A system for monitoring the height difference between a roll-on roll-off wharf and a ship deck

Abstract

The application provides a kind of roll-on roll-off wharf and ship deck height difference monitoring system.The system includes wharf side intelligent monitoring unit, wharf height and water level monitoring module, deck elevation intelligent solution module, system integration and intelligent display unit and intelligent early warning and decision support module;Through high-precision dual-shaft servo holder synchronous control posture coaxial laser ranging and multispectral vision module data acquisition, combined with plane fitting and multi-source data fusion algorithm, the ship deck elevation, inclination angle and wharf-deck real-time height difference are solved, and multi-source data control and hierarchical intelligent early warning are realized.The application realizes high-precision real-time automatic monitoring of roll-on roll-off operation height difference, effectively improves operation safety and efficiency, and adapts to the integrated application requirements of smart port.

Description

Technical Field

[0001] This invention relates to the technical field of port terminal equipment and ship loading and unloading auxiliary systems, and in particular to a monitoring system for the height difference between a roll-on / roll-off terminal and a ship deck. Background Technology

[0002] With the deepening of my country's efforts to build a strong transportation nation and the comprehensive upgrading of smart port construction, roll-on / roll-off (Ro-Ro) shipping, as a core mode of transportation for the automotive industry chain, the intermodal transport of heavy cargo, and cross-border trade logistics, occupies an increasingly important position in the domestic and international dual-circulation logistics system due to its outstanding advantages of high loading and unloading efficiency, low cargo damage, and adaptability to the transshipment of various types of goods. The automation and intelligent upgrading of port Ro-Ro operations has become a core development direction for improving the comprehensive operational capabilities of ports, strengthening full-process safety management, and adapting to the high-quality development needs of modern shipping logistics.

[0003] In the entire process of berthing and loading / unloading operations for roll-on / roll-off (Ro-Ro) ships, Ro-Ro vehicles and materials need to be transferred between the dock surface and the ship's deck via ramps. The real-time elevation difference between the dock surface and the ship's deck, as well as the three-dimensional tilt of the ship's deck, are core control parameters that directly define the safety boundaries of Ro-Ro operations and determine operational efficiency. Affected by multiple factors such as tidal fluctuations, adjustments in ship ballast water, and dynamic changes in ship draft during loading and unloading, the ship's deck elevation and the dock-deck elevation difference are in constant dynamic change. Once the elevation difference and deck tilt exceed the safety threshold, it can easily lead to major safety accidents such as ramp structure damage, vehicle overturning, and personnel casualties. Therefore, achieving high-precision, real-time, and continuous automated monitoring of the dock-deck elevation difference is a core requirement for the safety management of the entire Ro-Ro operation process.

[0004] Currently, elevation difference control in roll-on / roll-off (Ro-Ro) operations still relies primarily on manual visual observation and experience-based judgment. This approach suffers from inherent drawbacks such as low measurement accuracy, delayed response, significant susceptibility to subjective human factors, and ineffective operation at night or in adverse weather conditions like rain and fog. Existing publicly available monitoring solutions mostly employ single-point ranging using a single laser ranging or image recognition device. They lack a complete system integrating multi-source front-end sensing, dynamic elevation benchmark calibration, high-precision deck attitude calculation, integrated control, and intelligent early warning. Consequently, they cannot simultaneously achieve three-dimensional tilt state calculation of the ship's deck, real-time closed-loop monitoring of the elevation difference between the dock and deck, and full-process safety control. Furthermore, they suffer from poor adaptability to harsh environments, weak anti-interference capabilities, insufficient calculation accuracy, and inability to meet the integration requirements of automated operations in smart ports, making it difficult to comprehensively guarantee the safety and efficiency of Ro-Ro operations. Summary of the Invention

[0005] In view of the above problems, a system for monitoring the height difference between a roll-on / roll-off terminal and a ship's deck surface is proposed to overcome or at least partially solve these problems, comprising: The intelligent monitoring unit on the dock side includes a dual-axis servo gimbal, a rotatable laser ranging module and a multispectral vision module coaxially mounted on the dual-axis servo gimbal, and an edge computing node. The multispectral vision module is used to identify the effective planar area of ​​the ship's deck. The edge computing node is used to plan at least four non-collinear laser measuring points located within the effective planar area using a uniform distribution strategy, and generate gimbal control commands. The rotatable laser ranging module is used to execute ranging according to the control commands. During ranging, the multispectral vision module synchronously verifies the laser landing position. If it identifies that the laser has landed in an invalid area, the edge computing node replans and supplements the measuring points. The wharf deck height and water level monitoring module includes a radar level gauge fixed to the wharf hydraulic structure for real-time measurement of the vertical height from the wharf deck to the water surface. The intelligent deck elevation calculation module communicates with the intelligent monitoring unit on the dock side and the dock height and water level monitoring module. It is used to filter the distance measurement data by using the effective plane area identification results, retain only the distance measurement data where the laser point is located in the effective area, and calculate the absolute elevation and three-dimensional tilt angle of the ship deck surface by using a plane fitting algorithm based on at least 4 non-collinear effective measurement points after filtering, and calculate the real-time height difference between the dock surface and the ship deck surface. The system integration and intelligent display unit, including a data processing host and a display terminal, is used for the fusion processing, credibility assessment and visualization of end-to-end monitoring data; The intelligent early warning and decision support module communicates with the system integration and intelligent display unit to pre-store graded safety thresholds, complete real-time data comparison and early warning, and generate roll-on / roll-off operation adjustment suggestions.

[0006] Optionally, the multispectral vision module integrates a visible light imaging unit, an infrared imaging unit, and a thermal imaging unit, supporting the fusion recognition of visible light, infrared, and thermal imaging. It also incorporates a deep learning-based artificial intelligence recognition model to automatically identify the effective planar area of ​​the ship's deck, the edge of the vehicle passage, and obstacles in the work area.

[0007] Optionally, the intelligent deck elevation calculation module is also equipped with a ship-side data communication interface, which is used to access the real-time attitude data of the shipborne inertial measurement unit and / or the status data of the ship's ballast system.

[0008] Optionally, the intelligent deck elevation calculation module also uses a Kalman filter algorithm to fuse multi-point ranging data collected by the rotatable laser ranging module, deck effective plane area verification data output by the high-definition multispectral vision module, real-time attitude data accessed by the ship's end data communication interface, and / or the status data of the ship's ballast system.

[0009] Optionally, the ro-ro terminal and ship deck elevation difference monitoring system also includes a mobile monitoring and wireless sensor network, which includes mobile wireless cameras deployed in the ro-ro operation channel and wireless tilt sensors deployed on the ship deck. The mobile wireless cameras and wireless tilt sensors are connected to the system via a wireless network for distributed supplementary monitoring of the operation scenario.

[0010] Optionally, the system integration and intelligent display unit also integrates a digital twin server. The digital twin server is used to build a digital twin model of the dock-ship roll-on / roll-off operation scenario, and combine the tidal change pattern and ship ballast adjustment parameters to complete the prediction, simulation and visualization of the dock-deck height difference change trend.

[0011] Optionally, the intelligent early warning and decision support module is equipped with a communication interface for the augmented reality display terminal. The communication interface is used to push the real-time height difference between the dock surface and the ship deck surface, the three-dimensional tilt angle, the vertical height from the dock surface to the water surface, early warning information, and virtual safety boundary information to the augmented reality display terminal.

[0012] Optionally, the ro-ro terminal and ship deck elevation difference monitoring system also includes a data storage module. This module is used to perform hash-based on-chain storage of the system's raw monitoring data, calculation results, early warning event logs, and operational records throughout the entire process; wherein: The raw monitoring data includes multi-point ranging data collected by the rotatable laser ranging module, raw imaging data and deck area identification data collected by the multispectral vision module, and vertical height data from the dock surface to the water surface collected by the radar level gauge; the solution results include the absolute elevation data of the ship's deck surface, three-dimensional tilt angle data, and real-time height difference data between the dock surface and the ship's deck surface output by the intelligent deck surface elevation solution module.

[0013] Optionally, the raw monitoring data also includes real-time attitude data from the shipborne inertial measurement unit accessed through the ship's data communication interface, ship ballast system status data, as well as video data and tilt measurement data collected by mobile monitoring and wireless sensor networks; the solution results also include elevation change trend prediction data output by the digital twin server.

[0014] Optionally, the system also includes a standardized data communication interface for connecting to external monitoring equipment, including anemometers, ship draft detectors, and automatic identification system terminals.

