Unmanned ship task planning and fuel supply cooperative scheduling system

Abstract

The invention provides an unmanned ship task planning and fuel supply cooperative scheduling system, and relates to the technical field of intelligent planning, and the system comprises an obtaining module which is used for obtaining the task information of a plurality of unmanned ships; based on the task information, an initial navigation path from the current position to the task target position is generated for each unmanned ship by adopting a segmented path planning method; the construction module is used for collecting and fusing environment and ship state data along the initial navigation path of each unmanned ship, and constructing environment situation information and a space reference frame of a unified space-time standard; the calculation module is used for extracting three-dimensional space feature data along the air route based on the space reference framework and calculating space boundary features; and dividing the air route space into a plurality of analysis areas according to the boundary features, and correspondingly associating the environment situation information to each area. According to the invention, dynamic adaptation of path optimization and supply scheduling is realized.

Description

Technical Field

[0001] This invention relates to the field of intelligent planning technology, and in particular to a coordinated scheduling system for unmanned vessel mission planning and refueling. Background Technology

[0002] With the large-scale application of unmanned vessels in fields such as marine environmental monitoring and resource exploration, there are increasingly more scenarios where multiple unmanned vessels work together to perform complex tasks over long periods of time and across sea areas. Endurance and refueling efficiency have become key factors restricting the continuity of operations. However, in existing technologies, the path planning of unmanned vessels focuses on obstacle avoidance and mission timeliness, while refueling scheduling is often carried out independently of path planning, lacking a dynamic linkage mechanism.

[0003] For example, when multiple unmanned surface vessels (USVs) are conducting environmental monitoring missions in a certain sea area, some vessels may experience insufficient fuel reserves during their voyage. The existing dispatch system cannot dynamically adjust the location and order of refueling points based on the real-time navigation paths of each vessel and changes in the marine environment, nor can it simultaneously optimize the navigation routes of oil receiving vessels to adapt to the refueling arrangements. This results in poor coordination between the refueling vessels and the oil receiving vessels, affecting the progress of the mission. The core technical defect of this problem is that the existing system has not built a deep collaborative framework for mission planning and fuel refueling. It is difficult to integrate real-time environmental data, vessel status information, and refueling resource conditions for global optimization, resulting in insufficient flexibility and efficiency in collaborative operations, and failing to meet the actual needs of multi-vehicle collaboration in complex sea areas. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a collaborative scheduling system for unmanned vessel mission planning and refueling, so as to achieve dynamic adaptation of path optimization and refueling scheduling.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] The first aspect is the unmanned vessel mission planning and refueling coordination scheduling system, including:

[0007] The acquisition module is used to acquire mission information from multiple unmanned surface vessels (USVs); based on the mission information, a segmented path planning method is used to generate an initial navigation path for each USV from its current position to the mission target position.

[0008] The module is used to collect and fuse environmental and vessel status data along the initial navigation path of each unmanned vessel to build a unified spatiotemporal benchmark for environmental situation information and spatial reference framework.

[0009] The calculation module is used to extract three-dimensional spatial feature data along the route based on a spatial reference frame, calculate spatial boundary features, divide the route space into multiple analysis areas according to the boundary features, and associate environmental situation information with each area accordingly.

[0010] The optimization module is used to comprehensively evaluate the multi-dimensional features of each region, obtain the trajectory correction strategy, and then dynamically optimize and smooth the initial navigation path to obtain the optimized global reference navigation path.

[0011] The judgment module is used to determine whether refueling is needed based on the optimized navigation path and the current fuel status of each unmanned vessel. If so, it generates a refueling scheduling plan.

[0012] The processing module is used to dispatch unmanned refueling vessels to corresponding refueling points according to the fuel supply scheduling scheme, and to perform precise fuel refueling by controlling the robotic arm to dock with the oil receiving vessel's fuel inlet.

[0013] The monitoring module is used to monitor the real-time status of all ships during navigation and refueling, and to provide fault warnings and comprehensive status assessments. If a sudden emergency event is detected, it will autonomously execute emergency operations according to the event type.

[0014] Furthermore, mission information for multiple unmanned surface vessels (USVs) is acquired; based on this mission information, a segmented path planning method is used to generate an initial navigation path for each USV from its current position to the mission objective position, including:

[0015] Receive mission information from multiple unmanned vessels issued by the mission platform;

[0016] By analyzing the received mission information, the mission start point, mission target point, and mission priority parameters corresponding to each unmanned vessel are extracted.

[0017] Based on the extracted mission start point and mission target point, a segmented path planning method is used to generate an initial navigation path for each unmanned vessel from its current position to the mission target position.

[0018] Furthermore, along the initial navigation paths of each unmanned vessel, environmental and vessel status data are collected and fused to construct a unified spatiotemporal benchmark for environmental situation information and a spatial reference framework, including:

[0019] The system receives raw environmental data collected in real time by the multi-source sensors carried by each unmanned vessel during navigation via a satellite communication network. The raw environmental data includes positioning information, lidar point cloud data, millimeter-wave radar obstacle detection data, and visual environmental feature data.

[0020] Based on the original environmental data, heterogeneous data are synchronized in time and transformed in space by using a unified spatiotemporal benchmark to generate a standardized environmental dataset with spatiotemporal consistency.

[0021] By fusing standardized environmental datasets with pre-loaded digital marine map information, a three-dimensional environmental situation information covering the navigation paths of each unmanned vessel is constructed.

[0022] Based on three-dimensional environmental situation information, combined with the historical navigation trajectory data and current real-time status parameters of each unmanned vessel, a set of ship dynamic characteristic parameters is calculated. The ship dynamic characteristic parameter set is then spatiotemporally correlated with the three-dimensional environmental situation information to obtain ship-environment interaction reference data.

[0023] By performing integrity verification and outlier detection processing on the ship-environment interaction reference data, outlier data points are removed and data interpolation compensation is performed, ultimately generating high-precision environmental situation information and spatial reference framework with a unified spatiotemporal benchmark.

[0024] Furthermore, based on a spatial reference frame, three-dimensional spatial feature data along the route are extracted, and spatial boundary features are calculated. The route space is then divided into multiple analysis regions based on these boundary features, and environmental situation information is correlated to each region accordingly, including:

[0025] By using high-precision environmental situation information and a spatial reference frame with a unified spatiotemporal benchmark, the original three-dimensional spatial features along the routes of each unmanned vessel are extracted from the spatial reference frame. The original three-dimensional spatial features include water depth and topography data, obstacle spatial distribution data, water flow vector field data, and meteorological influence parameters.

[0026] By performing data preprocessing operations on the original three-dimensional spatial feature data to eliminate noise and outliers, a standardized three-dimensional spatial feature dataset is formed. The standardized three-dimensional spatial feature dataset is then input into a three-dimensional convex hull construction algorithm to calculate the precise boundary feature data of the airway space.

[0027] Based on the precise boundary feature data of the airway space, combined with the navigation performance parameters and safety margin requirements of each unmanned vessel, the airway space is dynamically divided into multiple analysis regions with different analysis granularities.

[0028] The constructed environmental situation information is accurately matched and associated with multiple analysis regions according to the principle of spatiotemporal consistency, and a corresponding set of environmental feature parameters is assigned to each analysis region.

[0029] Furthermore, by comprehensively evaluating the multi-dimensional characteristics of each region, a trajectory correction strategy is derived, thereby dynamically optimizing and smoothing the initial navigation path to obtain an optimized global reference navigation path, including:

[0030] By receiving the set of environmental characteristic parameters assigned to each analysis area, the navigation risk coefficient, energy efficiency coefficient and obstacle avoidance safety coefficient of each analysis area are quantitatively evaluated in multiple dimensions based on the set of environmental characteristic parameters, and the regional navigation characteristic evaluation matrix is ​​obtained.

[0031] Based on the regional navigation feature evaluation matrix, combined with the navigation performance constraints and mission priority parameters of each unmanned vessel, a segmented weighted cost function is constructed. The optimal trajectory correction offset for each segment is calculated through a dynamic weight allocation mechanism, and a trajectory correction strategy set is obtained.

[0032] The set of trajectory correction strategies is applied segment by segment to the initial navigation path, and the path nodes are dynamically offset to obtain a preliminary optimized navigation path sequence.

[0033] By performing Bézier curve smoothing on the initially optimized navigation path sequence, sharp turns and discontinuities in the path are eliminated, resulting in a global reference navigation path with continuous curvature variation.

[0034] Furthermore, based on the optimized navigation path and the current fuel status of each unmanned vessel, it is determined whether refueling is necessary. If so, a refueling schedule is generated, including:

[0035] By receiving optimized global reference navigation path data and the current real-time fuel status information of each unmanned vessel, the estimated fuel consumption of each unmanned vessel in each segment is calculated based on the global reference navigation path data, and a fuel balance prediction curve is generated by combining the current real-time fuel status information.

