's motion state such as the speed and steering angle. The working condition type refers to different operation modes divided according to the navigation state. Preferably, it is specifically divided into four typical scenarios: low-speed cruise, normal navigation, high-speed navigation, and emergency obstacle avoidance. The division basis mainly considers the different requirements for the timeliness of control response in different speed ranges.

[0076] The working condition-frequency mapping table is used to store the preset values of the reference control frequency corresponding to different working conditions. Its mapping relationship is determined through the dynamic characteristic test of the unmanned boat 1 to ensure that the update frequency of the control instruction and the motion state maintain the best match under each working condition. The reference output frequency is used as the adjustment reference, and each adjustment process is recorded through the frequency adjustment log information, which is convenient for the system to trace and analyze the frequency change rule. When the speed increase is used as the trigger condition, the system gradually increases the output frequency according to the preset rule to ensure that the control instruction is synchronized with the speed change.

[0077] The smooth transition process realizes the seamless switching through the frequency switching transition interval. Optionally, the frequency switching transition interval is set to the duration range of 20%-30% of the difference between the current frequency and the target frequency to ensure that the actuator can smoothly adapt to the frequency change. The progressive adjustment algorithm uses the S-shaped curve transition function to realize the continuity of the frequency change, and calculates the instantaneous frequency value in the transition interval through non-linear interpolation to avoid control jitter caused by sudden frequency changes.

[0078] This embodiment enables the output frequency of the obstacle avoidance instruction to always maintain the best match with the current navigation conditions through the dynamic frequency adjustment mechanism. By dynamically adjusting the output frequency of the obstacle avoidance instruction, the update rate of the control instruction is always adapted to the current navigation conditions, and the continuity of the instruction is ensured by combining the smooth transition process, effectively improving the timeliness of the obstacle avoidance response and the control stability of the unmanned boat 1 under different navigation conditions.

[0079] In some embodiments, the underwater obstacle avoidance system of the unmanned boat based on the distributed wide-beam sonar further includes a reverberation suppression module and a zero-distance blind area compensation module. The reverberation suppression module is used to establish a seabed-water surface multipath reflection channel model, use an adaptive notch filter to suppress the reverberation in a specific frequency band, eliminate false echoes, and adjust the signal pulse width of the transmitted signal to optimize the range resolution; the zero-distance blind area compensation module is used to establish an error compensation model based on the hull vibration, fuse inertial data to predict the position of the proximal obstacle, and use the time gating algorithm to isolate the self-interference signal.

[0080] In this embodiment, the reverberation suppression module analyzes the characteristics of the acoustic wave reflection path by establishing a seabed-water surface multipath reflection channel model, and this model includes the propagation characteristic parameters of the direct wave, the sea surface reflection wave, and the seabed reflection wave. The adaptive notch filter suppresses reverberation in a specific frequency band by dynamically adjusting the filtering characteristics, and its filtering parameters are automatically updated according to the spectrum of the reverberation signal monitored in real time. The signal pulse width adjustment is achieved by a pulse generation circuit, and the transmitted pulse width is dynamically adjusted according to the current detection requirement to optimize the range resolution.

[0081] The zero-range blind area compensation module establishes an error compensation model based on the hull vibration, and this model fuses the attitude data output by the inertial measurement unit and is used to predict the displacement trajectory of the obstacle near the hull. The time gating algorithm isolates the self-interference signal by precisely controlling the signal reception period, and its gating timing is synchronized with the transmitted pulse to ensure the accurate extraction of the effective echo signal. The module realizes the reliable prediction of the position of the obstacle in the zero-range blind area by fusing the sonar measurement data and the inertial navigation data.

[0082] In this embodiment, the adaptive notch filter of the reverberation suppression module and the signal pulse width adjustment effectively reduce the multipath reflection interference. Combining the hull vibration error compensation model and the time gating algorithm of the zero-range blind area compensation module significantly improves the detection accuracy of the near obstacle and the system reliability.