[0015] This invention, through a multi-module collaborative monitoring system for the height difference between the ro-ro terminal and the ship's deck, achieves automated, high-precision, real-time monitoring of the height difference between the terminal and the deck, as well as the three-dimensional tilt state of the ship's deck. It solves the inherent defects of traditional manual observation, such as low accuracy, slow response, and poor adaptability to harsh working conditions. It also makes up for the shortcomings of existing single monitoring schemes, which cannot achieve closed-loop control of the entire process. By integrating multi-source data and performing precise calculations, it improves the reliability of monitoring, effectively enhances the safety control capabilities and loading and unloading efficiency of ro-ro operations, and can seamlessly adapt to the integration needs of automated operations in smart ports, providing stable technical support for the safe and efficient operation of ro-ro transportation. Attached Figure Description

[0016] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the present invention will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the overall architecture of a roll-on / roll-off terminal and ship deck height difference monitoring system provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the intelligent monitoring unit structure on the dock side provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the deck surface plane fitting and elevation calculation process provided in an embodiment of the present invention; Figure 4 This is a flowchart of multi-source data fusion and credibility assessment provided in the embodiments of the present invention; Figure 5 This is a schematic diagram of the digital twin model and prediction display interface provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of intelligent early warning and AR-assisted display provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the deployment of a mobile monitoring and wireless sensor network provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of a real-world application scenario of the system provided in this embodiment of the invention for summarizing roll-on / roll-off operations. Detailed Implementation

[0018] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0019] Reference Figures 1 to 7 This invention provides a system for monitoring the height difference between a roll-on / roll-off terminal and a ship's deck, which may specifically include: The intelligent monitoring unit on the dock side includes a dual-axis servo gimbal, a rotatable laser ranging module and a multispectral vision module coaxially mounted on the dual-axis servo gimbal, and an edge computing node. The multispectral vision module is used to identify the effective planar area of ​​the ship's deck. The edge computing node is used to plan at least four non-collinear laser measuring points located within the effective planar area using a uniform distribution strategy, and generate gimbal control commands. The rotatable laser ranging module is used to execute ranging according to the control commands. During ranging, the multispectral vision module synchronously verifies the laser landing position. If it identifies that the laser has landed in an invalid area, the edge computing node replans and supplements the measuring points. The wharf deck height and water level monitoring module includes a radar level gauge fixed to the wharf hydraulic structure for real-time measurement of the vertical height from the wharf deck to the water surface. The intelligent deck elevation calculation module communicates with the intelligent monitoring unit on the dock side and the dock height and water level monitoring module. It is used to filter the distance measurement data by using the effective plane area identification results, retain only the distance measurement data where the laser point is located in the effective area, and calculate the absolute elevation and three-dimensional tilt angle of the ship deck surface by using a plane fitting algorithm based on at least 4 non-collinear effective measurement points after filtering, and calculate the real-time height difference between the dock surface and the ship deck surface. The system integration and intelligent display unit, including a data processing host and a display terminal, is used for the fusion processing, credibility assessment and visualization of end-to-end monitoring data; The intelligent early warning and decision support module communicates with the system integration and intelligent display unit to pre-store graded safety thresholds, complete real-time data comparison and early warning, and generate roll-on / roll-off operation adjustment suggestions.

[0020] In one or more embodiments of the present invention, the intelligent monitoring unit on the wharf side can be preferentially fixedly installed on an unobstructed mooring pier or dedicated monitoring column at the front edge of the roll-on / roll-off wharf. The unit is configured with a dual-axis servo gimbal, a rotatable laser ranging module, a multispectral vision module, and an edge computing node. The rotatable laser ranging module and the multispectral vision module adopt a rigid coaxial mounting structure, so that the laser emission optical axis of the rotatable laser ranging module and the imaging optical axis of the multispectral vision module are completely coincident. The two modules are fixed together on the action end of the dual-axis servo gimbal, and the attitude is synchronously controlled by the dual-axis servo gimbal. The horizontal rotation and pitch adjustment action commands of the dual-axis servo gimbal are synchronously sent to the common mounting reference base of the two modules, ensuring that the two modules always remain coaxial and synchronous during the full range of attitude adjustment, without relative angular deviation, and avoiding misalignment between the laser measuring point and the imaging field of view.

[0021] During operation, the rotatable laser ranging module projects a laser beam onto the deck of the berthed ship according to a pre-set measurement point plan, completing distance data acquisition at multiple points on the deck and achieving multi-point ranging. Simultaneously, a multispectral vision module, triggered synchronously with the laser ranging action, acquires imaging data of the corresponding area of ​​the deck. First, an imaging feature extraction algorithm identifies the precise pixel coordinates of the laser projection point in the image, completing laser point location identification. Simultaneously, imaging feature analysis identifies the physical boundaries and operational access areas of the deck, excluding non-operational areas such as vehicles and fixed equipment, thus completing the identification of the effective planar area of ​​the deck and providing direct verification of the effectiveness of the laser measurement points. Based on this, edge computing nodes are directly connected to a dual-axis servo gimbal, a rotatable laser ranging module, and a multispectral vision module via wired links. This allows them to acquire all raw data collected from the front end in real time. The raw data is first preprocessed locally, including outlier removal from laser ranging data, noise reduction and filtering of multispectral imaging data, and validity verification of gimbal attitude data. Simultaneously, a unified high-precision hardware timescale is matched to all front-end acquired data, completing the timescale alignment of multi-source front-end data. This ensures that the time base of all data is completely consistent, eliminating deviations caused by time differences in data acquisition from different modules. The preprocessed and timescale-aligned data can be directly transmitted to the back end, reducing the computing power pressure and transmission latency of back-end data processing.

[0022] In one or more embodiments of the present invention, the wharf surface height and water level monitoring module may include a radar level gauge. The radar level gauge can be rigidly fixed and installed at a stable position on the water-facing side of the wharf hydraulic structure. Preferred locations include the top of the vertical bank at the wharf's front edge, the unobstructed water-facing end of the mooring pier, and other main hydraulic structures of the wharf. The installation location should avoid areas affected by collisions during ship berthing operations, wharf drainage outlets, and locations with severe water turbulence to prevent external factors from interfering with the measurement process. During installation, a hot-dip galvanized anti-corrosion stainless steel fixing bracket is used to rigidly connect the radar level gauge to the wharf hydraulic structure to ensure that the radar level gauge does not loosen or vibrate during operation. At the same time, the installation attitude of the radar level gauge is strictly calibrated so that the central axis of the electromagnetic wave transmitting antenna of the radar level gauge is always perpendicular to the still water surface, ensuring that the electromagnetic wave propagation path completely coincides with the path of the vertical height to be measured, and avoiding measurement errors caused by attitude deviation.

[0023] After installation, a high-precision total station can be used to complete the benchmark calibration, accurately measuring the fixed vertical height difference between the radar level gauge antenna's reference transmitting surface and the dock working surface. This calibrated fixed height difference parameter is pre-stored in the radar level gauge's processing unit to complete the module's initial calibration. During module operation, the radar level gauge uses the frequency-modulated continuous wave radar measurement principle to continuously emit high-frequency continuous electromagnetic waves towards the water surface below. After the electromagnetic waves contact the water surface, they are reflected, and the reflected echo is received by the radar level gauge's antenna unit. By calculating the time difference between electromagnetic wave transmission and reception, as well as the propagation speed of electromagnetic waves in the air, the real-time vertical distance from the radar antenna's reference transmitting surface to the water surface is accurately calculated. Combined with the pre-stored fixed height difference parameter between the antenna reference surface and the dock working surface, data conversion is completed, and the real-time vertical height data from the dock working surface to the water surface is obtained.

[0024] Furthermore, to ensure the stability and accuracy of the measurement data, the sampling frequency of the radar level gauge can be adjusted within the range of 1Hz to 10Hz according to the actual needs of roll-on / roll-off operations. At the same time, it has a built-in moving average filtering algorithm to smooth the instantaneous fluctuation measurement values ​​caused by waves and swells on the water surface, eliminating measurement errors caused by random fluctuations on the water surface, and continuously outputting stable and high-precision real-time data of the vertical height from the dock surface to the water surface. The module adopts a sealed design with an IP67 or higher protection level, which is suitable for outdoor operation environments with high salt spray and high humidity in port docks, avoiding the impact of marine environmental corrosion on equipment performance and measurement accuracy. At the same time, the module's measurement data can be periodically checked and calibrated by leveling to eliminate zero drift caused by long-term operation of the equipment and ensure long-term stability of measurement accuracy.

