Multi-sensor pulse synchronous control method for unmanned ship

Abstract

The invention provides an unmanned ship multi-sensor pulse synchronous control method, which relates to the technical field of ocean exploration, and comprises the following steps: S1, receiving acoustic sensor working parameters input by a user, and distributing an initial pulse emission time window for each sensor; s2, adopting a field programmable gate array as a control core, receiving a time reference signal calibration clock, dynamically adjusting emission delay of each sensor based on real-time attitude and water depth data, and generating and distributing a trigger signal; and S3, automatically executing a comprehensive coping strategy including alarm, fault-tolerant scheduling and data recording when monitoring that the emission of the sensor is abnormal. By establishing a nanosecond time reference system and combining time-frequency interference evaluation and a cable delay pre-compensation mechanism, accurate synchronization of pulse emission of multiple acoustic sensors is realized, signal crosstalk and superimposed interference of the multiple sensors are reduced, and measurement accuracy and reliability of submarine topography and geomorphy and stratigraphic structure data are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of marine exploration technology, and in particular to a multi-sensor pulse synchronization control method for unmanned surface vessels. Background Technology

[0002] Currently, comprehensive acoustic surveys of seabed topography and shallow structure in nearshore shallow waters primarily rely on work platforms equipped with multi-source acoustic detection systems. However, conventional large vessels, due to their deep draft, cannot easily access shallow waters, while smaller vessels, limited by space and functionality, typically cannot integrate multibeam echo sounders, side-scan sonar, and shallow seismic profilers for coordinated operation. Against this backdrop, unmanned surface vessels (USVs), with their shallow draft, high maneuverability, and low cost, have become crucial platforms for comprehensive, multi-dimensional seabed information detection in nearshore areas. By integrating various acoustic sensors onto a single USV platform, integrated detection of seabed topography, geomorphology, and stratigraphy can be conducted simultaneously. However, when multiple acoustic systems operate simultaneously within the limited space of a USV, crosstalk due to similar operating frequencies and overlapping transmitted signal spectra, as well as superimposed interference during signal propagation caused by a lack of coordination in pulse transmission timing, commonly affects the accuracy of measurement data and the quality of acoustic images. In existing technologies, the independent working sequence of each sensor or a simple time-division multiplexing mechanism is usually relied upon, which lacks comprehensive consideration of dynamic factors such as platform movement and signal propagation delay in complex marine environments and high-precision synchronous control capabilities.

[0003] Regarding the aforementioned and existing related technologies, the inventors believe that the following shortcomings often exist: Existing multi-sensor collaborative measurement schemes for unmanned surface vessels (USVs) mainly suffer from insufficient pulse transmission timing coordination capabilities and a lack of dynamic interference suppression mechanisms. Specifically, this manifests as: a lack of a unified high-precision time reference and centralized synchronization control mechanism, making it difficult to achieve nanosecond-level precision pulse transmission control; failure to consider the real-time impact of platform attitude and water depth changes on the sound wave propagation path, causing the preset timing to fail in actual underwater acoustic environments; and a lack of proactive avoidance strategies for spectral overlap interference and adaptive fault-tolerant scheduling capabilities under abnormal conditions, resulting in insufficient measurement reliability, data quality, and collaborative efficiency of the system in real, dynamic marine operating environments. Summary of the Invention

[0004] The technical problem to be solved by this invention is: how to study interference elimination methods to address the problem of mutual interference of acoustic signals when multiple data acquisition systems are running simultaneously, and improve the performance of unmanned surface vessel (USV) measurement systems. This has become an urgent problem to be solved in the current development of USV measurement systems. To this end, we propose a multi-sensor pulse synchronization control method for USVs.

[0005] To achieve the above objectives, this application adopts the following technical solution: a pulse synchronization control method for multiple acoustic sensors of an unmanned surface vessel (USV), comprising the following steps: S1, receiving acoustic sensor operating parameters input by the user, and allocating an initial pulse emission time window for each acoustic sensor according to the parameters, wherein the acoustic sensors include a multi-beam, side-scan, and shallow-profile system; S2, using a field-programmable gate array (FPGA) as a pulse synchronization control unit, receiving a time reference signal to calibrate the internal clock, and receiving real-time attitude and water depth data of the USV to automatically adjust the pulse emission delay of each acoustic sensor, generating and distributing trigger signals to drive the acoustic sensors to emit pulses in sequence; S3, when an abnormal emission of an acoustic sensor is detected, automatically executing a comprehensive response strategy including alarm, fault-tolerant scheduling, and data recording.