[0036] Based on the fuel balance prediction curve, a safe fuel threshold parameter is set. When the predicted fuel balance is lower than the safe fuel threshold, the fuel replenishment demand determination mechanism is triggered to obtain a list of unmanned vessels that need fuel replenishment and a replenishment priority sequence.

[0037] Based on the list of unmanned vessels requiring fuel replenishment and the priority sequence of replenishment, combined with the intersection of the navigation paths of each unmanned vessel, the marine environmental conditions, and the current location and available fuel inventory of the unmanned refueling vessel, the final replenishment point location and replenishment time window are determined through a spatiotemporal conflict resolution algorithm, resulting in a preliminary fuel replenishment scheduling plan.

[0038] By conducting multi-objective optimization and evaluation of the preliminary fuel replenishment scheduling plan, and comprehensively considering the timeliness of replenishment, navigation interference and fuel type matching, a final fuel replenishment scheduling plan is generated through a dynamic weight adjustment mechanism. The final fuel replenishment scheduling plan includes the receiver vessel identifier, replenishment point coordinates, fuel type, replenishment quantity, estimated arrival time and priority parameters.

[0039] Furthermore, through a fuel supply scheduling plan, unmanned refueling vessels are dispatched to corresponding refueling points, and their robotic arms are docked with the receiving vessel's fuel inlet to perform precise fuel refueling, including:

[0040] By analyzing the oil receiving vessel identification information, refueling point location, fuel type and quantity parameters in the fuel supply scheduling scheme, the mission instructions of the unmanned refueling vessel are obtained.

[0041] Based on the mission execution instructions, the designated unmanned refueling vessel is scheduled to obtain a dedicated route to the refueling point from its current position using a segmented path planning method, and the navigation status and estimated arrival time of the unmanned refueling vessel are monitored in real time through a satellite communication network.

[0042] When the unmanned refueling vessel reaches the preset safe distance range of the refueling point, the relative position and attitude information of the unmanned refueling vessel and the receiving vessel are obtained, and high-precision berthing control commands are obtained.

[0043] Based on the high-precision berthing control command, the robot arm's motion trajectory is planned by identifying the type of oil port on the receiving vessel in real time, determining the oil port specifications and spatial coordinates.

[0044] By planning the motion trajectory of the robotic arm, the end effector of the dual redundant six-degree-of-freedom robotic arm is controlled to move towards the target oil port. By sensing the contact force and visual feedback information in real time, the motion posture of the robotic arm is dynamically adjusted to complete the precise docking of the refueling nozzle and the oil port.

[0045] After confirming successful docking, the corresponding fuel tank is activated according to the fuel type and quantity requirements in the fuel supply scheduling plan. The fuel flow is monitored in real time by a high-precision mass flow meter, and a precise fuel refueling operation is performed until the preset refueling amount is reached.

[0046] Furthermore, during navigation and refueling, the real-time status of all vessels is monitored, and fault warnings and comprehensive status assessments are conducted. If a sudden emergency event is detected, emergency operations are autonomously executed according to the event type, including:

[0047] The system receives real-time ship operation status data uploaded by each unmanned vessel and unmanned refueling vessel through a satellite communication network, and generates a real-time ship status vector set based on the ship operation status data.

[0048] By extracting multi-dimensional features from the real-time status vector set of ships and performing pattern matching in combination with a pre-set fault feature library, the fault risk index of each ship is calculated. When the fault risk index exceeds a pre-set threshold, a fault warning signal is obtained and the abnormal ship is marked.

[0049] Based on fault warning signals and real-time ship status vector sets, a comprehensive status assessment is conducted on abnormal ships to calculate their navigation safety level. At the same time, the collaborative operation capability of normal ships is dynamically assessed, generating a ship status assessment report that includes both safety level and operational capability.

[0050] Based on the ship condition assessment report, identify the types of sudden emergency events, match the pre-set emergency operation strategy library based on the event type, and obtain the corresponding emergency operation instruction set;

[0051] The emergency operation instruction set is issued to the abnormal vessel and its cooperating vessels, controlling the abnormal vessel to perform emergency shutdown, safe anchoring and return to port operations. At the same time, cooperating vessels are dispatched to perform auxiliary rescue tasks. The effectiveness of the emergency operation is verified by real-time feedback data, and the emergency strategy is dynamically adjusted according to the effectiveness until the vessel's status returns to normal.

[0052] In a second aspect, a computing device includes:

[0053] One or more processors;

[0054] A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to execute the system.

[0055] Thirdly, a computer-readable storage medium storing a program that, when executed by a processor, performs the system.

[0056] The above-described solution of the present invention has at least the following beneficial effects:

[0057] By employing a series of technical means, including segmented path planning, multi-source data fusion and 3D convex hull construction algorithms, segmented weighted cost functions and Bézier curve smoothing techniques, spatiotemporal conflict resolution and dynamic weight adjustment mechanisms, dual-redundant six-degree-of-freedom robotic arm precise docking schemes, and full-process real-time monitoring and emergency coordination strategies, this technology effectively overcomes the technical problems of existing technologies, such as the disconnect between unmanned vessel mission planning and refueling, the difficulty in integrating multi-source heterogeneous data to achieve accurate environmental and vessel interaction adaptation, frequent spatiotemporal conflicts during refueling in multi-vessel collaborative operations, and the lack of global coordination in emergency response. This achieves deep dynamic adaptation between unmanned vessel swarm mission planning and refueling, improves the accuracy of path optimization and refueling decisions, enhances the continuity and efficiency of multi-vessel collaborative operations, strengthens the system's fault tolerance and safety in complex marine environments, and increases the practical application value of unmanned vessel swarms. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of the unmanned vessel mission planning and fuel replenishment collaborative scheduling system provided in an embodiment of the present invention.

[0059] Figure 2 This is a flowchart illustrating the process of a collaborative scheduling system for unmanned surface vessel mission planning and refueling provided by an embodiment of the present invention. It comprehensively evaluates the multi-dimensional characteristics of each region to obtain a trajectory correction strategy, thereby dynamically optimizing and smoothing the initial navigation path to obtain an optimized global reference navigation path. Detailed Implementation

[0060] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0061] like Figure 1 As shown, embodiments of the present invention propose an unmanned surface vessel (USV) mission planning and refueling coordination scheduling system, comprising:

[0062] The acquisition module is used to acquire mission information from multiple unmanned surface vessels (USVs); based on the mission information, a segmented path planning method is used to generate an initial navigation path for each USV from its current position to the mission target position.

[0063] The module is used to collect and fuse environmental and vessel status data along the initial navigation path of each unmanned vessel to build a unified spatiotemporal benchmark for environmental situation information and spatial reference framework.

[0064] The calculation module is used to extract three-dimensional spatial feature data along the route based on a spatial reference frame, calculate spatial boundary features, divide the route space into multiple analysis areas according to the boundary features, and associate environmental situation information with each area accordingly.

[0065] The optimization module is used to comprehensively evaluate the multi-dimensional features of each region, obtain the trajectory correction strategy, and then dynamically optimize and smooth the initial navigation path to obtain the optimized global reference navigation path.

[0066] The judgment module is used to determine whether refueling is needed based on the optimized navigation path and the current fuel status of each unmanned vessel. If so, it generates a refueling scheduling plan.

[0067] The processing module is used to dispatch unmanned refueling vessels to corresponding refueling points according to the fuel supply scheduling scheme, and to perform precise fuel refueling by controlling the robotic arm to dock with the oil receiving vessel's fuel inlet.

[0068] The monitoring module is used to monitor the real-time status of all ships during navigation and refueling, and to provide fault warnings and comprehensive status assessments. If a sudden emergency event is detected, it will autonomously execute emergency operations according to the event type.

[0069] In this embodiment of the invention, by employing techniques such as segmented path planning, multi-source data fusion and unified spatiotemporal benchmark construction, three-dimensional convex hull construction and route area division, multi-dimensional feature evaluation and dynamic path optimization and smoothing, intelligent judgment of fuel replenishment demand and spatiotemporal conflict resolution scheduling, precise scheduling of unmanned refueling vessels and docking and refueling with dual-redundant six-degree-of-freedom robotic arms, and real-time monitoring and emergency collaborative response throughout the entire process, the invention effectively overcomes the technical problems in existing technologies, such as the disconnect between unmanned vessel mission planning and fuel replenishment, the difficulty in integrating multi-source heterogeneous data to achieve precise adaptation between the environment and the vessel, frequent spatiotemporal conflicts during multi-vessel collaborative operations, and the lack of global coordination in emergency response. This results in deep dynamic adaptation of unmanned vessel swarm mission planning and fuel replenishment, improving the accuracy of path optimization and replenishment decisions, the continuity and efficiency of multi-vessel collaborative operations, and enhancing the system's fault tolerance and safety in complex marine environments, thereby increasing the practical application value of unmanned vessel swarms.