[0025] In one or more embodiments of the present invention, the intelligent deck elevation calculation module can be deployed in an edge computing node or a back-end data processing host on the dock side. This module can establish bidirectional real-time communication connections with the dock-side intelligent monitoring unit and the dock surface height and water level monitoring module via wired Ethernet or low-latency 5G wireless communication links, using a stable TCP / IP transmission protocol, ensuring the synchronization and reliability of data transmission. During operation, the module can receive in real-time multi-point ranging data from the rotatable laser ranging module (pre-processed and time-aligned), laser point coordinate data from the multispectral vision module, and the identification results of the effective planar area of ​​the ship's deck from the dock-side intelligent monitoring unit. Simultaneously, it receives real-time vertical height data from the dock surface to the water surface from the dock surface height and water level monitoring module. All received data carries a unified hardware timescale, ensuring complete consistency of the time base of the multi-source data and fundamentally avoiding calculation errors caused by timing misalignment.

[0026] After receiving the data, the intelligent deck elevation calculation module can first perform validity verification and adaptive weighting processing of the input data through a multi-source data fusion algorithm. This algorithm adopts an adaptive weighted fusion model adapted to the roll-on / roll-off operation scenario. The core principle is to accurately filter the laser ranging data based on the identification results of the effective plane area of ​​the ship deck output by the multispectral vision module, retaining only the ranging data where the laser point falls within the effective plane area, and completely eliminating invalid ranging data that falls on obstacles or non-deck areas. At the same time, according to the spatial distribution uniformity of the effective laser measuring points and the continuous variance value of a single set of ranging data, a corresponding weight coefficient is assigned to each effective measuring point. The more uniform the distribution of measuring points and the higher the stability of the ranging data, the larger the weight coefficient, and vice versa. This eliminates the interference of abnormal ranging values ​​on the subsequent calculation results and completes the fusion and optimization processing of multi-source data.

[0027] After data fusion and optimization, the intelligent deck elevation calculation module can calculate the spatial attitude of the ship's deck based on at least four non-collinear effective weighted laser measurement points using a plane fitting algorithm. This algorithm employs a least-squares plane fitting model adapted to the dynamic scenario of the ship's deck. During implementation, a unified spatial rectangular coordinate system is first established, with the fixed reference point pre-set at the wharf work surface as the origin. The X-axis is horizontally parallel to the wharf shoreline, the Y-axis is horizontally perpendicular to the wharf shoreline, and the Z-axis is vertically upward. Then, combining the horizontal rotation angle, pitch angle, and distance measurement values ​​of the dual-axis servo gimbal corresponding to each effective laser measurement point, the three-dimensional spatial coordinates of each measurement point in the spatial rectangular coordinate system are calculated. Based on the three-dimensional coordinates of effective measuring points, the optimal plane equation of the ship deck space is obtained by fitting using the least squares method. The plane equation adopts the general expression Ax+By+Cz+D=0, where A, B, C, and D are the coefficients of the plane equation obtained by fitting. The module calculates the pitch angle and roll angle of the ship deck surface relative to the horizontal plane through the normal vector (A,B,C) of the plane equation. The two together constitute the three-dimensional tilt angle of the ship deck surface. At the same time, combined with the real-time vertical height data of the dock surface to the water surface output by the dock surface height and water level monitoring module, and the preset absolute elevation of the dock reference point, the reference absolute elevation of the fitted ship deck plane is calculated, that is, the absolute elevation of the ship deck surface.

[0028] Ultimately, the intelligent deck elevation calculation module, based on the calculated absolute elevation of the ship's deck and the real-time elevation benchmark of the dock working surface, calculates the difference between the benchmark elevation of the dock working surface and the benchmark absolute elevation of the ship's deck through elevation difference calculation. This difference represents the real-time elevation difference between the dock surface and the ship's deck. The entire calculation process is executed synchronously with the front-end data acquisition process, and the calculation frequency is consistent with the sampling frequency of the laser ranging module, achieving continuous real-time calculation. The calculated absolute elevation of the ship's deck, three-dimensional tilt angle, and real-time elevation difference data can be output to the back-end system in real time, providing accurate basic data for subsequent visualization and early warning control.

[0029] In one or more embodiments of the present invention, the integrated intelligent display unit is deployed in the central control room of the roll-on / roll-off terminal, including a data processing host and a display terminal. The data processing host can be a rack-mounted industrial computer with redundant power supply and wide operating temperature capability, adapted to the long-term uninterrupted operation requirements of the port central control room, and installed in a standard cabinet in the central control room. The display terminal adopts a high-brightness, wide-viewing-angle LCD screen, which is fixed to the visual dispatching area of ​​the central control room by a wall-mounted or floor-standing bracket, and establishes a direct connection with the data processing host through HDMI 2.0 or DP high-definition digital interface to ensure zero-latency and high-definition image transmission. The data processing host is connected to the edge computing nodes and code of the terminal-side intelligent monitoring unit through a wired gigabit Ethernet ring network. The radar level gauge of the head and water level monitoring module and the intelligent deck elevation calculation module establish a stable full-duplex communication link, using Modbus TCP or a custom low-latency transmission protocol to acquire full-link monitoring data in real time. The full-link monitoring data specifically includes laser multi-point ranging data, multispectral visual recognition data, and pre-processed time-stamped aligned data output by the intelligent monitoring unit on the wharf side; vertical height data from the wharf surface to the water surface output by the wharf surface height and water level monitoring module; and absolute elevation data of the ship's deck surface, three-dimensional tilt angle data, and real-time height difference data between the wharf surface and the ship's deck surface output by the intelligent deck elevation calculation module. This ensures that the data processing host can completely cover all monitoring data throughout the entire operation process without any data link omissions.

[0030] After receiving the end-to-end monitoring data, the data processing host first performs the fusion processing of the end-to-end data. Specifically, it first performs secondary time stamp alignment on all access data. Using the standard time output by the high-precision GPS / BeiDou timing module built into the data processing host as a unified benchmark, it recalibrates the timestamps of monitoring data with different sampling frequencies and transmission delays to ensure that the time benchmarks of all monitoring data within the same sampling period are completely consistent. This eliminates the problem of data timing misalignment caused by transmission delays and sampling time differences of different front-end devices. Then, it performs dimensional normalization processing on the time-stamped end-to-end data. Monitoring data with different physical dimensions and different numerical ranges are mapped to a unified numerical range of 0-1 through a linear normalization algorithm, eliminating the dimensional differences of different types of data and providing a unified processing benchmark for subsequent credibility assessment.

[0031] After fusion processing, the data processing host performs a multi-dimensional credibility assessment on the end-to-end monitoring data. Specifically, a three-level verification mechanism can be adopted. The first level is data integrity verification, checking the frame structure integrity and data field integrity of the end-to-end monitoring data cycle by cycle. Cycle data with packet loss, disconnection, or missing fields are marked as invalid data and the anomaly type is recorded. The second level is data logical consistency verification, comparing the logical matching degree of related data from the same source within the same cycle. This includes the matching degree between laser ranging measurement point data and the multispectral visual effective plane area identification results, and the logical consistency between the changing trend of the vertical height from the dock surface to the water surface and the real-time elevation difference changing trend. Data exceeding a preset logical deviation threshold is marked as low-credibility data. The third level... To verify data stability, the sliding variance and fluctuation range of each monitoring data point are calculated over 3-5 consecutive sampling periods. Abrupt data with fluctuation ranges exceeding a preset stability threshold are marked as data to be reviewed. After completing the three-level verification, the data processing host assigns a confidence weight value in the range of 0-1 to the end-to-end data for each valid period. The weight value is positively correlated with the verification result. At the same time, the marking information of abnormal data and low-confidence data is synchronously attached to the corresponding data frame to complete the confidence assessment of the end-to-end data. The core principle of this three-level verification mechanism is to quantify the reliability of the end-to-end monitoring data through multi-dimensional cross-validation, eliminate or mark abnormal data, and avoid invalid data interfering with subsequent operation judgments, thereby ensuring the accuracy and reliability of monitoring results from the data level.

[0032] The display terminal synchronizes data with the data processing host in real time, completing the visualization of the entire monitoring data chain. In specific implementation, the display interface adopts a modular partition layout. The refresh frequency of all displayed content is consistent with the front-end data sampling and processing frequency, with a minimum of 1Hz, to ensure the real-time performance of the displayed content. The first area of ​​the interface is the core monitoring parameter visualization area, which uses digital instruments combined with real-time dynamic curves to intuitively display four core parameters: vertical height from the dock surface to the water surface, absolute elevation of the ship deck surface, three-dimensional tilt angle of the ship deck surface, and real-time height difference between the dock surface and the ship deck surface. The dynamic curves can display the parameter change trends over the past 15 minutes, and operators can zoom in and out of the time axis and retrieve parameter change curves for any historical period using touch or keyboard and mouse operations. The second area of ​​the interface is the entire data status visualization area, which uses a list format combined with status color bars. The interface displays the operational status of each front-end acquisition device, the transmission status of the entire data chain, and the data reliability assessment results. Normal status is marked with green, low reliability with yellow, and abnormal / invalid status with red. Operators can intuitively grasp the operational status and reliability of the entire data chain. The third area of ​​the interface is a raw data traceability visualization area, allowing operators to retrieve the corresponding raw laser ranging data, multispectral visual recognition results, and raw water level measurement data by selecting the corresponding timestamp, enabling traceability and verification of monitoring results. The entire visualization display interface adopts a UI design with clear and eye-catching fonts and icons, adapting to the long-distance viewing needs of the port control room. It also allows operators to customize the interface layout, display parameters, and refresh rate according to operational needs, meeting the personalized display requirements of different operational scenarios.