[0006] Further, step S1 includes the following steps: S11, the user inputs the operating parameters of each acoustic sensor, the operating parameters including at least the acoustic sensor model, measurement period, pulse emission delay, signal propagation speed, and operating bandwidth; S12, the received acoustic sensor operating parameters are transmitted through the system communication bus; S13, the received parameters are parsed and stored, and the initial pulse emission time window sequence is calculated according to the scheduling principle of long period first and short period later and the measurement period and bandwidth of each acoustic sensor. The emission time windows of adjacent acoustic sensors are set during the allocation process, wherein a propagation time corresponding to more than 3 times the signal bandwidth is reserved between the emission time windows; S14, the emission time windows of each acoustic sensor generated in S13 are displayed for user confirmation, the final timing scheduling scheme is loaded into its internal memory, and the system enters the ready state, waiting for the start command.

[0007] Furthermore, the calculation of the initial pulse emission time window sequence specifically involves the following operations: obtaining the identification, duty cycle, center frequency and bandwidth, and maximum detection distance of all cooperating acoustic sensors; establishing an optimization model based on multi-objective constrained optimization theory, with the objectives of maximizing time utilization efficiency, minimizing the total system cycle length, and ensuring that the isolation between any two acoustic sensor signals in the time and frequency domains exceeds a safety threshold; the optimization model employs an interference evaluation function to quantify the potential interference risk between any two acoustic sensors; and assigning a suitable emission time offset to each acoustic sensor based on the solution of this model, generating a baseline acoustic sensor timing schedule table, wherein the optimization model formula is: ,in, Acoustic sensor and The overall interference risk value between them , Acoustic sensor and The center frequency, , Acoustic sensor and bandwidth, For frequency domain coupling functions, describing in frequency and of Spectral overlap intensity at that location For time-domain conflict functions, for The launch time for Launch time, for The transmitted signal The propagation delay that causes interference For maximum detection range, These are the weighting coefficients for the time-domain conflict terms. Typical pulse duration, This is the preset protection time interval.

[0008] Furthermore, the calculation of the initial pulse emission time window sequence also takes into account the signal transmission delay caused by the different cable lengths from each acoustic sensor to the transducer. Specifically, the following operations are performed: Before system deployment, the physical length of the cable between each acoustic sensor and its underwater transducer unit is accurately measured, and the length parameter is stored as the acoustic sensor's operating parameter. The pre-stored signal transmission speed in the cable is called up, and the inherent cable transmission delay of each acoustic sensor is calculated according to the formula. When constructing the initial pulse emission time window, the acoustic sensor timing schedule table is updated according to the calculated inherent cable transmission delay as a negative offset.

[0009] Further, step S2 includes: S21, using a field-programmable gate array (FPGA) as a processing unit to continuously receive the standard time signal from the internal GPS / BeiDou dual-mode positioning module, and using it to calibrate the internal oven-controlled crystal oscillator clock source, establishing and maintaining a unified time reference with a stability of 10⁻⁹ for the entire system; S22, using attitude sensors and depth sensors to collect real-time data on the unmanned surface vessel's roll, pitch, and bow angles, as well as real-time depth data, and filtering and calibrating them, then transmitting the processed reliable data in real-time via the communication bus; S23, if the change in attitude or depth data exceeds a preset threshold, such as a 5° change in current attitude or a 1-meter change in depth, using the FPGA to dynamically calculate the acoustic values ​​of each acoustic sensor. The additional time delay deviation of the wave signal due to the change in the propagation path is adjusted in real time according to this deviation, and the timing schedule of the acoustic sensor pulses is updated. S24, strictly according to the calibrated time base and the updated timing schedule, at a predetermined time accurate to nanoseconds, a pulse emission trigger signal for each acoustic sensor is generated, and the trigger signal is distributed in parallel to the acoustic sensor adapter module through the high-speed bus. S25, the acoustic sensor adapter module performs level conversion and impedance matching on the trigger signal through its dedicated signal conditioning circuit to form a stable emission control signal compatible with the electrical interface of each acoustic sensor, and outputs it to each acoustic sensor to drive it to emit pulses accurately according to the adjusted timing.

[0010] Furthermore, the use of a field-programmable gate array (FPGA) as the pulse synchronization control unit specifically performs the following operations: the FPGA receives the standard time signal from the BeiDou dual-mode positioning module of the Global Positioning System (GPS), calibrates the oven-controlled crystal oscillator to the system's unified time reference through a digital phase-locked loop (PLL), executes a parallel triggering algorithm based on the calibrated clock signal, marks the precise time for each triggering event using a hardware timestamp mechanism, and independently configures a dedicated delay compensator and pulse generator for each acoustic sensor channel. Multiple nanosecond-level precision pulse trigger signals are simultaneously generated through hardware logic. Each trigger signal is synchronously distributed to the corresponding acoustic sensor control module with a synchronization accuracy of 10 nanoseconds through a high-speed peripheral component interconnect bus, and microsecond-level dynamic reconfiguration of the timing strategy is achieved through register mapping.