[0070] In a preferred embodiment of the present invention, mission information of multiple unmanned surface vessels (USVs) is acquired; based on the mission information, a segmented path planning method is used to generate an initial navigation path for each USV from its current position to the mission target position, including:

[0071] The system receives mission information from multiple unmanned vessels issued by the mission platform. Specifically, the mission platform is a dedicated scheduling device deployed in a shore-based centralized scheduling center. The device establishes a stable connection with all participating unmanned vessels through a satellite communication network and continuously receives mission information from multiple unmanned vessels issued by the platform. The received information includes each unmanned vessel's unique equipment identifier, specific mission type, mission execution deadline, detailed geographical location information of the mission start point, detailed geographical location information of the mission target point, and explanatory information related to the mission's urgency.

[0072] By analyzing and extracting the mission start point, mission target point, and mission priority parameters for each unmanned surface vessel (USV) from the received mission information, the process involves: First, performing structured analysis on the received mission information from multiple USVs. This is done by differentiating each USV based on its unique equipment identifier to ensure accurate matching of information with the corresponding vessel. Then, extracting the specific data of each USV's mission start point from the differentiated mission information, including its latitude and longitude coordinates, as well as the latitude and longitude coordinates of the mission target point, and the starting and ending points of the vessel's navigation. Next, based on the mission urgency description in the mission information, the mission priority is divided into three levels: high priority, medium priority, and low priority. Missions requiring rapid response, such as emergency rescue and important data collection, are assigned high priority; routine environmental monitoring and daily patrols are assigned medium priority; and secondary tasks, such as auxiliary data transmission and equipment status verification, are assigned low priority. This process completes the extraction and classification of all key parameters.

[0073] Based on the extracted mission start point and mission target point, a segmented path planning method is used to generate an initial navigation path for each unmanned surface vessel (USV) from its current position to the mission target position. Specifically, this involves: generating an initial navigation path for each USV individually; first, using the extracted coordinates of the mission start point and mission target point, determining the overall range and general direction of the global navigation; then, referring to existing basic geographic information of the sea area, initially judging the sea area characteristics along the navigation route; and finally, using the segmented path planning method to design the specific path, dividing the global path from the start point to the target point into multiple continuous sub-paths according to differences in the sea environment. In areas with no obvious obstacles and open sea, the sub-segment length is set relatively long to reduce the computational complexity of path planning. In areas near ports, waterways, islands, or areas with potential obstacles such as reefs and aquatic plants, the sub-segment length is set relatively short to improve path accuracy. For each sub-segment, a basic straight path connecting the starting and ending points of the sub-segment is planned to ensure smooth connection between the sub-segment paths without obvious turning points or conflicts. Finally, all sub-segment paths are integrated to form a complete initial navigation path that starts from the current position of the unmanned vessel, passes through each sub-segment in sequence, and finally arrives at the mission target position.

[0074] In this embodiment of the invention, by employing the technical means of receiving mission information from multiple unmanned vessels issued by the mission platform, parsing and extracting the mission start point, mission target point, and mission priority parameters corresponding to each vessel, and generating initial navigation paths for each unmanned vessel based on these key information using a segmented path planning method, the technical problems of insufficient precise extraction and differentiated adaptation of core parameters of single-vehicle missions and insufficient targeted path generation in existing multi-vehicle mission planning are effectively overcome. This achieves personalized and orderly generation of initial navigation paths for multiple unmanned vessels, ensuring that high-priority tasks are preferentially adapted during the path planning stage.

[0075] In a preferred embodiment of the present invention, environmental and vessel status data are collected and fused along the initial navigation path of each unmanned vessel to construct a unified spatiotemporal benchmark environmental situation information and spatial reference framework, including:

[0076] The system receives raw environmental data collected in real time by multi-source sensors on each unmanned surface vessel (USV) during navigation via a satellite communication network. This raw environmental data includes positioning information, lidar point cloud data, millimeter-wave radar obstacle detection data, and visual environmental feature data. Specifically, as each USV navigates along its generated initial navigation path, its onboard multi-source sensors continuously collect raw environmental data in real time. These sensors include positioning sensors, lidar sensors, millimeter-wave radar sensors, and visual acquisition sensors. The collected raw environmental data specifically includes the real-time latitude, longitude, and altitude information of the vessel obtained by the positioning sensors; the three-dimensional spatial point cloud data of surrounding objects obtained by the lidar sensors; the distance, speed, and direction of motion information of obstacles detected by the millimeter-wave radar sensors; and the visual environmental feature data such as coastline morphology, navigation mark positions, and other vessel outlines extracted from marine environmental images and videos captured by the visual acquisition sensors.

[0077] Based on the raw environmental data, heterogeneous data are synchronized in time and transformed in space using a unified spatiotemporal reference to generate a standardized environmental dataset with spatiotemporal consistency. Specifically, after the receiving end acquires all raw environmental data, data processing is performed based on the unified spatiotemporal reference, which adopts an internationally accepted time standard and geodetic coordinate system. First, the heterogeneous data collected by different sensors are synchronized in time, and a unified standard timestamp is added to each set of data to eliminate time deviations caused by differences in sensor response speeds. Then, spatial coordinate transformation is performed, converting the relative coordinate data collected by lidar, millimeter-wave radar, and visual sensors into absolute coordinate data in a unified geodetic coordinate system to ensure that various types of data are comparable in spatial location. After the above processing, a standardized environmental dataset with spatiotemporal consistency is generated.

[0078] By fusing standardized environmental datasets with pre-loaded digital marine map information, a three-dimensional environmental situation information covering the navigation paths of each unmanned vessel is constructed. Specifically, this involves fusing the standardized environmental datasets with pre-loaded digital marine map information. The pre-loaded digital marine map contains basic information such as water depth distribution, seabed topography, navigation planning areas, restricted navigation areas, and the location of known fixed obstacles in the marine area. During the fusion process, real-time dynamic information from the standardized environmental dataset, such as temporarily appearing floating obstacles and real-time water flow direction, is accurately overlaid onto the corresponding spatial location on the digital marine map. At the same time, combined with the navigation path range of each unmanned vessel, relevant environmental data covering all unmanned vessel navigation paths are selected to construct a three-dimensional environmental situation information.

[0079] Based on 3D environmental situation information, combined with the historical navigation trajectory data and current real-time status parameters of each unmanned surface vessel (USV), a set of ship dynamic characteristic parameters is calculated. This set of ship dynamic characteristic parameters is then spatiotemporally correlated with the 3D environmental situation information to obtain ship-environment interaction reference data. Specifically, this includes: based on the constructed 3D environmental situation information, further in-depth processing is performed on the historical navigation trajectory data and current real-time status parameters of each USV. The historical navigation trajectory data includes the USV's past navigation paths, speed variation patterns, and steering adjustment records in similar sea areas or similar missions. The current real-time status parameters cover the ship's current speed, heading angle, fuel consumption, etc. Data such as fuel consumption rate, propulsion operating parameters, and hull attitude are collected. Through comprehensive analysis and calculation of this data, a set of ship dynamic characteristic parameters is obtained, specifically including parameters reflecting the ship's motion characteristics such as acceleration, steering torque, sailing resistance coefficient, and fuel efficiency coefficient. Subsequently, the set of ship dynamic characteristic parameters is spatiotemporally correlated with three-dimensional environmental situation information to obtain the correspondence between the ship's dynamic state and surrounding environmental elements at a specific time point and spatial location. For example, the influence of water flow velocity at a certain location on the ship's sailing resistance, and the change of the ship's steering torque under specific terrain conditions, etc., ultimately forming interactive reference data between the ship and the environment.

[0080] By performing integrity verification and outlier detection processing on the ship-environment interaction reference data, removing outlier data points and performing data interpolation compensation, a high-precision environmental situation information and spatial reference framework with a unified spatiotemporal benchmark is finally generated. Specifically, this includes: performing integrity verification and outlier detection processing on the ship-environment interaction reference data. Integrity verification checks whether the data covers all navigation time periods and key spatial locations to determine if there are any missing data points. If data is missing in a certain time period or location, the missing area is marked. Outlier detection determines whether various data exceed a reasonable normal range. For example, if the positioning data deviation exceeds the normal error range or the obstacle distance data changes abruptly, these are identified as outlier data points and then removed. For data missing areas and blank areas formed after removing outlier data, normal data from adjacent time periods or adjacent spatial locations are used for interpolation compensation. Through reasonable interpolation calculation methods, data gaps are filled to ensure data continuity and integrity. After the above series of processing, a high-precision environmental situation information and spatial reference framework with a unified spatiotemporal benchmark is finally generated.

[0081] In this embodiment of the invention, by employing technical means such as receiving raw environmental data from multiple sources of sensors via satellite communication networks, synchronizing heterogeneous data in time and transforming spatial coordinates with a unified spatiotemporal reference, constructing three-dimensional environmental situation information by fusing digital maps of the sea area, forming ship-environment interactive reference data by associating ship dynamics characteristic parameter sets, and optimizing data quality through integrity verification and outlier processing, the invention effectively overcomes the technical problems of spatiotemporal inconsistency of multi-source heterogeneous data, lack of precise correlation between environment and ship status, and weak decision support caused by insufficient data reliability in the prior art. Thus, it successfully generates high-precision environmental situation information and spatial reference framework with a unified spatiotemporal reference.