[0033] In one or more embodiments of the present invention, the intelligent early warning and decision support module can be deployed as a software functional module within the system integration and intelligent display unit or the data processing host, or it can be deployed as an independent redundant industrial control computer. This adapts to the high reliability requirements of 24 / 7 uninterrupted operation at port terminals. The module establishes a full-duplex, low-latency, stable communication connection with the system integration and intelligent display unit via a wired gigabit Ethernet link. The communication transmission can use the TCP / IP industrial communication protocol, supporting bidirectional data interaction. The module can synchronously acquire the full-link effective monitoring data output by the system integration and intelligent display unit in real time, after fusion processing and reliability assessment. Specifically, this includes real-time elevation difference data between the dock surface and the ship deck surface, three-dimensional tilt angle data of the ship deck surface, and vertical height data from the dock surface to the water surface. The data synchronization frequency is consistent with the sampling frequency and calculation frequency of the front-end monitoring data, with a minimum of 1Hz, ensuring that the real-time data acquired by the module is completely synchronized with the on-site operation status, without transmission delay or timing misalignment issues. This provides an accurate and real-time data foundation for subsequent early warning comparison and decision suggestion generation.

[0034] The intelligent early warning and decision support module has a built-in non-volatile storage unit for pre-storing graded safety thresholds. In practice, the graded safety thresholds adopt a three-tiered progressive architecture conforming to port roll-on / roll-off operation industry safety standards, divided into three levels: early warning, alarm, and emergency. Each level corresponds to setting threshold boundaries for three core control parameters: real-time height difference between the wharf deck and the ship's deck, the ship's deck trim angle, and the ship's deck roll angle. Specifically, the early warning threshold is the operational safety reminder boundary, with a corresponding parameter value of 80% of the safe operation limit; the alarm threshold is the operational risk control boundary, with a corresponding parameter value of 95% of the safe operation limit; and the emergency threshold is the operation stop line, with a corresponding parameter value of the safe operation limit specified in the safety operation standards. The pre-storage and configuration of graded safety thresholds are achieved through… The system integration and intelligent display unit's display terminal is completed. Operators can customize and modify the threshold parameters for three levels according to the vessel type parameters, the type of materials in the roll-on / roll-off operation, the on-site tidal conditions, and the dock operation specifications. The configured graded safety thresholds will be permanently stored in the module's non-volatile storage unit and will not be lost after the device is powered off or restarted. At the same time, the module has built-in threshold rationality verification logic, which can verify the compliance of the upper and lower limits of the configured thresholds to avoid warning failures caused by incorrect threshold configuration. The core principle of the three-level progressive threshold is to achieve progressive identification and control of operational risks through gradient safety boundary division, replacing the traditional single threshold control mode. This avoids missed risk assessment and reduces the interference of invalid warnings on normal operations.

[0035] During operation, the intelligent early warning and decision support module, based on pre-stored graded safety thresholds, compares real-time data and triggers early warnings. Specifically, the module first performs a pre-validation of the synchronously acquired real-time monitoring data. Only when the data's credibility weight reaches the module's preset acceptable threshold is the data included in the comparison range. Invalid data with insufficient credibility is directly filtered and marked as abnormal to avoid false warnings. After validity verification, the module compares the real-time elevation difference, pitch angle, and roll angle data of the current cycle with the pre-stored three-level graded safety thresholds parameter by parameter at a frequency of once per cycle to accurately determine the risk level corresponding to the current operation status. Simultaneously, the module has a built-in early warning anti-shake verification mechanism, which only triggers warnings when three consecutive sampling cycles... Only when the comparison results all exceed the threshold boundary of the same level will the corresponding level of early warning be officially triggered, eliminating false early warnings caused by parameter changes due to instantaneous ship swaying and water surface wave fluctuations. After the early warning is triggered, the module will push the early warning information such as the early warning level, the type of out-of-tolerance parameter, the out-of-tolerance range, and the trigger time to the system integration and intelligent display unit in real time. The information will be displayed in a pop-up window on the display terminal. At the same time, it will be linked to the audible and visual alarm in the central control room. Different early warning levels correspond to different audible and visual prompt modes. The early warning level uses yellow light and intermittent prompt sound, the alarm level uses orange light and high-frequency intermittent prompt sound, and the emergency level uses red light and continuous audible and visual alarm. This will achieve differentiated early warning prompts for different risk levels, allowing on-site operators to quickly identify the risk level and risk type.

[0036] Upon triggering an early warning, the intelligent early warning and decision support module generates matching, actionable adjustment suggestions for roll-on / roll-off (Ro-Ro) operations based on the warning level, the type of deviation parameter, and the deviation magnitude. In practice, the module utilizes a standardized operation adjustment rule library built upon Ro-Ro operation safety regulations and port operational experience. Each adjustment rule in the library corresponds one-to-one with the warning level, deviation parameter type, and deviation magnitude. The intelligent early warning and decision support module generates targeted, directly executable operation adjustment suggestions by matching the rule entries corresponding to the current warning information, rather than generalized prompts. For example, when a warning is triggered indicating that the roll angle exceeds the warning level threshold, the module will generate specific operational suggestions for ballast water adjustment based on the direction of the roll angle deviation. When an emergency high-level warning is triggered... When an over-limit warning is issued, the intelligent warning and decision support module will generate clear suggestions for operation start-up and shutdown control. The generated operation adjustment suggestions will be bound to the corresponding warning information and pushed synchronously to the display terminal of the system integration and intelligent display unit, and displayed synchronously in the warning zone. At the same time, the module will record all warning events and corresponding adjustment suggestions completely and store them synchronously in the data processing host of the system integration and intelligent display unit, supporting subsequent operation traceability and safety audit. The core principle of this rule-matching suggestion generation is to solidify the standardized safety management process of port roll-on / roll-off operations into executable rule items, replacing the traditional operation adjustment mode that relies on manual experience. This ensures that the generated adjustment suggestions accurately match the on-site risks, can directly guide on-site operations, and quickly eliminate safety hazards.

[0037] In a preferred embodiment of the present invention, the multispectral vision module integrates a visible light imaging unit, an infrared imaging unit, and a thermal imaging unit, supporting the fusion recognition of visible light, infrared, and thermal imaging. It also incorporates a deep learning-based artificial intelligence recognition model for automatically identifying the effective planar area of ​​the ship's deck, the edge of the vehicle passage, and obstacles in the work area.

[0038] Specifically, the multispectral vision module can be implemented using an integrated coaxial structure. The module integrates a visible light imaging unit, an infrared imaging unit, and a thermal imaging unit. The optical axes of these three imaging units coincide with the laser emission optical axis of the matching rotatable laser ranging module, and are jointly fixed to the action end of a dual-axis servo gimbal, adjusting their attitude synchronously with the gimbal. The entire multispectral vision module is encapsulated in an IP67-rated salt spray-proof sealed shell, suitable for outdoor port operations. The trigger ends of the three imaging units are electrically linked to the laser ranging module, enabling synchronized triggering of image acquisition and laser ranging actions, ensuring precise matching between the image and the laser measurement point position.

[0039] During operation, the multispectral vision module supports the fusion recognition of visible light, infrared, and thermal imaging. In specific implementation, the module has a built-in adaptive multispectral image fusion algorithm that can adjust the working weight of the three imaging units in real time according to the ambient light and visibility: during the day, high-resolution visible light imaging is used as the basis, with infrared and thermal imaging used to supplement features; at night, in rain, fog, backlight, and other adverse conditions, infrared contour imaging and thermal imaging temperature difference recognition are used as the core, and visible light detail information is fused to generate a highly recognizable fused image, solving the problem of single imaging mode failure in adverse environments.

[0040] The multispectral vision module also incorporates a deep learning-based artificial intelligence recognition model. This model can be pre-trained and optimized using a massive dataset of multispectral images from roll-on / roll-off (Ro-Ro) operations, enabling real-time edge-side inference. The multispectral vision module inputs the fused image data into the model, automatically and accurately identifying three types of targets: first, the effective planar area of ​​the ship's deck, outputting the pixel coordinates of the area's boundaries; second, the edges of the Ro-Ro vehicle passageway, outputting the linear boundary parameters of the passageway; and third, obstacles in the work area, outputting the obstacle's outline and location information. The recognition results are output to the edge computing node in real time, providing a basis for verifying the validity of laser measurement points. The single-frame inference latency does not exceed 100ms, meeting the requirements of real-time operations.