[0011] Further, step S23 specifically performs the following operations: receiving filtered real-time attitude data and water depth data provided by the dynamic monitoring module; when the detected attitude or water depth change exceeds a set threshold, executing a dynamic compensation algorithm based on the time-varying characteristics of the propagation path; calculating the precise compensation time of each acoustic sensor by solving the time-varying integral equation; wherein the module calculates the spatial position vector of the acoustic sensor under wave disturbance in real time through coordinate transformation, calculates the propagation characteristics of sound waves in the time-varying medium by coupling sound velocity profile disturbance, predicts pulse propagation behavior by analyzing path change acceleration to achieve feedforward compensation, and applies the calculated compensation time to the reconfiguration of the field-programmable gate array timing scheduler to complete the nanosecond-level precise adjustment of the pulse emission time of each acoustic sensor; wherein the compensation time calculation formula is:

[0012] ;

[0013] Acoustic sensor The final compensation time amount obtained from the calculation; To calculate the start time; For system control cycle; Based on water depth The reference speed of sound as it changes over time; The sound velocity profile disturbance is calculated from real-time water temperature and salinity data; It is a time-varying rotation matrix composed of roll, pitch, and yaw data; Acoustic sensor Initial installation position vector in the hull coordinate system; The translation vector is composed of the displacement of the unmanned surface vessel's center of mass and the change in water depth. This is the gain coefficient for path change acceleration; This refers to the acceleration due to path change.

[0014] Further, step S3 includes the following steps: S31, real-time monitoring and acquisition of transmission status confirmation signals from each acoustic sensor; S32, comparison of the received transmission status confirmation signals with the internal timing schedule table to determine whether any acoustic sensor has failed to provide a valid feedback signal within a preset time window, thereby identifying acoustic sensors with transmission anomalies; S33, upon identification of an anomaly, sending an alarm signal containing the ID of the abnormal acoustic sensor, the anomaly type, and the time of occurrence, and activating an audible and visual alarm to notify the user and display detailed alarm information; S34, the system immediately triggers a three-level fault-tolerant response mechanism, writes all fault events and scheduling adjustment records into a non-volatile memory, and updates the system topology status diagram in real time; S35, timestamps all relevant data of this abnormal event and stores it in its integrated storage module.

[0015] Furthermore, the system's adaptive fault-tolerant processing three-level fault-tolerant response mechanism includes: The first level response involves querying a pre-stored backup window mapping table based on the type of the faulty acoustic sensor and the current system operating mode, automatically activating the corresponding backup transmission time window, and attempting to re-trigger the acoustic sensor; the second level response involves recalculating the time window allocation for the remaining normal acoustic sensors using an isolation algorithm based on dynamic resource planning after consecutive retry failures, maintaining the overall system sampling rate by dynamically compressing the protection interval and optimizing the transmission sequence; the third level response involves initiating a game theory-based multi-agent resource allocation model when multiple acoustic sensors fail simultaneously, determining the optimal acoustic sensor scheduling strategy by solving for the Nash equilibrium point; wherein, the isolation algorithm formula is: ;in, Acoustic sensor At any moment The proportion of time slot resources obtained through reallocation; Acoustic sensor Data quality assessment coefficient within the current task cycle; Acoustic sensor Real-time task priority weights; This represents the total number of acoustic sensors that are functioning normally in the system. This is the initial timestamp at which the system detected the fault. The system's fault-tolerant response time constant; Acoustic sensor In time The load dynamic change function; This is the time integration variable.

[0016] Further, step S35 specifically performs the following operations: The system packages the abnormal timestamp, acoustic sensor ID, abnormal type code, and corresponding environmental status data into a structured log and stores it in a non-volatile storage module. It then periodically performs offline data analysis. By performing time-series-based pattern mining and association rule analysis on historical abnormal records, an abnormal prediction model is established. Based on the output of this model, the health status scores of each acoustic sensor are dynamically adjusted. When the failure probability of a specific acoustic sensor or combination of acoustic sensors exceeds a preset threshold, the system automatically updates the parameter configuration in the fault tolerance strategy library, including optimizing the allocation scheme of the backup time window, adjusting the weight coefficient of the dynamic resource planning algorithm, and correcting the sensitivity threshold of abnormal judgment. These verified optimized parameters are then loaded into the time-series scheduling table through online updates.

[0017] The technical effects and advantages of this invention are as follows: 1. By combining a high-precision synchronous control module with GPS / BeiDou dual-mode positioning and a constant-temperature crystal oscillator, a nanosecond-level time reference system is established. Based on the FPGA hardware parallel processing architecture, multi-channel synchronous trigger control is implemented, improving the synchronization accuracy of multi-sensor pulse transmission to the 10-nanosecond level. This ensures complete isolation of acoustic signals from multi-beam, side-scan, and shallow profiling systems during propagation, eliminates signal crosstalk, and significantly reduces noise and stripe interference in acoustic images, thereby significantly improving the accuracy and reliability of seabed topography and stratigraphic structure data. 2. Based on a real-time timing adjustment mechanism using dynamic monitoring data, the system collects platform motion and environmental data in real time through attitude sensors and depth sensors. When the attitude change exceeds ±5° or the depth change exceeds 1 meter, the system automatically calculates the propagation path change and adjusts the transmission delay, solving the synchronization deviation problem caused by platform sway and depth changes, and ensuring the stable operation of the system in complex marine environments.