[0082] In a preferred embodiment of the present invention, based on a spatial reference frame, three-dimensional spatial feature data along the route are extracted, and spatial boundary features are calculated. The route space is divided into multiple analysis regions according to the boundary features, and environmental situation information is correspondingly associated with each region, including:

[0083] By utilizing a high-precision environmental situation information and spatial reference framework with a unified spatiotemporal benchmark, raw three-dimensional spatial feature data along the routes of each unmanned vessel (UV) is extracted from the spatial reference framework. This raw three-dimensional spatial feature data includes water depth and topography data, obstacle spatial distribution data, water flow vector field data, and meteorological impact parameters. Specifically, based on the established high-precision environmental situation information and spatial reference framework with a unified spatiotemporal benchmark, raw three-dimensional spatial feature data is extracted along the specific routes of each UV. The extraction process strictly adheres to the principle of spatiotemporal consistency, ensuring that the data accurately corresponds to the spatial range and temporal dimension of the route. Water depth and topography data covers the actual water depth, seabed topographic relief height, and topographic type information at different longitude and latitude coordinates along the route. Obstacle spatial distribution data includes the precise geographical location, three-dimensional size parameters, shape characteristics, and attribute identifiers indicating whether various obstacles are dynamic. Water flow vector field data records the magnitude, direction, and frequency of water flow changes at different locations within the route area. Meteorological impact parameters include key meteorological indicators such as real-time wind speed, wind direction, visibility, precipitation intensity, and wave height.

[0084] By performing data preprocessing operations on the raw 3D spatial feature data to eliminate noise and outliers, a standardized 3D spatial feature dataset is formed. This standardized dataset is then input into a 3D convex hull construction algorithm to calculate the precise boundary feature data of the route space. Specifically, this includes: systematically preprocessing the extracted raw 3D spatial feature data; firstly, using smoothing techniques to eliminate noise interference in the data, such as applying a moving average to continuously collected water depth data to mitigate the impact of random fluctuations; subsequently, outlier detection and removal, with normal value ranges set for various data types. The criteria for judging outliers in water depth data are exceeding the historical average water depth of the area by ±5 meters, and outliers in water flow velocity... The judgment criteria are as follows: exceeding 10 meters per second; for meteorological parameters, the judgment criteria for wind speed anomalies are exceeding 25 meters per second. Data exceeding these ranges are considered anomalies and are removed. After noise elimination and anomaly processing, the data undergoes format unification and magnitude standardization. Data from different sources and units are converted into a unified format and magnitude, forming a standardized three-dimensional spatial feature dataset with a regular structure and reliable data. The standardized dataset is input into a three-dimensional convex hull construction algorithm. The algorithm performs wrapping calculations on all data points within the airway space to determine the three-dimensional contour boundary of the airway space, obtaining precise boundary feature data containing key parameters such as the maximum and minimum longitude, maximum and minimum latitude, maximum and minimum water depth of the airway space.

[0085] Based on precise boundary feature data of the airway space, and combined with the navigation performance parameters and safety margin requirements of each unmanned surface vessel (USV), the airway space is dynamically divided into multiple analysis regions with different analytical granularities. Specifically, this includes: fully considering the navigation performance parameters and safety margin requirements of each USV, based on precise boundary feature data of the airway space. USV navigation performance parameters include inherent attributes such as maximum speed, minimum turning radius, maximum draft, and wave resistance level; safety margin requirements cover key indicators ensuring navigation safety, such as a minimum safe distance of 300 meters from obstacles, a minimum safe water depth of 2 meters in the navigation area, and a speed redundancy of 1 meter per second to cope with changes in water flow; and based on the data, the airway space is further divided into multiple analysis regions with different analytical granularities. The route space is dynamically divided based on the complexity of the marine environment and the level of navigation risk. In areas with open seas, no obstacles, gentle currents, and stable weather conditions, the granularity of the analysis area is set to a larger value, with each 10 nautical miles forming an independent analysis area to reduce computational load. In areas with dense obstacles, such as near ports, channel junctions, or around islands, or in high-risk areas with rapid currents and variable weather conditions, the granularity of the analysis area is set to a smaller value, with each 2 nautical miles forming an independent analysis area to improve the accuracy of subsequent assessments and decisions. During the division process, it is ensured that the boundaries of each analysis area are continuous, non-overlapping, and completely cover the entire route space, forming multiple analysis areas with different granularities.

[0086] The constructed environmental situation information is precisely matched and correlated with multiple analysis areas according to the principle of spatiotemporal consistency. A corresponding set of environmental feature parameters is assigned to each analysis area. Specifically, this includes: matching and correlating the previously constructed high-precision environmental situation information with the divided analysis areas according to the principle of spatiotemporal consistency; the principle of spatiotemporal consistency requires that the time dimension of the environmental situation information completely corresponds to the spatial range of the analysis area, that is, each analysis area is only matched with the environmental situation information within its spatial range and corresponding time period; during the correlation process, a unique set of environmental feature parameters is assigned to each analysis area. The environmental feature parameter set comprehensively integrates detailed water depth and topography data, specific information on the spatial distribution of obstacles, real-time data of water flow vector fields, and key parameters of meteorological influence within the area, while also including relevant data on ship dynamics interaction in the area, ensuring that the environmental characteristics of each analysis area are clear, complete, and accurate.

[0087] In this embodiment of the invention, because the original three-dimensional spatial feature data along the route is extracted from a high-precision spatial reference frame, and after preprocessing to eliminate noise and outliers, the precise boundary features are obtained through a three-dimensional convex hull construction algorithm. Then, the route areas with different analysis granularities are dynamically divided in combination with the unmanned vessel's navigation performance parameters and safety margin requirements. The environmental situation information is accurately matched and associated with each area according to the principle of spatiotemporal consistency, and corresponding environmental feature parameter sets are assigned. Therefore, the technical problems of vague route space boundary definition, fixed area division granularity lacking adaptability, and inaccurate association between environmental information and route areas leading to insufficient targeting of subsequent optimization decisions in the prior art are effectively overcome. Thus, the refined and differentiated division of the route space is realized, making the environmental feature parameters of each area clear.

[0088] like Figure 2 As shown, in another preferred embodiment of the present invention, a trajectory correction strategy is obtained by comprehensively evaluating the multi-dimensional features of each region, thereby dynamically optimizing and smoothing the initial navigation path to obtain an optimized global reference navigation path, including:

[0089] By receiving the environmental characteristic parameter set assigned to each analysis area, a multi-dimensional quantitative assessment of the navigation risk coefficient, energy efficiency coefficient, and obstacle avoidance safety coefficient within each analysis area is conducted based on the environmental characteristic parameter set, resulting in a regional navigation characteristic assessment matrix. Specifically, this includes: relying on the complete environmental characteristic parameter set assigned to each analysis area, conducting multi-dimensional quantitative assessments to accurately calculate the navigation risk coefficient, energy efficiency coefficient, and obstacle avoidance safety coefficient for each analysis area; the navigation risk coefficient assessment combines meteorological influence parameters, water flow vector field data, and obstacle distribution within the area. When the wind speed exceeds 15 meters per second, the wave height exceeds 2 meters, or the obstacle density is greater than 5 obstacles per square kilometer, the risk coefficient increases, ultimately quantifying to 0. Values ​​ranging from 0 to 10 are used to determine the risk level. Values ​​of 6 and above indicate high risk. The energy efficiency coefficient is calculated based on the angle between the current direction and the ship's direction of travel, as well as the influence of water depth and topography on navigation resistance. The coefficient is higher when sailing downstream and in suitable water depth, and is quantified as a value from 0 to 10. Values ​​of 3 and below indicate low energy efficiency. The obstacle avoidance safety coefficient is determined based on the shortest distance between the obstacle and the preset route, and the obstacle's movement status. The coefficient decreases when the distance is less than 300 meters or when there is a risk of intersection between the dynamic obstacle's trajectory and the obstacle. The coefficient is quantified as a value from 0 to 10, and values ​​of 4 and below indicate low safety. The three coefficients are arranged in the order of the analysis area to form a regional navigation characteristic assessment matrix that includes the area number, navigation risk coefficient, energy efficiency coefficient, and obstacle avoidance safety coefficient.