[0041] In a preferred embodiment of the present invention, the intelligent calculation module for deck elevation is further provided with a ship-end data communication interface, which is used to access the real-time attitude data of the shipborne inertial measurement unit and / or the status data of the ship's ballast system.

[0042] Specifically, the ship-side data communication interface can be integrated into the hardware carrier of the deck elevation intelligent calculation module, adapting to the shore communication scenario of roll-on / roll-off vessels. At the hardware level, it reserves standard industrial Ethernet interface and RS485 serial interface. At the wireless communication level, it integrates a marine-specific VHF digital communication unit and a 5G wireless communication unit, while being compatible with the marine general IEC61162 and NMEA0183 protocols and the industrial standard ModbusTCP protocol. It can adapt to the access requirements of shipborne equipment of different ship types and brands. The interface adopts an IP65 protection level design and can be stably deployed in the dockside outdoor cabinet or the central control room cabinet.

[0043] After the vessel is berthed, operators can use the configuration interface of the intelligent deck elevation calculation module to complete the communication pairing and protocol adaptation between the ship's data communication interface and the target shipboard equipment. The interface supports two modes: single device independent access and multi-device synchronous access, fully covering the "and / or" access requirements: it can be connected to the shipboard inertial measurement unit alone to receive real-time attitude data such as ship roll, pitch, and heading output by the unit; it can also be connected to the ship's ballast system alone to receive real-time status data such as ballast tank level, valve status, and ballast water adjustment output by the system; and it can also be connected to two types of equipment simultaneously to receive two types of real-time data.

[0044] The ship-side data communication interface has built-in data frame verification and time-stamp alignment logic. For each set of incoming ship-side data, it first performs frame structure integrity and numerical range rationality checks, and removes invalid data with packet loss or exceeding the limit. Then, it matches the valid data with the unified time reference of the deck elevation intelligent calculation module. After completing the time-stamp alignment, it is input into the calculation module in real time. The data processing delay of the entire interface process does not exceed 50ms, ensuring the synchronization of the incoming data and the data collected on the dock side, and meeting the operational requirements of real-time calculation.

[0045] In a preferred embodiment of the present invention, the intelligent deck elevation calculation module further uses a Kalman filter algorithm to fuse multi-point ranging data collected by the rotatable laser ranging module, deck effective plane area verification data output by the high-definition multispectral vision module, real-time attitude data accessed by the ship end data communication interface, and / or ship ballast system status data.

[0046] Specifically, the Kalman filter algorithm can be deployed as an embedded executable program within the computing unit of the intelligent deck elevation calculation module. Before the algorithm is activated, initialization configuration is completed. Based on the dynamic change characteristics of the attitude of the roll-on / roll-off vessel, the appropriate system state equation and observation equation are preset. At the same time, based on the inherent measurement accuracy of each data source, the corresponding process noise covariance matrix and observation noise covariance matrix are preset, thus establishing a stable benchmark framework for multi-source data fusion calculation.

[0047] During algorithm operation, three types of observation data sources are received in real time: the first type is multi-point ranging data of the ship's deck surface collected by the rotatable laser ranging module; the second type is the deck effective plane area verification data output by the high-definition multispectral vision module; and the third type is the real-time attitude data of the shipborne inertial measurement unit and the status data of the ship's ballast system accessed by the ship's data communication interface. The algorithm has built-in dual-mode adaptation logic, which supports the operation mode of accessing a single type of ship-end data alone or accessing two types of ship-end data simultaneously, fully covering the equipment configuration scenarios of different ship types.

[0048] The algorithm first performs unified time-scale alignment and dimension normalization preprocessing on all input data, and then completes the fusion processing through a closed-loop iteration of prediction and update: the prediction stage predicts the prior estimate of the deck attitude at the current moment based on the optimal estimate of the deck attitude at the previous moment, combined with the ship's ballast system status data; the update stage uses laser ranging data, visual verification data, and real-time shipboard attitude data as observations, calculates the optimal Kalman gain, iteratively corrects the prior estimate, and finally outputs the unbiased optimal estimate of the deck attitude. This algorithm effectively suppresses measurement noise and random errors from a single data source through complementary fusion of multi-source data, improves solution accuracy and anti-interference robustness, and its operation frequency is perfectly matched with the front-end data sampling frequency, meeting the requirements of real-time operations.

[0049] In a preferred embodiment of the present invention, the ro-ro terminal and ship deck elevation difference monitoring system further includes a mobile monitoring and wireless sensor network, which includes a mobile wireless camera deployed in the ro-ro operation channel and a wireless tilt sensor deployed on the ship deck. The mobile wireless camera and the wireless tilt sensor are connected to the system via a wireless network for distributed supplementary monitoring of the operation scenario.

[0050] Specifically, the mobile monitoring and wireless sensor network can be implemented using a low-latency wireless self-organizing network architecture, consisting of mobile wireless cameras deployed in the roll-on / roll-off (Ro-Ro) operation channel, wireless tilt sensors deployed on the ship deck, and a matching wireless gateway. The wireless gateway is deployed in the dock control room or at the forefront monitoring pillar, and connects to the core data link of the elevation difference monitoring system via wired gigabit Ethernet, providing network access and data forwarding services for the front-end devices. The wireless network can use the port-specific encrypted Wi-Fi 6 protocol, supporting concurrent access by multiple devices, and has the ability to resist multipath interference and adapt to high salt spray environments, and can stably cover the entire Ro-Ro operation area of ​​the dock and the deck working surface of berthed ships.

[0051] The portable wireless camera adopts a magnetic quick-installation mobile base design, which does not require fixed civil engineering installation. Before the roll-on / roll-off operation begins, the operator can quickly deploy it in unobstructed locations such as guardrails and pillars on both sides of the channel according to the direction of the operation channel and the transshipment operation points. The camera has a built-in high-sensitivity starlight-level imaging unit and infrared supplementary light module, which supports all-weather operation scene monitoring. The real-time video stream collected is synchronously uploaded to the system integration and intelligent display unit through the wireless network for visual verification of the operation site, full recording of the transshipment process, and supplementation of the blind spots of the fixed monitoring unit on the dock side.

[0052] The wireless tilt sensor features a portable design with a built-in high-capacity battery. Before operation, it can be temporarily deployed at key points on the ship's deck, such as the ends of the work channels and gangway connections, to collect real-time deck tilt data. The sensor incorporates a high-precision MEMS tilt measurement unit, and the collected tilt data is uploaded in real-time via a wireless network to the deck elevation intelligent calculation module for distributed cross-verification of the deck's tilt status. All front-end devices automatically connect to the wireless gateway after power-on, allowing for flexible adjustment of deployment points according to operational needs. This enables distributed supplementary monitoring of the operational scenario, improving the comprehensiveness and reliability of the system's monitoring.

[0053] In a preferred embodiment of the present invention, the system integration and intelligent display unit also integrates a digital twin server. The digital twin server is used to construct a digital twin model of the dock-ship roll-on / roll-off operation scenario, and combine the tidal change law and ship ballast adjustment parameters to complete the prediction, simulation and visualization of the dock-deck height difference change trend.

[0054] Specifically, the digital twin server can be a rack-mounted high-performance server deployed in a standard cabinet in the terminal control room. It establishes a two-way low-latency communication link with the data processing host of the system integration and intelligent display unit via gigabit Ethernet. It can acquire full-link monitoring data, ship basic parameters and terminal environmental data in real time. The server has a built-in lightweight digital twin engine adapted to port roll-on / roll-off operation scenarios, providing stable computing power support for model building and prediction simulation.

[0055] When the server constructs a digital twin model of a wharf-ship roll-on / roll-off operation scenario, it first imports high-precision 3D modeling data of the wharf's hydraulic structure, the front-end operation area, and the location of monitoring equipment to establish a wharf environmental twin base. Then, based on the ship's hull drawings, deck structure, and ballast system parameters, it generates a 3D twin of the ship that perfectly matches the actual ship's size and structure. At the same time, it maps and binds the real-time data of the wharf-side monitoring unit, water level monitoring module, and deck elevation calculation module to the corresponding data nodes of the twin model one by one, achieving millisecond-level real-time synchronous mapping between the physical operation scenario and the digital twin model, ensuring that the model's state is completely consistent with the actual on-site state.