[0018] 3. Full-path synchronous control and signal integrity assurance: Precise timing management is implemented throughout the entire link from electrical triggering to acoustic emission. The dedicated signal conditioning circuit of the sensor adapter module ensures stable transmission of the trigger signal and controls the signal transmission error within ±1 nanosecond, enabling true synchronous emission of the transducer ends of each sensor. This provides an industry-leading synchronous control solution for comprehensive marine surveys of unmanned surface vessels. Attached Figure Description

[0019] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts:

[0020] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system block diagram of the present invention. Detailed Implementation

[0021] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0022] Reference Figure 1 , Figure 2As shown, the present invention provides a technical solution: a pulse synchronization control method for multiple acoustic sensors of an unmanned surface vessel (USV), comprising the following steps: S1, receiving acoustic sensor operating parameters input by the user, and allocating an initial pulse emission time window for each acoustic sensor according to the parameters. The acoustic sensors include a multi-beam, side-scan, and shallow-profile system. S2, using a field-programmable gate array (FPGA) as the pulse synchronization control unit, receiving a time reference signal to calibrate the internal clock, and receiving real-time attitude and water depth data of the USV to automatically adjust the pulse emission delay of each acoustic sensor, generating and distributing trigger signals to drive the acoustic sensors to emit pulses in sequence. S3, when an abnormal emission of an acoustic sensor is detected, automatically executing a comprehensive response strategy including alarm, fault-tolerant scheduling, and data recording.

[0023] Further, step S1 includes the following steps: S11, the user inputs the operating parameters of each acoustic sensor, the operating parameters including at least the acoustic sensor model, measurement period, pulse emission delay, signal propagation speed, and operating bandwidth; S12, the received acoustic sensor operating parameters are transmitted through the system communication bus, the bus including RS485, EtherCAT, or CAN bus; S13, the received parameters are parsed and stored, and the initial pulse emission time window sequence is calculated according to the scheduling principle of long period first and short period later and the measurement period and bandwidth of each acoustic sensor. The emission time windows of adjacent acoustic sensors are set during the allocation process, wherein the propagation time corresponding to more than 3 times the signal bandwidth is reserved between the emission time windows; S14, the emission time windows of each acoustic sensor generated in S13 are displayed for user confirmation, the final timing scheduling scheme is loaded into its internal memory, and the system enters the ready state, waiting for the start command.

[0024] Furthermore, step S11 also includes automatic sensor identification and parameter verification: by reading the identifiers of each sensor device and comparing them with a preset database, the accuracy and compatibility of parameter configuration are ensured, and synchronization failure caused by parameter misconfiguration is avoided.

[0025] Furthermore, step S12 specifically performs the following operations: The user-confirmed parameters are packaged into a data packet with a specific frame format. This frame format includes a start bit, sensor ID, parameter type, parameter value, and CRC cyclic redundancy check code. Based on the preset sensor type and interface protocol, the data packet is distributed to the corresponding physical communication port via a multi-protocol communication controller. For example, an EtherCAT bus is allocated for multi-beam sonar, and an RS485 bus is allocated for side-scan sonar. During transmission, the bus status is continuously monitored to ensure the integrity of data transmission. Upon receiving the packet, a CRC check is immediately performed. If the check fails, a retransmission is requested through the human-machine interface module until all parameters are accurately received and stored in the specified register address, laying a reliable data foundation for subsequent parameter parsing and strategy generation.

[0026] Furthermore, the calculation of the initial pulse emission time window sequence specifically involves the following operations: obtaining the identification, duty cycle, center frequency and bandwidth, and maximum detection distance of all cooperating acoustic sensors; establishing an optimization model based on multi-objective constrained optimization theory, with the objectives of maximizing time utilization efficiency, minimizing the total system cycle length, and ensuring that the isolation between any two acoustic sensor signals in the time and frequency domains exceeds a safety threshold; the optimization model employs an interference evaluation function to quantify the potential interference risk between any two acoustic sensors; and assigning a suitable emission time offset to each acoustic sensor based on the solution of this model, generating a baseline acoustic sensor timing schedule table, wherein the optimization model formula is:

[0027] ,in, Acoustic sensor and The overall interference risk value between them , Acoustic sensor and The center frequency, , Acoustic sensor and bandwidth, For frequency domain coupling functions, describing in frequency and of Spectral overlap intensity at that location For time-domain conflict functions, for The launch time for Launch time, for The transmitted signal The propagation delay that causes interference For maximum detection range, These are the weighting coefficients for the time-domain conflict terms. Typical pulse duration, This is the preset protection time interval.