[0090] Based on the regional navigation feature evaluation matrix, and combined with the navigation performance constraints and mission priority parameters of each unmanned surface vessel (USV), a segmented weighted cost function is constructed. The optimal trajectory correction offset for each segment is calculated through a dynamic weight allocation mechanism, resulting in a trajectory correction strategy set. Specifically, this includes: constructing a segmented weighted cost function based on the regional navigation feature evaluation matrix and the navigation performance constraints and mission priority parameters of each USV; USV navigation performance constraints include inherent parameters such as a maximum turning angle of 30 degrees, a minimum speed of 2 knots, and a maximum range of 500 nautical miles; mission priority parameters are divided into high priority, medium priority, and low priority. Low priority tasks are assigned different weighting benchmarks. A dynamic weighting allocation mechanism flexibly adjusts weights based on regional navigation characteristics and mission requirements: high-priority tasks have an energy efficiency coefficient weight of 0.4, a navigation risk coefficient weight of 0.4, and an obstacle avoidance safety coefficient weight of 0.2 in the evaluation matrix, prioritizing mission timeliness and safety; medium-priority tasks have all three coefficient weights set to 0.33, balancing risk, energy consumption, and safety; low-priority tasks have an energy efficiency coefficient weight of 0.5, a navigation risk coefficient weight of 0.2, and an obstacle avoidance safety coefficient weight of 0.3, focusing on reducing fuel consumption. For high-risk areas, the navigation risk coefficient weight is temporarily increased to 0.5; for areas with low energy efficiency, the energy efficiency coefficient weight is increased to 0.45. This is achieved through a weighted cost function. ,in, It is the first The weighting of navigation risk coefficients for each segment. It is the first Weighting of energy efficiency coefficients for each flight segment It is the first The obstacle avoidance safety coefficient weight for each flight segment It is the first The navigation risk coefficient for each segment. It is the first Energy efficiency coefficient for each flight segment It is the first Obstacle avoidance safety coefficient for each flight segment These are the segment index parameters. The optimal track correction offset for each segment is calculated, with the offset range controlled within 0 to 500 meters. This ensures that the corrected path does not deviate from the core range of the original route while avoiding unfavorable areas, forming a set of track correction strategies that includes the correction direction, offset distance, and implementation priority for each segment.

[0091] The trajectory correction strategy set is applied segment by segment to the initial navigation path, and the path nodes are dynamically offset to obtain a preliminary optimized navigation path sequence. Specifically, this involves: applying the trajectory correction strategy set segment by segment to the initial navigation path, and dynamically offsetting each node in the path. For segments corresponding to high-risk areas, the path nodes are offset towards the low-risk area side according to the offset direction in the correction strategy set. The offset distance is determined based on the risk level: 300 to 500 meters for high-risk areas and 150 to 300 meters for medium-risk areas. For segments with low energy efficiency, the path nodes are adjusted to reduce the angle between the ship's navigation direction and the current direction to within 30 degrees, reducing navigation resistance. For segments with insufficient obstacle avoidance safety factors, the nodes are adjusted to bypass obstacles based on their location and movement, ensuring a minimum distance of 300 meters. During the adjustment process, the continuity between nodes in each segment is maintained to avoid path breaks, forming a preliminary optimized navigation path sequence.

[0092] By performing Bézier curve smoothing on the initially optimized navigation path sequence, sharp corners and discontinuities in the path are eliminated, resulting in a global reference navigation path with continuous curvature changes. Specifically, this involves: performing Bézier curve smoothing on the initially optimized navigation path sequence to eliminate sharp corners and discontinuities; using the nodes of the initially optimized path as control points during the process; setting curve parameters based on node spacing and corner size; adding intermediate control points in node areas with corners greater than 30 degrees to make the curve transition smoother; using curve fitting technology to transform the broken line corners in the original path into continuously changing curvature, ensuring that the rate of change of the ship's turning angle does not exceed 5 degrees per second, avoiding increased fuel consumption and maneuverability caused by sharp corners; after smoothing, the path maintains its adaptability to the environmental characteristics of each analysis area, without deviating from the optimization objective, ultimately forming a global reference navigation path with continuous curvature changes that balances safety, energy economy, and mission timeliness.

[0093] In this embodiment of the invention, by receiving environmental feature parameter sets of each analysis area and performing multi-dimensional quantitative evaluation of navigation risk, energy efficiency, and obstacle avoidance safety to obtain a regional navigation feature evaluation matrix, constructing a segmented weighted cost function by combining unmanned vessel navigation performance constraints and task priorities, calculating the optimal trajectory correction offset for each segment through a dynamic weight allocation mechanism, adjusting the initial path nodes segment by segment, and then smoothing through Bézier curves, the technical means effectively overcome the technical problems in existing path optimization, such as lack of comprehensive consideration of multi-dimensional regional features, fixed and rigid weight allocation, insufficient targeted path adjustment, and sharp turns leading to high navigation energy consumption and high safety risks. Thus, dynamic and precise trajectory optimization is achieved, ensuring that the path avoids high-risk areas, adapts to task priorities and navigation performance constraints, and has continuous curvature, reducing navigation energy consumption and control difficulty.

[0094] In a preferred embodiment of the present invention, based on the optimized navigation path and the current fuel status of each unmanned vessel, it is determined whether refueling is required. If so, a refueling scheduling plan is generated, including:

[0095] By receiving optimized global reference navigation path data and the current real-time fuel status information of each unmanned vessel (UV), the projected fuel consumption of each UV in each segment is calculated based on the global reference navigation path data. A fuel remaining quantity prediction curve is then generated by combining this with the current real-time fuel status information. Specifically, this involves: first, receiving optimized global reference navigation path data and current real-time fuel status information from each UV. The global reference navigation path data includes detailed information such as the specific length of each segment, environmental characteristic parameters of the areas traversed, and projected navigation speed. The current real-time fuel status information includes the vessel's current remaining fuel quantity, fuel consumption rate for the current segment, and fuel consumption rate. Key data such as fuel tank capacity are collected; combined with environmental conditions for each segment, the estimated fuel consumption of each unmanned vessel in the corresponding segment is calculated. For segments sailing downstream with relatively small waves, the consumption rate is calculated at 10 liters per nautical mile; for segments sailing upstream or with waves exceeding 1.5 meters, the consumption rate is calculated at 15 liters per nautical mile; for segments navigating areas with dense obstacles requiring frequent turns, the consumption rate is calculated at 12 liters per nautical mile. The estimated fuel consumption for each segment is added up sequentially, and combined with the current remaining fuel, a fuel remaining prediction curve is generated in chronological order of navigation. The curve clearly shows the trend of the estimated remaining fuel at different points in time during the entire voyage.

[0096] Based on the fuel balance prediction curve, a safe fuel threshold parameter is set. When the predicted fuel balance is lower than the safe fuel threshold, a fuel replenishment demand determination mechanism is triggered to obtain a list of unmanned vessels that need fuel replenishment and a replenishment priority sequence. Specifically, based on the generated fuel balance prediction curve, a unified safe fuel threshold parameter of 500 liters is set. This threshold comprehensively considers the additional fuel demand of the vessel in response to sudden environmental changes such as a sudden increase in wind and waves, the minimum fuel consumption for the remaining voyage, and the amount of fuel required for emergency maneuvers. The system continuously monitors the fuel remaining quantity prediction curve. When the predicted fuel remaining quantity of a certain unmanned vessel falls below 500 liters at any time point, the fuel replenishment demand determination mechanism is immediately triggered, and the vessel is added to the list of unmanned vessels requiring fuel replenishment. Subsequently, the vessels in the list are sorted according to their mission priority parameters and fuel shortage level to form a replenishment priority sequence. Vessels with higher mission priority are ranked first. If mission priorities are the same, vessels with predicted fuel remaining quantities below 300 liters are prioritized over vessels with remaining quantities between 300 and 500 liters. If fuel remaining quantities are similar, vessels with more critical navigation progress, such as those closer to the mission objective and with more pressing mission deadlines, are given higher priority.

[0097] Based on the list of unmanned surface vessels (USVs) requiring refueling and their refueling priority sequence, and considering the intersection points of each USV's navigation paths, marine environmental conditions, and the current location and available fuel inventory of the unmanned refueling vessels, a spatiotemporal conflict resolution algorithm is used to determine the final refueling point location and refueling time window, resulting in a preliminary fuel refueling scheduling plan. Specifically, this plan includes: comprehensively collecting navigation path data for each USV based on the list of USVs requiring refueling and their refueling priority sequence; identifying the intersection points of all USV navigation paths through path analysis, which serve as the core selection range for candidate refueling points; and simultaneously screening the marine environmental conditions of candidate refueling points, eliminating areas with wave heights exceeding 2 meters, water depths less than 3 meters, or dense obstacles, and selecting candidate refueling points with calm waves, suitable water depths, and no navigational safety hazards. Based on the current location of the unmanned refueling vessels, the travel distance and estimated travel time from each unmanned refueling vessel to each candidate refueling point are calculated, and the candidate refueling point that is closest and can be reached the fastest is selected first. At the same time, the available fuel inventory of the unmanned refueling vessels is checked to ensure that the type of fuel they carry is completely matched with the needs of the receiving vessels. In the case of multiple receiving vessels selecting the same candidate refueling point, the refueling time window is adjusted through a spatiotemporal conflict resolution algorithm. If the time interval between the estimated arrival time of two receiving vessels at the same refueling point is less than 30 minutes, the refueling time of the vessel with lower priority is postponed by 40 minutes, or its travel path is slightly adjusted to extend the refueling time interval to more than 40 minutes. Finally, the final refueling point coordinates and refueling time window of each receiving vessel are determined, forming a preliminary fuel refueling scheduling plan.