[0083] Adopting the above technical solution, compared with the prior art, the beneficial effects of the present invention are as follows: The present invention effectively improves the reliability of underwater environment detection by the collaborative work of the non-directional wide-angle detection characteristics of the distributed wide-beam sonar group 2 and the adaptive filtering process. The sonar detection module adopts a split design of a wide-beam transmitter 21 and a wide-beam receiver 22, combines the modulation signal transmission strategy of asynchronous same-frequency or synchronous different-frequency, and performs dynamic noise reduction processing through adaptive filtering based on the NLMS algorithm to obtain environment detection information with improved signal-to-noise ratio. The navigation compensation module obtains the motion state of the unmanned boat 1 through a tightly coupled system constructed by a MEMS-IMU and a dual-frequency GNSS receiver, and performs data fusion using an improved robust UKF algorithm based on SVD to compensate for the motion error of the environment detection information and obtain environment correction information. The safety domain construction module constructs an ellipsoidal space safety area 3 of the current unmanned boat 1 according to the environment correction information, and calculates the safe navigation boundary in real time to obtain safe navigation information. The obstacle avoidance decision-making module obtains the real-time navigation state and safe navigation information, and generates a multi-level obstacle avoidance strategy with priorities. The execution control module converts the multi-level obstacle avoidance strategy into an obstacle avoidance instruction including a heading control instruction and a speed control instruction, and dynamically adjusts the output frequency to adapt to different navigation conditions. The reverberation suppression module uses an adaptive notch filter to suppress reverberation in a specific frequency band, and the zero-distance blind area compensation module establishes an error compensation model based on the hull vibration to jointly improve the detection accuracy of proximal obstacles. Each module realizes collaborative control through a processing unit with an industrial-grade MCU as the core, and finally achieves the real-time matching of the dynamic activity space 4 of the unmanned boat 1 and the ellipsoidal space safety area 3.

[0084] In addition, each functional unit in various embodiments of the present invention may be integrated in a processing unit, may exist physically alone for each unit, or two or more units may be integrated in one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of a software functional unit.

[0085] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods in various embodiments of the present invention. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical disks and other various media that can store program codes.

[0086] The above are only some embodiments of the present invention, and thus do not limit the protection scope of the present invention. Any equivalent device or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be similarly included in the patent protection scope of the present invention.

Claims

1. An unmanned surface vehicle underwater obstacle avoidance system based on a distributed wide-beam sonar, characterized in that, Including: A sonar detection module, including a wide-beam sonar group, each wide-beam sonar group including a wide-beam transmitter and a wide-beam receiver. The sonar detection module is used to collect original environmental information and perform dynamic noise reduction processing on the original environmental information using adaptive filtering to obtain environmental detection information; A navigation compensation module, used to obtain the motion state of the unmanned boat and perform motion error compensation on the environmental detection information to obtain environmental correction information; A safety domain construction module, used to construct an ellipsoidal space safety area of the current unmanned boat according to the environmental correction information and calculate the safety navigation boundary corresponding to the ellipsoidal space safety area in real time to obtain safety navigation information; An obstacle avoidance decision-making module, used to obtain the real-time navigation state and safety navigation information of the current unmanned boat and generate a multi-level obstacle avoidance strategy with priorities; An execution control module, used to convert the multi-level obstacle avoidance strategy into an obstacle avoidance instruction, the obstacle avoidance instruction including a heading control instruction and a speed control instruction, and dynamically adjust the output frequency of the obstacle avoidance instruction to adapt to different navigation conditions.

2. The unmanned surface vehicle underwater obstacle avoidance system based on a distributed wide-beam sonar according to claim 1, wherein, The sonar detection module is used to collect original environmental information and perform dynamic noise reduction processing on the original environmental information using adaptive filtering to obtain environmental detection information, including: Controlling the wide-beam transmitter to emit acoustic wave signals. The wide-beam transmitter uses linear frequency modulation, and the transmission power of the wide-beam transmitter is generated according to the reverberation intensity monitored in real time; Receiving underwater echo signals through the wide-beam receiver, and performing signal amplification processing, band-pass filtering processing, and analog-to-digital conversion processing on the underwater echo signals to obtain original environmental information; Performing adaptive filtering processing on the original environmental information, including: Performing primary filtering on the original environmental information using the normalized least mean square algorithm to obtain first filtering information; Performing secondary filtering on the first filtering information using a convolutional neural network model to obtain second filtering information; Performing time delay estimation on the second filtering information using a pulse compression algorithm, correcting the earliest echo arrival time, to obtain third filtering information; Calculating the distance of the obstacle according to the sound speed and the third filtering information, and generating environmental detection information including the position of the obstacle, the distance of the obstacle, and the intensity of the obstacle; And, in the case of communication anomalies, temporarily storing the environmental detection information in a local cache and recording the data acquisition timestamp and the signal quality index.

3. The unmanned surface vehicle underwater obstacle avoidance system based on a distributed wide-beam sonar according to claim 2, wherein The convolutional neural network model is obtained through the following steps of training: Constructing an initial network model and inputting sample data for training. The sample data includes sample echo signals. The initial network model includes a feature extraction layer and a feature classification layer. The feature extraction layer is used to perform multi-scale feature extraction on the sample echo signals to obtain sample target echoes and sample noise components. The feature classification layer is used to distinguish the sample target echoes from the sample noise components; Repeatedly training the initial network model until the classification accuracy reaches a preset threshold to obtain the convolutional neural network model.