[0028] Furthermore, the calculation of the initial pulse emission time window sequence also takes into account the signal transmission delay caused by the different cable lengths from each acoustic sensor to the transducer. Specifically, the following operations are performed: Before system deployment, the physical length of the cable from the output of the high-precision synchronization control module to its underwater transducer unit of each acoustic sensor is accurately measured, and the length parameter is stored as the acoustic sensor's operating parameter. The pre-stored signal transmission speed in the cable is then called, which is typically 2 × 10⁻⁶. 8 The inherent cable transmission delay of each acoustic sensor is calculated according to the formula. When constructing the initial pulse transmission time window, the acoustic sensor timing schedule table is updated based on the calculated inherent cable transmission delay as a negative offset.

[0029] Further, step S2 includes: S21, using a field-programmable gate array (FPGA) as a processing unit to continuously receive the standard time signal from the internal GPS / BeiDou dual-mode positioning module, and using it to calibrate the internal oven-controlled crystal oscillator clock source to establish and maintain a unified time reference with a stability of 10⁻⁹ for the entire system; S22, using attitude sensors and depth sensors to collect real-time data on the unmanned surface vessel's roll, pitch, and bow angles, as well as real-time depth data, and filtering and calibrating them, then transmitting the processed reliable data in real-time via a communication bus; S23, if the change in attitude or depth data exceeds a preset threshold, based on the current attitude and depth, using the FPGA to dynamically calculate the propagation path of the acoustic signals from each acoustic sensor. The additional time delay deviation caused by the change is adjusted in real time according to this deviation, and the timing schedule table of the acoustic sensors is updated. S24, strictly according to the calibrated time base and the updated timing schedule table, at a predetermined time accurate to nanoseconds, a pulse emission trigger signal for each acoustic sensor is generated, and the trigger signal is distributed in parallel to the acoustic sensor adapter module through a high-speed bus including the PCIe bus. S25, the acoustic sensor adapter module performs level conversion and impedance matching on the trigger signal through its dedicated signal conditioning circuit to form a stable emission control signal compatible with the electrical interface of each acoustic sensor, and outputs it to each acoustic sensor to drive it to emit pulses accurately according to the adjusted timing.

[0030] Furthermore, the use of a field-programmable gate array (FPGA) as the pulse synchronization control unit includes: the FPGA internally employs a multi-stage pipeline and parallel processing architecture, and implements signal generation logic accurate to the clock cycle through a hardware description language to ensure stable synchronization performance even under the worst sea conditions. It receives the standard time signal from the BeiDou dual-mode positioning module of the Global Positioning System, calibrates the oven-controlled crystal oscillator to the system's unified time reference through a digital phase-locked loop, executes a parallel triggering algorithm based on the calibrated clock signal, marks the precise time for each triggering event using a hardware timestamp mechanism, and independently configures a dedicated delay compensator and pulse generator for each acoustic sensor channel. It simultaneously generates multiple nanosecond-level precision pulse trigger signals through hardware logic, synchronously distributes each trigger signal to the corresponding acoustic sensor control module with 10 nanosecond-level synchronization accuracy through a high-speed peripheral component interconnect bus, and realizes microsecond-level dynamic reconfiguration of timing strategies through register mapping.

[0031] Further, step S23 specifically performs the following operations: receiving filtered real-time attitude data and water depth data provided by the dynamic monitoring module; when the detected attitude or water depth change exceeds a set threshold, executing a dynamic compensation algorithm based on the time-varying characteristics of the propagation path; calculating the precise compensation time of each acoustic sensor by solving the time-varying integral equation; wherein the module calculates the spatial position vector of the acoustic sensor under wave disturbance in real time through coordinate transformation, calculates the propagation characteristics of sound waves in the time-varying medium by coupling sound velocity profile disturbance, predicts pulse propagation behavior by analyzing path change acceleration to achieve feedforward compensation, and applies the calculated compensation time to the reconfiguration of the field-programmable gate array timing scheduler to complete the nanosecond-level precise adjustment of the pulse emission time of each acoustic sensor; wherein the compensation time calculation formula is: , Acoustic sensor The final compensation time amount is calculated. To calculate the start time, For the system control cycle, Based on water depth The reference speed of sound as a function of time. The sound velocity profile disturbance is calculated from real-time water temperature and salinity data. It is a time-varying rotation matrix composed of roll, pitch, and bow data. Acoustic sensor Initial installation position vector in the hull coordinate system The translation vector is composed of the displacement of the unmanned surface vessel's center of mass and the change in water depth. This is the path change acceleration gain coefficient. This refers to the acceleration due to path change.