[0098] By conducting a multi-objective optimization evaluation of the preliminary fuel replenishment scheduling plan, comprehensively considering replenishment timeliness, navigation interference, and fuel type matching, a final fuel replenishment scheduling plan is generated through a dynamic weight adjustment mechanism. The final fuel replenishment scheduling plan includes the receiver vessel identifier, replenishment point coordinates, fuel type, replenishment quantity, estimated arrival time, and priority parameters. Specifically, this includes: conducting a multi-objective optimization evaluation of the preliminary fuel replenishment scheduling plan, with evaluation indicators including replenishment timeliness, navigation interference, and fuel type matching; and establishing a dynamic weight allocation mechanism, where for high-priority receiver vessels, the weight for replenishment timeliness is set to 0.4, the weight for navigation interference is set to 0.3, and the weight for fuel type matching is set to 0. 3; For oil receivers with medium-priority missions, the weights of all three indicators are set to 0.33; for oil receivers with low-priority missions, the weights for navigation interference are set to 0.4, replenishment timeliness to 0.3, and fuel type matching to 0.3. In the replenishment timeliness assessment scheme, the time difference between the expected replenishment completion time and the oil receiver's fuel depletion warning time is considered, with a larger difference indicating better timeliness; the navigation interference assessment considers whether the replenishment point and process will affect the normal navigation of other vessels, whether it is close to busy waterways or other vessels' critical operating areas, with no interference resulting in the highest score; the fuel type matching must ensure that the fuel type of the unmanned refueling vessel is completely consistent with the needs of the oil receiver, and if they are inconsistent, the scheme will be directly eliminated. The initial plan is adjusted based on the evaluation results. If the refueling timeliness of a certain plan is insufficient, the sailing speed of the receiving vessel or the location of the refueling point can be fine-tuned to shorten the refueling waiting time. If the navigation interference is high, a new candidate refueling point with no surrounding vessels is selected. After multiple rounds of optimization, the final fuel refueling scheduling plan is generated. The plan includes the unique equipment identifier of the receiving vessel, the precise latitude and longitude coordinates of the refueling point, the specific type of fuel required, the precise refueling quantity, the estimated arrival time of the receiving vessel and the unmanned refueling vessel, and the priority parameters of the refueling task.

[0099] In this embodiment of the invention, by receiving optimized global reference navigation path data and real-time fuel status information of each unmanned vessel, calculating the expected fuel consumption of each segment and generating a fuel reserve prediction curve, setting a safe fuel threshold to trigger resupply demand determination to determine the resupply list and priority sequence, and combining the unmanned vessel path intersection points, marine environment, and unmanned refueling vessel status to determine resupply points and time windows through a spatiotemporal conflict resolution algorithm, and then generating a final scheduling scheme through multi-objective optimization evaluation that takes into account resupply timeliness, navigation interference, fuel type matching degree, and dynamic weight adjustment, this invention effectively overcomes the technical problems of inaccurate fuel resupply demand determination, unreasonable planning of resupply points and time windows, frequent spatiotemporal conflicts in multi-vessel resupply, and insufficient adaptability and practicality of scheduling schemes that do not comprehensively consider multiple dimensions of factors in the existing technology. This achieves accurate identification of fuel resupply demand, reasonable allocation of resupply resources, and effective resolution of spatiotemporal conflicts. The generated scheduling scheme is both targeted and optimized, providing a scientific and feasible execution basis for efficient coordination of unmanned vessel swarm mission planning and fuel resupply.

[0100] In a preferred embodiment of the present invention, a fuel replenishment scheduling scheme is used to schedule an unmanned refueling vessel to a corresponding refueling point, and to perform precise fuel refueling by controlling a robotic arm to dock with the receiving vessel's fuel inlet. This includes:

[0101] By analyzing the receiver vessel identification information, refueling point location, fuel type, and quantity parameters in the fuel replenishment scheduling plan, the mission instructions for the unmanned refueling vessel are obtained. Specifically, this includes: first, a comprehensive analysis of the fuel replenishment scheduling plan, extracting key information from each part of the plan; receiver vessel identification information, which is the unique equipment code of each receiver vessel, used to accurately locate the target vessel requiring replenishment; refueling point location information, which includes specific latitude and longitude coordinates, ensuring that the unmanned refueling vessel can accurately navigate to the designated area; fuel type information, which is clearly labeled as specific types such as diesel, to avoid fuel compatibility errors; and quantity parameters, which are the specific fuel refueling volumes, serving as the basis for accurate refueling; and finally, logically integrating the extracted information to form the mission instructions for the unmanned refueling vessel.

[0102] Based on the mission execution instructions, the designated unmanned refueling vessel is dispatched to obtain a dedicated route to the refueling point from its current location using a segmented path planning method. The navigation status and estimated arrival time of the unmanned refueling vessel are monitored in real time through a satellite communication network. Specifically, based on the generated mission execution instructions, the designated unmanned refueling vessel to undertake this refueling mission is determined. The unmanned refueling vessel departs from its current location and designs a dedicated route to the refueling point using a segmented path planning method. During the planning process, the route segments are divided according to the characteristics of the marine environment. In open, unobstructed waters with gentle currents, the route segment length is set to 10 nautical miles to improve navigation efficiency. In areas near islands, reefs, or busy waterways, the route segment length is set to 2 nautical miles to ensure navigation safety. Each segment has a planned final navigation route to ensure smooth transitions between segments. At the same time, the navigation status of the unmanned refueling vessel is monitored in real time via satellite communication network. The monitoring content includes real-time speed, heading, current position, fuel consumption, and equipment operating parameters. The estimated arrival time is dynamically updated based on the real-time navigation status. If the navigation speed fluctuates due to wind, waves, or changes in current, the estimated arrival time is adjusted in a timely manner to ensure a smooth connection between the receiving vessel and the unmanned refueling vessel.

[0103] When the unmanned refueling vessel reaches the preset safe distance range of the refueling point, the relative position and attitude information between the unmanned refueling vessel and the receiving vessel are acquired to obtain high-precision berthing control commands. Specifically, when the unmanned refueling vessel sails within the preset 500-meter safe distance range of the refueling point, the multi-source sensor collaborative working mode is activated; the relative position data between the unmanned refueling vessel and the receiving vessel are simultaneously collected through lidar, millimeter-wave radar, and visual sensors, including the straight-line distance, latitude difference, longitude difference, and vertical height difference between the two vessels; the attitude information of the two vessels is also collected, covering their respective pitch angle, roll angle, and heading angle; the collected data is transmitted to the control center in real time, and the control center performs comprehensive analysis and calculation on the data to generate high-precision berthing control commands, instructing the unmanned refueling vessel's berthing speed, turning angle, and adjustment frequency to ensure that the unmanned refueling vessel gradually approaches the receiving vessel with a stable attitude, ultimately keeping the two vessels in a relatively stationary state, and controlling the distance between the hulls within 0.3 meters.

[0104] Based on high-precision berthing control commands, the robot arm's motion trajectory is planned by identifying the type of oil port on the receiving vessel in real time, determining its specifications and spatial coordinates. Specifically, this includes: after berthing according to the high-precision berthing control commands, activating a vision recognition system to identify the oil port on the receiving vessel in real time; capturing images of the oil port area using vision sensors and determining the oil port type through image analysis technology; simultaneously, accurately measuring the specifications of the oil port, including its diameter, interface depth, and thread type, and calculating its precise coordinates in three-dimensional space; and combining the type, specifications, and spatial coordinates of the oil port to plan the motion trajectory of the dual-redundant six-degree-of-freedom robot arm. The trajectory planning process fully considers the robot arm's range of motion, joint rotation limits, and obstacle avoidance requirements, designing the optimal motion path from the robot arm's initial position to the target oil port to ensure that the robot arm does not collide with the hulls or equipment of either vessel during its movement.

[0105] By planning the motion trajectory of the robotic arm, the end effector of the dual-redundant six-degree-of-freedom robotic arm is controlled to move towards the target fuel inlet. Through real-time sensing of contact force and visual feedback, the robotic arm's motion posture is dynamically adjusted to achieve precise docking between the refueling nozzle and the fuel inlet. Specifically, this includes: controlling the end effector of the dual-redundant six-degree-of-freedom robotic arm to move towards the target fuel inlet according to the planned motion trajectory; the force sensor integrated into the end effector senses the contact force with the fuel inlet in real time, with a contact force threshold set to 5 Newtons. When the sensed contact force approaches or reaches 5 Newtons, the robotic arm's motion force is adjusted accordingly; simultaneously, the visual sensor continuously provides feedback on the relative positional deviation between the end effector and the fuel inlet, and the robotic arm's motion posture is dynamically adjusted based on the deviation data, including horizontal offset, vertical lifting and lowering, and angular rotation; through dynamic adjustment using a hybrid force and visual feedback, the refueling nozzle is precisely aligned with the fuel inlet, slowly inserted, and sealed, ensuring no fuel leakage during the docking process, thus achieving precise docking between the refueling nozzle and the fuel inlet.