[0056] The server combines tidal variation patterns and ship ballast adjustment parameters to predict, simulate, and visualize the trend of changes in the wharf-deck elevation difference. In practice, it first imports official tidal forecast data for the work area, fits the water level change curve for the future work period, and simultaneously connects the ship's ballast system adjustment plan parameters to establish a correlation calculation model between ship draft changes and deck elevation fluctuations. Based on the current real-time elevation difference benchmark data, it performs rolling prediction simulations of the wharf-deck elevation difference trend over the next 30 minutes, generating continuous prediction curves. The prediction results and the dynamically updated twin model are simultaneously pushed to the display terminal of the system integration and intelligent display unit. The ship's attitude change trend and elevation difference prediction curve are intuitively displayed in the 3D twin view, achieving a visual representation of the prediction results and providing forward-looking operational guidance for workers.

[0057] In a preferred embodiment of the present invention, the intelligent early warning and decision support module is provided with a communication interface of the augmented reality display terminal. The communication interface is used to push the real-time height difference between the dock surface and the ship deck surface, the three-dimensional tilt angle, the vertical height from the dock surface to the water surface, early warning information, and virtual safety boundary information to the augmented reality display terminal.

[0058] Specifically, the communication interface of the augmented reality display terminal is integrated into the hardware computing unit of the intelligent early warning and decision support module. It adopts a low-latency dual-link wireless architecture design. At the hardware level, it can integrate a Wi-Fi 6 industrial wireless unit and a Bluetooth 5.2 industrial communication unit, which is compatible with the wireless transmission standards of maritime operation scenarios and supports AES256 hardware encrypted transmission to avoid data leakage and interference. The interface reserves a standardized data push protocol, which can be adapted to mainstream augmented reality display terminals. Before operation, operators can complete the pairing and binding of the interface and the target terminal and configure the transmission parameters through the module's configuration interface to establish a stable low-latency communication link. The link transmission delay is controlled within 20ms to ensure the real-time performance of data push.

[0059] The interface is directly connected to the core data caching unit of the intelligent early warning and decision support module, which can synchronously acquire and encapsulate all operational data to be pushed in real time. Specifically, it includes five types of core data: real-time height difference between the dock surface and the ship deck surface, three-dimensional tilt angle of the ship deck surface, vertical height from the dock surface to the water surface, graded early warning information, and virtual safety boundary information. In view of the lightweight display requirements of augmented reality display terminals, the interface standardizes and encapsulates the raw data and aligns the time scale to ensure that the time base of each set of pushed data is completely synchronized with the on-site operation status. The data push refresh frequency is consistent with the sampling frequency of the front-end monitoring data, with a minimum of 1Hz, to ensure that the terminal display data is consistent with the on-site status.

[0060] After receiving data pushed by the interface, the augmented reality display terminal uses a built-in spatial positioning and registration algorithm to accurately match the virtual safety boundary information with the physical space of the dock and ship deck. This results in a fixed virtual safety boundary line and risk warning area being overlaid and displayed in the actual field of vision of the workers. At the same time, in the unobstructed corners of the field of vision, dynamic values ​​of elevation difference, three-dimensional tilt angle, and vertical height from the dock surface to the water surface are updated in real time using a floating digital instrument. When a warning message is received, a pop-up window is highlighted in the field of vision to display the warning level and core content, and warning signs are overlaid simultaneously, allowing on-site workers to monitor the safety status of the operation in real time without leaving the work scene.

[0061] In a preferred embodiment of the present invention, the ro-ro terminal and ship deck elevation difference monitoring system further includes a data storage module. This module is used to perform hash-based on-chain storage of the system's raw monitoring data, calculation results, early warning event logs, and operational records throughout the entire process; wherein: The raw monitoring data includes multi-point ranging data collected by the rotatable laser ranging module, raw imaging data and deck area identification data collected by the multispectral vision module, and vertical height data from the dock surface to the water surface collected by the radar level gauge; the solution results include the absolute elevation data of the ship's deck surface, three-dimensional tilt angle data, and real-time height difference data between the dock surface and the ship's deck surface output by the intelligent deck surface elevation solution module.

[0062] Specifically, the data storage module can be implemented using an encrypted embedded architecture. It is deployed in the data processing host of the system integration and intelligent display unit in the form of hardware encryption unit and standardized executable program. The data storage module can be equipped with a national cryptographic SM3 hash algorithm unit that complies with the national commercial cryptography standard. It has reserved a standard node interface for the port operation supervision alliance chain and can establish an encrypted communication link with compliant alliance chain nodes. At the same time, it can establish a real-time data synchronization channel with various functional modules of the system through the high-speed data bus inside the system to ensure complete and uninterrupted collection of operation data throughout the entire process.

[0063] During the operation of the data storage module, four types of data to be stored are collected synchronously throughout the entire system process at a preset fixed cycle: raw monitoring data, solution result data, early warning event logs, and operation records. The raw monitoring data strictly collects multi-point ranging data from the rotatable laser ranging module, raw imaging data and deck area identification data from the multispectral vision module, and vertical height data from the dock surface to the water surface from the radar level gauge. The solution result data strictly collects the absolute elevation data of the ship's deck surface, three-dimensional tilt angle data, and real-time height difference data between the dock surface and the ship's deck surface output by the intelligent deck surface elevation calculation module. For each set of completely collected data, the module first adds a unique high-precision timestamp and data type identifier, and then calculates a unique and irreversible hash digest value using the national cryptographic SM3 algorithm, ensuring the data's immutability from the source.

[0064] After the data storage module completes the hash digest calculation, it packages the data hash value, corresponding timestamp, and data type identifier to generate a storage block. This block is then pushed to the port operation supervision consortium blockchain in real time via an encrypted link to complete the on-chain storage. The original data is encrypted and stored in a local redundant storage array, achieving a one-to-one binding between the on-chain hash and the off-chain original data. When conducting subsequent operation safety audits or data traceability, the local original data for the corresponding time period is retrieved, the hash value is recalculated, and compared with the hash digest stored on the blockchain. This allows for rapid verification of the data's integrity and authenticity, meeting the compliance requirements for roll-on / roll-off operation safety supervision and accident traceability.

[0065] In a preferred embodiment of the present invention, the raw monitoring data also includes real-time attitude data of the shipborne inertial measurement unit accessed by the ship's data communication interface, ship ballast system status data, and video data and tilt measurement data collected by the mobile monitoring and wireless sensor network; the solution result data also includes elevation change trend prediction data output by the digital twin server.

[0066] Specifically, the data storage module, based on the original data acquisition framework, establishes real-time synchronous acquisition links with the ship's data communication interface, mobile monitoring, and wireless sensor network through a high-speed data bus with data source identification within the system. The link transmission uses the same sampling period and high-precision time reference as the basic storage data to ensure time alignment of all data. The module synchronously acquires extended raw monitoring data at fixed intervals, including real-time attitude data from the shipborne inertial measurement unit accessed through the ship's data communication interface, ship ballast system status data, and video data and tilt measurement data acquired by the mobile monitoring and wireless sensor network. During acquisition, the module incorporates frame structure integrity and numerical compliance verification logic to eliminate invalid data due to packet loss or exceeding numerical limits, ensuring the integrity and validity of the extended raw data.

[0067] The data storage module synchronously establishes a data connection with the digital twin server through a dedicated low-latency communication link, collecting extended solution result data in real time, namely the elevation difference change trend prediction data output by the digital twin server. During the collection process, the prediction data is matched with the timestamp of the corresponding operation period to ensure data traceability. The data storage module packages the aforementioned extended original monitoring data and extended solution result data with the basic storage data of the same period, adds a globally unique operation batch identifier, and generates a unique and irreversible hash digest for the entire data packet using the built-in national cryptographic SM3 hash algorithm. It simultaneously completes on-chain hash storage and local encrypted storage, realizing the immutable storage of the entire operation data. During subsequent auditing and traceability, the full-dimensional data can be retrieved to complete the verification, meeting the full-process supervision requirements of roll-on / roll-off operations.

[0068] In a preferred embodiment of the present invention, the system further includes a standardized data communication interface for accessing external monitoring equipment, including an anemometer, a ship draft detector, and an automatic identification system terminal.

[0069] Specifically, the standardized data communication interface can be integrated into the data processing host of the system integration and intelligent display unit. It adopts a modular design that is compatible with multiple protocols. At the hardware level, it reserves multiple RS485 serial interfaces with opto-isolation protection, gigabit Ethernet interfaces and LoRa wireless communication interfaces, which can be adapted to both wired and wireless access methods. At the software level, it has built-in standardized communication protocol libraries commonly used in the port and maritime fields, such as Modbus RTU / TCP, IEC61162, NMEA0183, etc., and supports plug-and-play device access configuration. The interface has anti-electromagnetic interference and surge protection capabilities, and can stably adapt to the high interference and high salt spray outdoor operating environment of port terminals.