[0032] Furthermore, step S25 specifically performs the following operations: After the field-programmable gate array (FPGA) emits a nanosecond-level trigger signal, the level conversion chip boosts the low-voltage LVCMOS signal output by the FPGA to the drive level required by each acoustic sensor, such as 3.3... The LVCMOS signal has been improved to ±12 or ±24 The drive level is adjusted, and simultaneously, using an impedance matching network typically composed of precision resistors and capacitors, the signal output impedance is adjusted to match the impedance of the transmission cable and the sensor input impedance, such as adjusting it to 50. To minimize signal reflection and distortion, the conditioned high-voltage, low-impedance drive signals are transmitted to the transmitters of the multi-beam, side-scan, and shallow-profile systems via shielded cables. The transducer units of each acoustic sensor generate a powerful and steep acoustic pulse the instant they receive the electrical pulse, and then complete the synchronous transmission of underwater sound waves according to the updated timing schedule.

[0033] Further, step S3 includes the following steps: S31, real-time monitoring and acquisition of the transmission status confirmation signals fed back by each acoustic sensor; S32, comparison of the received transmission status confirmation signals with the internal timing schedule table to determine whether any acoustic sensor has failed to provide a valid feedback signal within a preset time window, thereby identifying the acoustic sensor with transmission abnormality; S33, when an abnormality is identified, sending an alarm signal containing the abnormal acoustic sensor ID, abnormality type, and occurrence time, and activating an audible and visual alarm to notify the user and display detailed alarm prompt information; S34, the system immediately triggers a three-level fault-tolerant response mechanism, writes all fault events and scheduling adjustment records into a non-volatile memory, and updates the system topology status diagram in real time; S35, timestamps all relevant data of this abnormal event and stores it in its integrated storage module.

[0034] Furthermore, the system's adaptive fault-tolerant processing three-level fault-tolerant response mechanism includes: The first level response involves querying a pre-stored backup window mapping table based on the type of the faulty acoustic sensor and the current system operating mode, automatically activating the corresponding backup transmission time window, and attempting to re-trigger the acoustic sensor; the second level response involves recalculating the time window allocation for the remaining normal acoustic sensors using an isolation algorithm based on dynamic resource planning after consecutive retry failures, maintaining the overall system sampling rate by dynamically compressing the protection interval and optimizing the transmission sequence; the third level response involves initiating a game theory-based multi-agent resource allocation model when multiple acoustic sensors fail simultaneously, determining the optimal acoustic sensor scheduling strategy by solving for the Nash equilibrium point. The isolation algorithm formula is: ,in, Acoustic sensor At any moment The proportion of time slot resources obtained through reallocation Acoustic sensor The data quality assessment coefficient within the current task cycle. Acoustic sensor Real-time task priority weights This represents the total number of acoustic sensors that are functioning normally in the system. The initial timestamp at which the system detected the fault. The system's fault-tolerant response time constant. Acoustic sensor In time The load dynamic change function, This is the time integration variable.

[0035] Further, step S35 specifically performs the following operations: The system packages the abnormal timestamp, acoustic sensor ID, abnormal type code, and corresponding environmental status data into a structured log and stores it in a non-volatile storage module. It then periodically performs offline data analysis. By performing time-series-based pattern mining and association rule analysis on historical abnormal records, an abnormal prediction model is established. Based on the output of this model, the health status scores of each acoustic sensor are dynamically adjusted. When the failure probability of a specific acoustic sensor or combination of acoustic sensors exceeds a preset threshold, the system automatically updates the parameter configuration in the fault tolerance strategy library, including optimizing the allocation scheme of the backup time window, adjusting the weight coefficient of the dynamic resource planning algorithm, and correcting the sensitivity threshold of abnormal judgment. These verified optimized parameters are then loaded into the time-series scheduling table through online updates.

[0036] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A method for multi-acoustic sensor pulse synchronization control of an unmanned surface vessel, characterized in that, Includes the following steps: S1: Receives user-input acoustic sensor operating parameters and allocates an initial pulse emission time window for each acoustic sensor based on these parameters. The acoustic sensors include multi-beam, side-scan, and shallow-profile systems. S2: Employs a field-programmable gate array (FPGA) as the pulse synchronization control unit. It receives a time reference signal to calibrate the internal clock and receives real-time attitude and depth data from the unmanned surface vessel to automatically adjust the pulse emission delay of each acoustic sensor. It then generates and distributes trigger signals to drive the acoustic sensors to emit pulses sequentially. S3: When an abnormal emission from an acoustic sensor is detected, it automatically executes a comprehensive response strategy including alarm, fault-tolerant scheduling, and data logging.