[0106] After successful docking, the corresponding fuel tank is activated according to the fuel type and quantity requirements in the fuel supply scheduling plan. A high-precision mass flow meter monitors the fuel flow rate in real time, and precise fuel refueling is performed until the preset refueling volume is reached. Specifically, after confirming successful docking and proper sealing between the refueling nozzle and the fuel inlet, the fuel delivery system of the corresponding fuel tank is activated according to the fuel type requirements in the fuel supply scheduling plan. If the plan requires diesel fuel, the delivery pump of the dedicated diesel fuel tank is activated; if aviation kerosene is required, the delivery pump of the dedicated aviation kerosene fuel tank is activated. During delivery, the fuel flow rate is monitored in real time using a high-precision mass flow meter, and the volume of fuel delivered is accurately recorded. The monitored flow rate data is compared with the preset refueling volume in real time. When the cumulative delivered fuel volume reaches the preset refueling volume, the fuel delivery system is automatically shut down, and the refueling operation stops. The fuel delivery pressure and inlet temperature are monitored simultaneously during refueling. If an abnormal increase in pressure or a temperature exceeding the normal range occurs, an emergency shutdown procedure is immediately initiated to ensure a safe and controllable refueling process.

[0107] In this embodiment of the invention, by employing technical means such as analyzing key parameters of the fuel replenishment scheduling scheme to generate unmanned refueling vessel mission instructions, formulating dedicated routes for refueling vessels through segmented path planning and monitoring navigation status in real time with the help of satellite communication, generating high-precision berthing control instructions based on relative position and attitude information, real-time identification of oil port type to plan the movement trajectory of the robotic arm, dynamically adjusting the attitude through force and visual feedback of the dual-redundant six-degree-of-freedom robotic arm to complete precise docking, and using a high-precision mass flow meter to monitor flow in real time to perform precise refueling, the invention effectively overcomes the technical problems of insufficient route planning, low berthing and docking accuracy, weak robotic arm adaptability to different oil ports, inaccurate refueling volume control, and lack of real-time monitoring throughout the process leading to poor refueling connection in existing unmanned refueling processes. This achieves precise scheduling of unmanned refueling vessels, reliable berthing with receiving vessels, and efficient oil port adaptation and docking, ensuring the accuracy and safety of fuel refueling and providing solid technical support for the efficient implementation of unmanned vessel swarm fuel replenishment.

[0108] In a preferred embodiment of the present invention, during navigation and refueling, the real-time status of all vessels is monitored, and fault warnings and comprehensive status assessments are performed. If a sudden emergency event is detected, emergency operations are autonomously executed according to the event type, including:

[0109] The system receives real-time ship operation status data uploaded by various unmanned surface vessels (USVs) and unmanned refueling vessels via satellite communication networks. Based on this data, a real-time ship status vector set is generated. Specifically, during the entire process of USVs performing missions and unmanned refueling vessels conducting refueling, a real-time data transmission link is established via satellite communication networks to continuously receive ship operation status data uploaded by all participating USVs and unmanned refueling vessels. The data covers multiple core parameters, including real-time navigation parameters such as speed, heading, pitch angle, and roll angle; fuel parameters such as remaining fuel quantity, fuel pressure, and fuel temperature; power parameters such as engine speed, propeller power, and operating current; refueling equipment parameters such as robotic arm joint angle, refueling nozzle pressure, and sealing status; and environmental parameters such as real-time wind speed, wave height, and water current speed. All received data is categorized and organized according to the unique identifier of each vessel, and combined with the data collection timestamp, a real-time ship status vector set containing various operation status parameters for each vessel is generated.

[0110] By extracting multi-dimensional features from the real-time ship status vector set and performing pattern matching with a pre-set fault feature library, the fault risk index of each ship is calculated. When the fault risk index exceeds a preset threshold, a fault warning signal is issued and the abnormal ship is identified. Specifically, this involves: extracting multi-dimensional features from the generated real-time ship status vector set, including key features reflecting the stability of the ship's operating status such as the rate of change of speed, the amplitude of course fluctuations, the rate of decrease in fuel pressure, the deviation value of robotic arm joint motion, and the range of thruster power fluctuations; and simultaneously calling the pre-set fault feature library, which stores various common faults. Feature patterns, such as sudden drop in fuel pressure corresponding to fuel leakage, excessive joint movement deviation corresponding to robotic arm jamming, and abnormal power fluctuation corresponding to thruster failure, are used to identify and match the extracted multi-dimensional features with the feature patterns in the fault feature library one by one. The fault risk index of each ship is calculated based on the degree of matching. The fault risk index is set to a value of 0 to 100, where 0 to 30 is low risk, 31 to 60 is medium risk, and 61 to 100 is high risk. The preset fault risk threshold is 70. When the fault risk index of a ship exceeds 70, a fault warning signal is immediately generated and the unique identifier of the ship is accurately marked.

[0111] Based on fault warning signals and real-time ship status vector sets, a comprehensive status assessment is conducted on abnormal ships to calculate their navigation safety level. Simultaneously, the collaborative operation capabilities of normal ships are dynamically assessed, generating a ship status assessment report that includes both safety level and operational capabilities. Specifically, this includes: conducting a comprehensive status assessment of ships marked as abnormal based on generated fault warning signals and corresponding real-time ship status vector sets; calculating the ship's navigation safety level by combining the abnormal ship's fault characteristic parameters, fault risk index, and current navigation environment, with five levels: Level 1 (Safe and able to operate normally), Level 2 (Basic safety requiring close monitoring), Level 3 (Minor danger requiring adjustment of operational status), Level 4 (Moderate danger requiring preparation of emergency measures), and Level 5 (High danger requiring immediate cessation of operations); simultaneously, a dynamic assessment of the collaborative operation capabilities of normal ships that have not triggered fault warnings is conducted, with assessment indicators including the normal ship's remaining fuel, current navigation position, distance from the abnormal ship, power redundancy, and onboard emergency equipment such as oil booms and firefighting equipment, to determine their ability to assist in rescue or take over operations; and generating a ship status assessment report that includes the safety level of each abnormal ship, the risk impact range of suspected fault locations, and the collaborative operation capability level of each normal ship.

[0112] Based on the ship condition assessment report, the types of sudden emergency events are identified. A pre-set emergency operation strategy library is matched to these event types to obtain corresponding emergency operation instruction sets. Specifically, this includes: accurately identifying and classifying the types of sudden emergency events based on the generated ship condition assessment report. Emergency event types include fuel system failures such as fuel leaks and fuel pump malfunctions; power failures such as engine shutdowns and propeller failures; refueling equipment failures such as stuck robotic arms, inability to detach refueling nozzles, and seal failures; navigation safety events such as collision risks, grounding risks, and extreme weather effects; and other emergencies such as hull damage and fires. The pre-set emergency operation strategy library is invoked. This library pre-sets standardized operating procedures and handling plans for different types of emergency events. For example, for fuel leak events, the pre-set operating procedures include emergency shutdown, safe anchoring, activation of emergency leak-sealing equipment, and dispatching nearby vessels to deploy oil booms; for robotic arm jamming events, the pre-set operating procedures include joint reset, emergency retraction, and activation of the backup robotic arm. Based on the identified emergency event type, the corresponding emergency operation instruction set is matched from the strategy library. This instruction set includes the required operating steps, operating sequence, involved equipment, and key parameters such as anchoring position and shutdown timing.

[0113] Emergency operation instructions are issued to the malfunctioning vessel and its collaborating vessels. The malfunctioning vessel is then instructed to perform emergency shutdown, safe anchoring, and return to port. Simultaneously, collaborating vessels are dispatched to perform auxiliary rescue tasks. The effectiveness of the emergency operations is verified through real-time feedback data, and the emergency strategy is dynamically adjusted based on the results until the vessel's condition returns to normal. Specifically, this includes: accurately issuing the matched emergency operation instructions to the corresponding malfunctioning vessel and collaborating vessels via satellite communication network; upon receiving the instructions, the malfunctioning vessel immediately executes the corresponding operations. If it is a highly dangerous event such as a fuel leak, an emergency shutdown is first performed to cut off power output, and then the vessel's attitude is adjusted to a safe anchoring position according to the instructions to complete anchoring, while simultaneously activating the onboard emergency leak-sealing equipment; if it is a malfunction of the refueling equipment such as a stuck robotic arm, the refueling operation is immediately stopped, and the robotic arm is controlled to attempt to reset. If the reset fails, an emergency retreat procedure is initiated. After receiving instructions, the cooperating vessels execute auxiliary rescue tasks according to the instructions, such as dispatching the nearest normal vessel to the incident area to deploy oil booms, or dispatching unmanned vessels with repair capabilities to the scene with repair equipment to provide support. During the emergency operation, real-time feedback data uploaded by the abnormal vessel and cooperating vessels is continuously received, including the fault control status, such as whether the leak has been contained, whether the robotic arm has returned to normal, and whether the vessel's attitude is stable, as well as the progress of the rescue mission, such as the completion of the oil boom deployment and the arrival of repair equipment. The effectiveness of the emergency operation is verified based on the feedback data. If the fault is not effectively controlled, such as the leakage continues to increase, or the rescue progress does not meet expectations, the emergency strategy is dynamically adjusted, such as increasing the number of cooperating rescue vessels, adjusting the working positions of rescue vessels, and activating higher-level emergency equipment, until the abnormal vessel returns to normal and all vessels can resume normal operations or safely evacuate.