[0070] During operation, operators can physically connect the target external monitoring equipment to the standardized data communication interface through a matching hardware link according to the on-site monitoring needs of the roll-on / roll-off operation. Then, through the system's visual configuration interface, they can select the communication protocol of the corresponding device, complete the device address mapping and data channel configuration, and quickly complete the device access and data synchronization. The external monitoring equipment specifically includes an anemometer, a ship draft detector, and an automatic identification system (AIS) terminal. After the anemometer is connected, it can transmit the wind speed and direction data at the operation site in real time. After the ship draft detector is connected, it can transmit the bow and stern draft data of the ship in real time. After the AIS terminal is connected, it can synchronize the basic identity and navigation status data of the berthed ship in real time.

[0071] The standardized data communication interface has built-in data preprocessing logic, which performs real-time frame structure integrity verification and numerical compliance verification on the data from the external devices. It also performs time stamp alignment for valid data to match the system's unified high-precision time base, incorporates it into the system's full-link monitoring data system, and pushes it synchronously to various functional modules of the system to complete fusion processing, visualization display, and early warning linkage. The interface supports concurrent access from multiple external devices, which can flexibly expand the system's monitoring dimensions without changing the system's core architecture, effectively improving the system's scenario adaptability.

[0072] This invention, through a multi-module collaborative monitoring system for the height difference between the ro-ro terminal and the ship's deck, achieves automated, high-precision, real-time monitoring of the height difference between the terminal and the deck, as well as the three-dimensional tilt state of the ship's deck. It solves the inherent defects of traditional manual observation, such as low accuracy, slow response, and poor adaptability to harsh working conditions. It also makes up for the shortcomings of existing single monitoring schemes, which cannot achieve closed-loop control of the entire process. By integrating multi-source data and performing precise calculations, it improves the reliability of monitoring, effectively enhances the safety control capabilities and loading and unloading efficiency of ro-ro operations, and can seamlessly adapt to the integration needs of automated operations in smart ports, providing stable technical support for the safe and efficient operation of ro-ro transportation.

[0073] The above is the overall concept of the present invention. For ease of understanding, the present invention also provides the following embodiments.

[0074] Reference Figure 8 This embodiment is applied to the scenario of loading and unloading automobiles at a coastal roll-on / roll-off terminal. The complete implementation process is as follows.

[0075] In the system deployment and initialization phase, the intelligent monitoring unit is first installed on the top of the mooring pier at the front of the roll-on / roll-off terminal, where there is no obstruction and no risk of ship collision. The unit is fixed with hot-dip galvanized stainless steel bracket bolts. The unit's shell has an IP67 protection rating, which can adapt to the high salt spray and high humidity outdoor working environment of the port. The unit integrates a rotatable laser rangefinder, a multispectral vision module, and an edge computing node. The multispectral vision module includes a visible light camera and an infrared thermal imager. All sensing components are fixed to the motion end of a high-precision dual-axis servo gimbal. During installation, the optical axis is coaxially calibrated to ensure that the laser emission axis is completely aligned with the optical axis of the two imaging units. The gimbal achieves synchronous attitude control. The gimbal and the edge computing node are connected to the terminal's industrial ring network via wired gigabit Ethernet. At the same time, a 5G industrial wireless module is configured as a backup link to establish an AES256 encrypted TCP / IP communication link with the data processing host and digital twin server deployed in the terminal's central control room cabinet. The link transmission latency is controlled within 20ms. A radar level gauge is fixedly installed on the top of the water-facing side of the vertical pier wall, avoiding the drainage outlet and the turbulent water flow area. The antenna is calibrated so that its central axis is perpendicular to the water surface. A fixed height difference between the radar antenna reference plane and the pier working surface is calibrated using a high-precision total station. The signal from the radar level gauge is directly connected to the data processing host through a shielded twisted-pair cable. After the system is powered on, the operator uses the supporting control software in the central control room to input the absolute elevation of the dock benchmark point, the ship type parameters of the operating vessel, load the daily tide model released by the local maritime department, delineate the geographical coordinates of the operation monitoring area, and set the three-level safety thresholds corresponding to the elevation difference and tilt angle to complete the basic system initialization. Thirty minutes before the start of the roll-on / roll-off operation, the movable wireless camera is deployed on both sides of the guardrail of the roll-on / roll-off operation channel using a magnetic quick-installation base, and the wireless tilt sensor is temporarily deployed at the gangway connection position on the ship deck and at key positions at both ends of the operation channel. After both types of equipment are powered on, they automatically connect to the system's industrial wireless gateway through an encrypted Wi-Fi 6 self-organizing network to complete network registration and Beidou time synchronization. The equipment readiness status is synchronized to the control software interface in real time.

[0076] In the measurement and data acquisition process, before the ship berths and gangplank are connected, the system automatically starts after the ship's identity is matched through the connected Automatic Identification System (AIS) terminal. It can also be manually triggered remotely by the operator. After startup, the multispectral vision module first continuously scans the target ship's deck area at a frequency of 10Hz, simultaneously acquiring visible light, infrared, and thermal imaging images. The fused image is then input into a built-in deep learning algorithm model pre-trained on a massive roll-on / roll-off (Ro-Ro) operation scenario dataset. This model performs real-time inference on the edge, automatically identifying the effective planar area pixel boundaries of the ship's deck, the linear edges of vehicle passages, and the outlines and positions of obstacles such as vehicles and fixed equipment on the deck. The identification results are output to the edge computing node in real time. Based on the effective planar area identification results, the edge computing node intelligently plans laser measurement points using a uniform distribution strategy, ensuring that the number of measurement points is no less than four and all are non-collinear. All measurement points fall within the effective planar area, avoiding obstacles and deck edges, and simultaneously generating rotation control commands for the dual-axis servo gimbal. The dual-axis servo gimbal synchronously drives the rotatable laser rangefinder and multispectral vision module according to instructions, aiming at each planned measurement point in sequence. The rotatable laser rangefinder completes the precise distance measurement of the corresponding measurement point at a frequency of 20Hz, and simultaneously triggers the multispectral vision module to acquire the corresponding image of the measurement point, continuously verifying whether the laser point falls within the effective plane area. If the measurement point is identified as falling in an invalid area, a supplementary measurement point is immediately replanned to ensure that all distance measurement data comes from the effective deck plane. At the same time, environmental images and thermal imaging data of the operation area are continuously collected for subsequent traceability. Simultaneously, the radar level gauge continuously measures the height Hwater from the dock surface to the water surface at a frequency of 5Hz and uploads it to the data processing host in real time. If the operating vessel is equipped with shipborne equipment compatible with the system protocol, the ship's roll, trim, and bow attitude data output by the shipborne inertial measurement unit, as well as the ballast tank level, valve status, and ballast water adjustment status data of the ship's ballast system, can be encrypted and transmitted to the dock data processing host through a dedicated maritime VHF digital communication link or wireless interface to complete real-time ship-shore data synchronization.

[0077] In the data processing, fusion, and modeling stages, all sensor data collected from the front end first converges to the edge computing node. The node preprocesses all data, including outlier removal from laser ranging data, noise reduction and filtering of multispectral image data, and frame structure integrity verification of various data types. Then, using the standard time output by the BeiDou time synchronization module as a unified benchmark, high-precision timestamps are matched to all valid data to complete the time stamp alignment of multi-source data and eliminate timing misalignment caused by sampling time differences and transmission delays of different devices. After preprocessing, the edge computing node sends the laser ranging data, the deck effective plane area verification results output by the multispectral vision module, and data from the optional shipborne inertial measurement unit to the data processing host in real time through the industrial ring network. The data processing host runs an extended Kalman filter multi-source data fusion algorithm adapted to the dynamic scenario of roll-on / roll-off operations. First, a spatial rectangular coordinate system is established with the dock reference point as the origin. Combining the gimbal rotation angle and distance measurement values ​​corresponding to each effective laser measurement point, the three-dimensional spatial coordinates of each measurement point within the coordinate system are calculated. Using these coordinates as the observation basis, and combining the attitude data from the shipborne inertial measurement unit as the state prediction value, the complementary information from multiple data sources is fused through the prediction-update closed-loop iteration of the extended Kalman filter. This suppresses measurement noise and random errors from a single data source, and a high-confidence equation for the ship deck space plane is fitted, thereby calculating the absolute elevation H of the ship deck reference point. deck This includes the deck's three-dimensional tilt angles, namely the pitch and roll angles. Simultaneously, the data processing host receives H data uploaded by the radar level gauge. water The data, combined with the built-in tidal model, undergoes moving average correction to eliminate instantaneous errors caused by water surface wave fluctuations, resulting in an accurate wharf deck elevation benchmark. The data processing host synchronously pushes the real-time calculated deck absolute elevation, three-dimensional tilt angle, wharf deck water level data, and vessel identity and operational status data to the digital twin server at a frequency of 1Hz. The digital twin server has a built-in lightweight three-dimensional digital twin engine, pre-stored with high-precision three-dimensional models of the wharf hydraulic structure, operational area, and monitoring equipment, as well as standardized three-dimensional models of mainstream roll-on / roll-off vessels. Upon receiving real-time data, it calls the corresponding vessel's three-dimensional model and the wharf environment model to complete spatial coordinate registration, constructing a 1:1 real-time mapping of the operational scene to the physical scene, with a mapping delay of no more than 100ms. Simultaneously, based on the current vessel tilt trend, the future tidal level change curve of the tidal model, and the adjustment plan parameters of the vessel's ballast system, the server establishes a deck elevation change prediction model, performing rolling prediction simulation of the wharf-deck elevation difference change within the next 30 minutes, generating a continuous elevation difference prediction curve, and synchronously updating it to the digital twin.