2. The method for multi-acoustic sensor pulse synchronization control of an unmanned surface vessel according to claim 1, characterized in that: Step S1 includes the following steps: S11, the user inputs the operating parameters of each acoustic sensor, including at least the acoustic sensor model, measurement period, pulse emission delay, signal propagation speed, and operating bandwidth; S12, the received acoustic sensor operating parameters are transmitted through the system communication bus; S13, the received parameters are parsed and stored, and the initial pulse emission time window sequence is calculated according to the scheduling principle of long period first and short period later and the measurement period and bandwidth of each acoustic sensor. The emission time windows of adjacent acoustic sensors are set during the allocation process, wherein a propagation time corresponding to more than 3 times the signal bandwidth is reserved between the emission time windows; S14, the emission time windows of each acoustic sensor generated in S13 are displayed for user confirmation, the final timing scheduling scheme is loaded into its internal memory, and the system enters a ready state, waiting for the start command.

3. The method for multi-acoustic sensor pulse synchronization control of an unmanned surface vessel according to claim 2, characterized in that, The calculation of the initial pulse emission time window sequence specifically involves the following operations: obtaining the identification, duty cycle, center frequency and bandwidth, and maximum detection distance of all cooperating acoustic sensors; establishing an optimization model based on multi-objective constrained optimization theory, with the objectives of maximizing time utilization efficiency, minimizing the total system cycle length, and ensuring that the isolation between any two acoustic sensor signals in the time and frequency domains exceeds a safety threshold; the optimization model uses an interference evaluation function to quantify the potential interference risk between any two acoustic sensors; and assigning a suitable emission time offset to each acoustic sensor based on the solution of this model, generating a baseline acoustic sensor timing schedule table. The optimization model formula is as follows: ,in, Acoustic sensor and The overall interference risk value between them , Acoustic sensor and The center frequency, , Acoustic sensor and bandwidth, For frequency domain coupling functions, describing in frequency and of Spectral overlap intensity at that location For time-domain conflict functions, for The launch time for Launch time, for The transmitted signal The propagation delay that causes interference For maximum detection range, These are the weighting coefficients for the time-domain conflict terms. Typical pulse duration, This is the preset protection time interval.

4. The method for multi-acoustic sensor pulse synchronization control of an unmanned surface vessel according to claim 2, characterized in that, The calculation of the initial pulse emission time window sequence also takes into account the signal transmission delay caused by the different cable lengths from each acoustic sensor to the transducer. Specifically, the following operations are performed: Before system deployment, the physical length of the cable between each acoustic sensor and its underwater transducer unit is accurately measured, and the length parameter is stored as the acoustic sensor's operating parameter. The pre-stored signal transmission speed in the cable is called up, and the inherent cable transmission delay of each acoustic sensor is calculated according to the formula. When constructing the initial pulse emission time window, the acoustic sensor timing schedule table is updated according to the calculated inherent cable transmission delay as a negative offset.

5. The method for multi-acoustic sensor pulse synchronization control of an unmanned surface vessel according to claim 1, characterized in that: Step S2 includes: S21, using a field-programmable gate array (FPGA) as a processing unit to continuously receive the standard time signal from the internal GPS / BeiDou dual-mode positioning module, and using it to calibrate the internal oven-controlled crystal oscillator clock source to establish and maintain a unified time reference with a stability of 10⁻⁹ for the entire system; S22, using attitude sensors and depth sensors to collect real-time data on the unmanned surface vessel's roll, pitch, and bow angles, as well as real-time depth data, and filtering and calibrating them, then transmitting the processed reliable data in real-time via the communication bus; S23, if the change in attitude or depth data exceeds a preset threshold, based on the current attitude and depth, using the FPGA to dynamically calculate the propagation time of the acoustic signals from each acoustic sensor. The additional time delay deviation caused by path changes is adjusted in real time according to this deviation, and the timing schedule of each acoustic sensor is updated. S24, strictly according to the calibrated time base and the updated timing schedule, at a predetermined time accurate to nanoseconds, a pulse emission trigger signal for each acoustic sensor is generated, and the trigger signal is distributed in parallel to the acoustic sensor adapter module through a high-speed bus. S25, the acoustic sensor adapter module performs level conversion and impedance matching on the trigger signal through its dedicated signal conditioning circuit to form a stable emission control signal compatible with the electrical interface of each acoustic sensor, and outputs it to each acoustic sensor to drive it to emit pulses accurately according to the adjusted timing.

6. The method for multi-acoustic sensor pulse synchronization control of an unmanned surface vessel according to claim 1, characterized in that, The above-mentioned field-programmable gate array (FPGA) is used as the pulse synchronization control unit, and specifically performs the following operations: The FPGA receives the standard time signal from the BeiDou dual-mode positioning module of the Global Positioning System (GPS), calibrates the oven-controlled crystal oscillator to the system's unified time reference through a digital phase-locked loop (PLL), executes a parallel triggering algorithm based on the calibrated clock signal, marks the precise time for each triggering event using a hardware timestamp mechanism, and independently configures a dedicated delay compensator and pulse generator for each acoustic sensor channel. Multiple nanosecond-level precision pulse trigger signals are generated simultaneously through hardware logic. The trigger signals are synchronously distributed to the corresponding acoustic sensor control modules with a synchronization accuracy of 10 nanoseconds through a high-speed peripheral component interconnect bus, and the timing strategy is dynamically reconfigured at the microsecond level through register mapping.