4. The unmanned surface vehicle underwater obstacle avoidance system based on a distributed wide-beam sonar according to claim 1, wherein The navigation compensation module is used to obtain the motion state of the unmanned boat and perform motion error compensation on the environmental detection information to obtain environmental correction information, including: Collect the real-time attitude data of the unmanned boat through a MEMS inertial measurement unit, where the real-time attitude data includes roll angle, pitch angle, and heading angle; Obtain the position information and speed information of the unmanned boat through a dual-frequency GNSS receiver, and construct a tightly coupled navigation solution model with the position information, speed information, and the real-time attitude data; Based on an improved robust UKF algorithm, perform data fusion on the tightly coupled navigation solution model, and calculate the motion compensation parameters of the unmanned boat in the world coordinate system, where the motion compensation parameters include position offset and attitude correction; Perform coordinate transformation compensation on the environmental detection information according to the motion compensation parameters, including: Establish the conversion relationship between the hull coordinate system and the world coordinate system; Convert the obstacle distance in the environmental detection information from the hull coordinate system to the world coordinate system; Compensate for the sonar measurement error caused by the hull movement; Output the environmental correction information including the accurate position of the obstacle in the world coordinate system; Moreover, when the GNSS signal is lost, switch to the pure inertial navigation mode and record the navigation status flag bit.

5. The unmanned surface vehicle underwater obstacle avoidance system based on a distributed wide-beam sonar according to claim 1, characterized in that, The safety domain construction module is used to construct the ellipsoidal space safety area of the current unmanned boat according to the environmental correction information, including: Obtain the positions of the obstacles in the environmental correction information one by one, and extract the azimuth angle and distance information of the obstacles relative to the unmanned boat; Taking the first actual position of the wide-beam transmitter and the second actual position of the wide-beam receiver at the current moment as the two foci, calculate the ellipsoid parameters based on the time difference of acoustic wave propagation, including: Take the shortest propagation distance corresponding to the original environmental information of the earliest echo as the major axis of the ellipsoid; Determine the focal length of the ellipsoid according to the distance between the transceiver transducers, and calculate the minor axis of the ellipsoid according to the real-time draft depth of the unmanned boat; Obtain the ellipsoid parameters according to the major axis, minor axis, and focal length of the ellipsoid; Construct a dynamic ellipsoidal safety model in the world coordinate system according to the ellipsoid parameters, including: Establish an ellipsoid equation with the transceiver transducers as the two foci; Update the position of the ellipsoidal space according to the real-time navigation data; Visualize the ellipsoidal space safety area.

6. The unmanned surface vehicle underwater obstacle avoidance system based on a distributed wide-beam sonar according to claim 5, characterized in that, Calculate the safety navigation boundary corresponding to the ellipsoidal space safety area in real time to obtain safety navigation information, including: Calculate the spatial relationship between the ellipsoidal space safety area and the hull contour of the unmanned boat to generate an initial safety navigation boundary; Output the initial safety area dataset of the ellipsoidal space safety area according to the initial safety navigation boundary, where the initial safety area dataset includes an initial boundary coordinate set and an initial nearest obstacle warning level; Based on the initial boundary coordinate set, establish a three-dimensional space grid map, where the three-dimensional space grid map includes multiple grid cells; Mark the grid cells as safety areas, warning areas, or dangerous areas one by one according to the initial nearest obstacle warning level; Obtain the dynamic parameters of the unmanned boat, and calculate the navigable area according to the dynamic parameters, where the dynamic parameters include the current speed, steering angle, roll angle, and pitch angle of the unmanned boat, including: Generate the minimum turning radius according to the current speed and steering angle of the unmanned boat; Generate the influence coefficient of the current minimum turning radius on the actual operable space according to the roll angle and pitch angle; Generate the navigable area of the three-dimensional space grid map based on the minimum turning radius, influence coefficient, and preset safety margin threshold; Perform multi-level boundary analysis in the navigable area, including: Calculate the minimum safety distance from the ellipsoid surface to the nearest obstacle to obtain the primary boundary; Predict the flight path envelope within a preset short time interval in the future based on the primary boundary and the current speed to generate the secondary boundary; Generate a dynamic safety corridor based on the primary boundary and the secondary boundary to obtain the safe navigation boundary; And, when it is detected that the width of the dynamic safety corridor is less than a preset multiple of the hull width, automatically trigger the speed reduction mode and re-plan the safe navigation boundary; Generate safe navigation information based on the safe navigation boundary, and the safe navigation information includes the optimal safe heading suggestion, the maximum allowable speed in each direction, the emergency braking distance parameter, and the multi-level warning status identifier.