[0078] In the information output, early warning, and decision support stages, all processing results, predicted data, and real-time monitoring images are centrally visualized on the large-screen display terminal in the control room. The interface adopts a modular partition design, and the refresh rate is consistent with the front-end data sampling frequency to ensure the real-time nature of the displayed content. The core parameter area uses digital instruments combined with dynamic curves to display the real-time elevation difference between the dock and the deck, the deck's pitch and roll angles, and the real-time water level from the dock surface to the water surface. The visualization area displays a 3D real-time view of the digital twin, supporting zoom and rotation of the viewpoint, and simultaneously displays the elevation difference prediction trend curve. The video monitoring area adopts a picture-in-picture layout, simultaneously displaying real-time images from multispectral vision modules and multiple movable wireless cameras. The early warning area uses status color codes to indicate the system's operating status, and simultaneously displays alarm information and recommended operation suggestions. During system operation, the real-time calculated elevation difference and inclination angle data are compared with preset multi-level safety thresholds cycle by cycle. Only when the values ​​of three consecutive sampling cycles exceed the corresponding threshold boundary is the corresponding level of warning triggered to avoid false warnings caused by instantaneous ship rolling. When the values ​​approach or exceed the threshold, the system automatically triggers the audible and visual alarm in the central control room. Different warning levels correspond to different audible and visual prompt modes. At the same time, the specific alarm type is displayed in a highlighted pop-up window in the warning area, such as "excessive elevation difference" or "excessive inclination". Based on the built-in roll-on / roll-off operation safety standard rule library, the system matches and generates corresponding recommended operation measures, such as "it is recommended to adjust the starboard ballast water to reduce the inclination angle". These early warning messages, core monitoring parameters, and virtual safety boundary information are simultaneously pushed in real time to the augmented reality glasses worn by on-site operation managers through the communication interface reserved by the intelligent early warning and decision support module. The glasses, using a built-in spatial positioning algorithm, accurately align the virtual safety boundary line with the physical space on-site, displaying it overlaid in the operator's real field of vision. Simultaneously, in unobstructed areas of vision, core parameters are refreshed in real time as floating instruments, and warning messages are highlighted upon receipt, providing immersive operation guidance. Throughout the entire operation process, the blockchain data storage module automatically collects all key data at fixed intervals, including raw sensor sampling data, fusion calculation results data, a complete record of early warning events, and operation logs. Using the national cryptographic SM3 algorithm, a unique and irreversible hash digest is generated for each set of data. The hash value, corresponding timestamp, and data identifier are packaged and uploaded to the port operation supervision consortium blockchain for on-chain storage. The original data is encrypted and stored in a local redundant storage array, achieving a one-to-one binding between the on-chain hash and the off-chain original data. This ensures the authenticity and immutability of data throughout the operation process, providing a reliable basis for subsequent safety audits, accident analysis, and insurance claims.

[0079] The above provides a detailed description of the height difference monitoring system between a roll-on / roll-off terminal and a ship's deck. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, those skilled in the art will recognize that, based on the ideas of this invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A system for monitoring the height difference between a roll-on / roll-off terminal and a ship's deck, characterized in that, The system includes: The intelligent monitoring unit on the dock side includes a dual-axis servo gimbal, a rotatable laser ranging module and a multispectral vision module coaxially mounted on the dual-axis servo gimbal, and an edge computing node. The multispectral vision module is used to identify the effective planar area of ​​the ship's deck. The edge computing node is used to plan at least four non-collinear laser ranging points located within the effective planar area using a uniform distribution strategy, and generate gimbal control commands. The rotatable laser ranging module is used to execute ranging according to the control commands. During ranging, the multispectral vision module synchronously verifies the laser landing position. If it identifies a landing in an invalid area, the edge computing node replans and supplements the ranging points. The wharf deck height and water level monitoring module includes a radar level gauge fixed to the wharf hydraulic structure for real-time measurement of the vertical height from the wharf deck to the water surface. The intelligent deck elevation calculation module is connected to the intelligent monitoring unit on the dock side and the dock height and water level monitoring module. It is used to filter the distance measurement data using the effective plane area identification results, retain only the distance measurement data where the laser point is located in the effective area, and calculate the absolute elevation and three-dimensional tilt angle of the ship deck surface through a plane fitting algorithm based on at least 4 non-collinear effective measurement points after filtering, and calculate the real-time height difference between the dock surface and the ship deck surface. The system integration and intelligent display unit, including a data processing host and a display terminal, is used for the fusion processing, credibility assessment and visualization of end-to-end monitoring data; The intelligent early warning and decision support module communicates with the system integration and intelligent display unit to pre-store graded safety thresholds, complete real-time data comparison and early warning, and generate roll-on / roll-off operation adjustment suggestions.

2. The system of claim 1, wherein, The multispectral vision module integrates a visible light imaging unit, an infrared imaging unit, and a thermal imaging unit, supporting the fusion recognition of visible light, infrared, and thermal imaging. It has a built-in deep learning-based artificial intelligence recognition model for automatically identifying the effective planar area of ​​the ship's deck, the edge of the vehicle passage, and obstacles in the work area.

3. The system according to claim 2, characterized in that, The intelligent deck elevation calculation module is also equipped with a ship-end data communication interface, which is used to access the real-time attitude data of the shipborne inertial measurement unit and / or the status data of the ship's ballast system.

4. The system of claim 3, wherein, The intelligent deck elevation calculation module also uses a Kalman filter algorithm to fuse multi-point ranging data collected by the rotatable laser ranging module, deck effective plane area verification data output by the high-definition multispectral vision module, real-time attitude data accessed by the ship's data communication interface, and / or the status data of the ship's ballast system.

5. The system according to claim 4, characterized in that, The system for monitoring the height difference between the roll-on / roll-off terminal and the ship deck also includes a mobile monitoring and wireless sensor network. The mobile monitoring and wireless sensor network includes mobile wireless cameras deployed in the roll-on / roll-off operation channel and wireless tilt sensors deployed on the ship deck. The mobile wireless cameras and wireless tilt sensors are connected to the system via a wireless network for distributed supplementary monitoring of the operation scenario.

6. The system of claim 5, wherein, The system integration and intelligent display unit also integrates a digital twin server, which is used to construct a digital twin model of the dock-ship roll-on / roll-off operation scenario, and combine tidal change patterns and ship ballast adjustment parameters to complete the prediction, simulation and visualization of the dock-deck height difference trend.

7. The system of claim 6, wherein, The intelligent early warning and decision support module is equipped with a communication interface for the augmented reality display terminal. The communication interface is used to push the real-time height difference between the dock surface and the ship deck surface, the three-dimensional tilt angle, the vertical height from the dock surface to the water surface, early warning information, and virtual safety boundary information to the augmented reality display terminal.

8. The system of claim 7, wherein, The ro-ro terminal and ship deck elevation difference monitoring system also includes a data storage module. This module is used to perform hash-based on-chain storage of the system's raw monitoring data, calculation results, early warning event logs, and operational records throughout the entire process. The raw monitoring data includes multi-point ranging data collected by the rotatable laser ranging module, raw imaging data and deck area identification data collected by the multispectral vision module, and vertical height data from the dock surface to the water surface collected by the radar level gauge; the calculation result data includes the absolute elevation data of the ship's deck surface, three-dimensional tilt angle data, and real-time height difference data between the dock surface and the ship's deck surface output by the intelligent deck surface elevation calculation module.

9. The system of claim 8, wherein, The raw monitoring data also includes real-time attitude data from the shipborne inertial measurement unit accessed through the ship's data communication interface, ship ballast system status data, as well as video data and tilt measurement data collected by mobile monitoring and wireless sensor networks; the solution result data also includes elevation change trend prediction data output by the digital twin server.

10. The system according to claim 9, characterized in that, The system also includes a standardized data communication interface for connecting to external monitoring equipment, including anemometers, ship draft detectors, and automatic identification system terminals.

Images