7. The method for multi-acoustic sensor pulse synchronization control of an unmanned surface vessel according to claim 5, characterized in that: Step S23 specifically involves the following operations: receiving filtered real-time attitude and water depth data provided by the dynamic monitoring module; when a change in attitude or water depth exceeds a set threshold, executing a dynamic compensation algorithm based on the time-varying characteristics of the propagation path; calculating the precise compensation time of each acoustic sensor by solving the time-varying integral equation; wherein the module calculates the spatial position vector of the acoustic sensor under wave disturbance in real time through coordinate transformation, calculates the propagation characteristics of sound waves in the time-varying medium by coupling sound velocity profile disturbance, predicts pulse propagation behavior by analyzing path change acceleration to achieve feedforward compensation, and applies the calculated compensation time to the reconfiguration of the field-programmable gate array timing scheduler to complete the nanosecond-level precise adjustment of the pulse emission time of each acoustic sensor; wherein the compensation time calculation formula is: , Acoustic sensor The final compensation time amount is calculated. To calculate the start time, For the system control cycle, Based on water depth The reference speed of sound as a function of time. The sound velocity profile disturbance is calculated from real-time water temperature and salinity data. It is a time-varying rotation matrix composed of roll, pitch, and bow data. Acoustic sensor Initial installation position vector in the hull coordinate system The translation vector is composed of the displacement of the unmanned surface vessel's center of mass and the change in water depth. This is the path change acceleration gain coefficient. This refers to the acceleration due to path change.

8. The method for multi-acoustic sensor pulse synchronization control of an unmanned surface vessel according to claim 1, characterized in that: Step S3 includes the following steps: S31, real-time monitoring and acquisition of transmission status confirmation signals from each acoustic sensor; S32, comparison of the received transmission status confirmation signals with the internal timing schedule table to determine if any acoustic sensor has failed to provide a valid feedback signal within a preset time window, thereby identifying acoustic sensors with transmission anomalies; S33, upon identification of an anomaly, sending an alarm signal containing the abnormal acoustic sensor ID, anomaly type, and occurrence time, and activating an audible and visual alarm to notify the user and display detailed alarm information; S34, the system immediately triggers a three-level fault-tolerant response mechanism, writes all fault events and scheduling adjustment records into a non-volatile memory, and updates the system topology status diagram in real time; S35, timestamps all relevant data of this abnormal event and stores it in its integrated storage module.

9. The method for multi-acoustic sensor pulse synchronization control of an unmanned surface vessel according to claim 8, characterized in that, The system's adaptive fault-tolerant processing three-level fault-tolerant response mechanism includes: The first level response involves querying a pre-stored backup window mapping table based on the type of the faulty acoustic sensor and the current system operating mode, automatically activating the corresponding backup transmission time window, and attempting to re-trigger the acoustic sensor; the second level response involves recalculating the time window allocation for the remaining normal acoustic sensors using an isolation algorithm based on dynamic resource planning after consecutive retry failures, maintaining the overall system sampling rate by dynamically compressing the protection interval and optimizing the transmission sequence; the third level response involves initiating a game theory-based multi-agent resource allocation model when multiple acoustic sensors fail simultaneously, determining the optimal acoustic sensor scheduling strategy by solving for the Nash equilibrium point; where the isolation algorithm formula is: ,in, Acoustic sensor At any moment The proportion of time slot resources obtained through reallocation Acoustic sensor The data quality assessment coefficient within the current task cycle. Acoustic sensor Real-time task priority weights This represents the total number of acoustic sensors that are functioning normally in the system. The initial timestamp at which the system detected the fault. The system's fault-tolerant response time constant. Acoustic sensor In time The load dynamic change function, This is the time integration variable.

10. The method for multi-acoustic sensor pulse synchronization control of an unmanned surface vessel according to claim 8, characterized in that: Step S35 specifically involves the following operations: The system packages the abnormal timestamp, acoustic sensor ID, abnormal type code, and corresponding environmental status data into a structured log and stores it in a non-volatile storage module. It then periodically performs offline data analysis. By performing time-series-based pattern mining and association rule analysis on historical abnormal records, an abnormal prediction model is established. Based on the output of this model, the health status scores of each acoustic sensor are dynamically adjusted. When the failure probability of a specific acoustic sensor or combination of acoustic sensors exceeds a preset threshold, the system automatically updates the parameter configuration in the fault tolerance strategy library, including optimizing the allocation scheme of the backup time window, adjusting the weight coefficient of the dynamic resource planning algorithm, and correcting the sensitivity threshold of abnormal judgment. These verified optimized parameters are then loaded into the time-series scheduling table through online updates.

Images