7. The unmanned surface vehicle underwater obstacle avoidance system based on a distributed wide-beam sonar according to claim 1, wherein The obstacle avoidance decision-making module is used to obtain the real-time navigation state and safe navigation information of the current unmanned boat, and generate a multi-level obstacle avoidance strategy with priorities, including: Construct an obstacle avoidance decision tree based on risk assessment. The obstacle avoidance decision tree includes a first-level response decision level, a second-level response decision level, and a third-level response decision level. The first-level response decision level is configured to immediately trigger an emergency braking instruction when it is detected that the collision time is less than 3 seconds. The second-level response decision level is configured to generate a heading correction strategy when the width of the dynamic safety corridor is less than 2 times the ship width. The third-level response decision level is configured to select an obstacle avoidance route based on the principle of optimal energy consumption when there are multiple feasible paths; Input the real-time navigation state and safe navigation information of the current unmanned boat into the obstacle avoidance decision tree to obtain the output result, and perform dynamic strategy optimization on the output result, including: Calculate the feasibility index of each obstacle avoidance plan in the output result in real time. The feasibility index includes the path smoothness score, the energy consumption cost score, and the task delay cost score; Use the fuzzy logic algorithm to weight and fuse the feasibility index to obtain the comprehensive score of each obstacle avoidance plan; Generate the priorities of multiple obstacle avoidance plans according to the comprehensive score and generate a multi-level obstacle avoidance strategy package.

8. The unmanned surface vehicle underwater obstacle avoidance system based on a distributed wide-beam sonar according to claim 1, characterized in that, The execution control module is used to convert the multi-level obstacle avoidance strategy into an obstacle avoidance instruction. The obstacle avoidance instruction includes a heading control instruction and a speed control instruction, including: Convert the path planning parameters of each obstacle avoidance path in the multi-level obstacle avoidance strategy into a heading control instruction, including: Extract the heading angle sequence of the obstacle avoidance path; Use the trajectory smoothing algorithm to generate a continuous steering instruction according to the heading angle sequence; Calculate the actual rudder angle control amount according to the rudder effect characteristics to obtain the heading control instruction; Convert the speed planning parameters of each obstacle avoidance path in the multi-level obstacle avoidance strategy into a speed control instruction, including: Analyze the recommended speed under the safety distance constraint to generate an acceleration curve that conforms to the propulsion characteristics; Calculate the final thrust control parameter according to the acceleration curve to obtain the speed control instruction; Generate an obstacle avoidance instruction according to the heading control instruction and the speed control instruction, and output it through a multi-mode output interface. The multi-mode output interface is configured as: Adopt a periodic instruction output mode in the normal cruise mode; Enable the event-triggered instruction output mode in the emergency obstacle avoidance mode; Switch to the degraded instruction output mode in case of system anomalies; And, perform safety verification processing on the obstacle avoidance instructions before output, including: Verify the valid range of the instruction parameters of the obstacle avoidance instructions; Limit the instruction change rate of the obstacle avoidance instructions; Check the coordination between multiple instructions of the obstacle avoidance instructions.

9. The underwater obstacle avoidance system for an unmanned boat based on a distributed wide-beam sonar according to claim 8, characterized in that, Dynamically adjust the output frequency of the obstacle avoidance instructions to adapt to different navigation conditions, including: Determine the reference output frequency according to the current navigation condition, including: Obtain the real-time navigation state parameters of the unmanned boat; Identify the type of the current working condition according to the real-time navigation state parameters; Query the working condition-frequency mapping table to determine the reference output frequency; Gradually increase the reference output frequency when the speed in the real-time navigation state parameters increases, and record the frequency adjustment log information; Perform smooth transition processing during the adjustment of the reference output frequency, including: Set the frequency switching transition interval and adopt a progressive adjustment algorithm to maintain the continuity of the obstacle avoidance instructions during output.

10. The unmanned surface vehicle underwater obstacle avoidance system based on a distributed wide-beam sonar according to claim 1, characterized in that, Also include: A reverberation suppression module, which is used to establish a seabed-water surface multipath reflection channel model, suppress the reverberation in a specific frequency band by using an adaptive notch filter, eliminate false echoes, and adjust the signal pulse width of the transmitted signal to optimize the range resolution; A zero-distance blind area compensation module, which is used to establish an error compensation model based on the hull vibration, fuse inertial data to predict the position of proximal obstacles, and use a time gating algorithm to isolate self-interference signals.

Images