[0114] In this embodiment of the invention, by employing a technique of receiving real-time ship operational status data via satellite communication network and generating a real-time ship status vector set, extracting multi-dimensional features from the vector set, and combining this vector set with a fault feature library pattern matching to calculate a fault risk index for fault early warning, comprehensive status assessment of abnormal ships, dynamic assessment of the collaborative operation capabilities of normal ships to generate ship status reports, matching a preset strategy library to obtain an operation instruction set based on the type of emergency event, issuing instructions to control abnormal ships to perform corresponding emergency operations and dispatching collaborative ships to carry out auxiliary rescue, and dynamically adjusting emergency strategies through real-time feedback data, this invention effectively overcomes the technical problems of existing technologies, such as lack of real-time ship status monitoring during navigation and refueling, delayed fault early warning, lack of coordination in emergency operations, inability to dynamically adapt strategies leading to high safety risks, and easy interruption of operations. This achieves accurate monitoring of the ship's operational status throughout the entire process and timely fault early warning, as well as rapid identification, accurate handling, and collaborative rescue of sudden emergency events, improving the system's fault tolerance and safety, and ensuring the continuity and stability of unmanned vessel swarm operations.

[0115] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the system as described above. All implementations in the above system embodiments are applicable to this embodiment and can achieve the same technical effects.

[0116] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the system as described above. All implementations in the above system embodiments are applicable to this embodiment and can achieve the same technical effects.

[0117] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An unmanned surface vessel (USV) mission planning and fuel replenishment coordinated scheduling system, characterized in that, include: The acquisition module is used to acquire mission information from multiple unmanned vessels. Based on mission information, a segmented path planning method is used to generate an initial navigation path for each unmanned vessel from its current position to the mission target position. The module is used to collect and fuse environmental and vessel status data along the initial navigation path of each unmanned vessel to build a unified spatiotemporal benchmark for environmental situation information and spatial reference framework. The calculation module is used to extract three-dimensional spatial feature data along the route based on a spatial reference frame and to calculate spatial boundary features. Based on boundary characteristics, the airway space is divided into multiple analysis regions, and environmental situation information is correlated to each region accordingly. The optimization module is used to comprehensively evaluate the multi-dimensional features of each region, obtain the trajectory correction strategy, and then dynamically optimize and smooth the initial navigation path to obtain the optimized global reference navigation path. The judgment module is used to determine whether refueling is needed based on the optimized navigation path and the current fuel status of each unmanned vessel. If so, it generates a refueling scheduling plan. The processing module is used to dispatch unmanned refueling vessels to corresponding refueling points according to the fuel supply scheduling scheme, and to perform precise fuel refueling by controlling the robotic arm to dock with the oil receiving vessel's fuel inlet. The monitoring module is used to monitor the real-time status of all ships during navigation and refueling, and to provide fault warnings and comprehensive status assessments. If a sudden emergency event is detected, it will autonomously execute emergency operations according to the event type.

2. The unmanned vessel mission planning and fuel replenishment coordinated scheduling system according to claim 1, characterized in that, Acquire mission information from multiple unmanned vessels; Based on mission information, a segmented path planning method is used to generate an initial navigation path for each unmanned surface vessel from its current position to the mission objective position, including: Receive mission information from multiple unmanned vessels issued by the mission platform; By analyzing the received mission information, the mission start point, mission target point, and mission priority parameters corresponding to each unmanned vessel are extracted. Based on the extracted mission start point and mission target point, a segmented path planning method is used to generate an initial navigation path for each unmanned vessel from its current position to the mission target position.

3. The unmanned vessel mission planning and fuel replenishment coordinated scheduling system according to claim 2, characterized in that, Along the initial navigation paths of each unmanned vessel, environmental and vessel status data are collected and fused to construct a unified spatiotemporal benchmark for environmental situation information and a spatial reference framework, including: The system receives raw environmental data collected in real time by the multi-source sensors carried by each unmanned vessel during navigation via a satellite communication network. The raw environmental data includes positioning information, lidar point cloud data, millimeter-wave radar obstacle detection data, and visual environmental feature data. Based on the original environmental data, heterogeneous data are synchronized in time and transformed in space by using a unified spatiotemporal benchmark to generate a standardized environmental dataset with spatiotemporal consistency. By fusing standardized environmental datasets with pre-loaded digital marine map information, a three-dimensional environmental situation information covering the navigation paths of each unmanned vessel is constructed. Based on three-dimensional environmental situation information, combined with the historical navigation trajectory data and current real-time status parameters of each unmanned vessel, a set of ship dynamic characteristic parameters is calculated. The ship dynamic characteristic parameter set is then spatiotemporally correlated with the three-dimensional environmental situation information to obtain ship-environment interaction reference data. By performing integrity verification and outlier detection processing on the ship-environment interaction reference data, outlier data points are removed and data interpolation compensation is performed, ultimately generating high-precision environmental situation information and spatial reference framework with a unified spatiotemporal benchmark.

4. The unmanned vessel mission planning and fuel replenishment coordinated scheduling system according to claim 3, characterized in that, Based on a spatial reference frame, three-dimensional spatial feature data along the route are extracted, and spatial boundary features are calculated. Based on boundary characteristics, the airway space is divided into multiple analysis regions, and environmental situation information is correlated to each region accordingly, including: By using high-precision environmental situation information and a spatial reference frame with a unified spatiotemporal benchmark, the original three-dimensional spatial features along the routes of each unmanned vessel are extracted from the spatial reference frame. The original three-dimensional spatial features include water depth and topography data, obstacle spatial distribution data, water flow vector field data, and meteorological influence parameters. By performing data preprocessing operations on the original three-dimensional spatial feature data to eliminate noise and outliers, a standardized three-dimensional spatial feature dataset is formed. The standardized three-dimensional spatial feature dataset is then input into a three-dimensional convex hull construction algorithm to calculate the precise boundary feature data of the airway space. Based on the precise boundary feature data of the airway space, combined with the navigation performance parameters and safety margin requirements of each unmanned vessel, the airway space is dynamically divided into multiple analysis regions with different analysis granularities. The constructed environmental situation information is accurately matched and associated with multiple analysis regions according to the principle of spatiotemporal consistency, and a corresponding set of environmental feature parameters is assigned to each analysis region.

5. The unmanned vessel mission planning and fuel replenishment coordinated scheduling system according to claim 4, characterized in that, By comprehensively evaluating the multi-dimensional characteristics of each region, a trajectory correction strategy is derived, thereby dynamically optimizing and smoothing the initial navigation path to obtain an optimized global reference navigation path, including: By receiving the set of environmental characteristic parameters assigned to each analysis area, the navigation risk coefficient, energy efficiency coefficient and obstacle avoidance safety coefficient of each analysis area are quantitatively evaluated in multiple dimensions based on the set of environmental characteristic parameters, and the regional navigation characteristic evaluation matrix is ​​obtained. Based on the regional navigation feature evaluation matrix, combined with the navigation performance constraints and mission priority parameters of each unmanned vessel, a segmented weighted cost function is constructed. The optimal trajectory correction offset for each segment is calculated through a dynamic weight allocation mechanism, and a trajectory correction strategy set is obtained. The set of trajectory correction strategies is applied segment by segment to the initial navigation path, and the path nodes are dynamically offset to obtain a preliminary optimized navigation path sequence. By performing Bézier curve smoothing on the initially optimized navigation path sequence, sharp turns and discontinuities in the path are eliminated, resulting in a global reference navigation path with continuous curvature variation.

6. The unmanned surface vessel mission planning and fuel replenishment coordinated scheduling system according to claim 5, characterized in that, Based on the optimized navigation path and the current fuel status of each unmanned vessel, determine whether refueling is needed. If so, generate a refueling schedule, including: By receiving optimized global reference navigation path data and the current real-time fuel status information of each unmanned vessel, the estimated fuel consumption of each unmanned vessel in each segment is calculated based on the global reference navigation path data, and a fuel balance prediction curve is generated by combining the current real-time fuel status information. Based on the fuel balance prediction curve, a safe fuel threshold parameter is set. When the predicted fuel balance is lower than the safe fuel threshold, the fuel replenishment demand determination mechanism is triggered to obtain a list of unmanned vessels that need fuel replenishment and a replenishment priority sequence. Based on the list of unmanned vessels requiring fuel replenishment and the priority sequence of replenishment, combined with the intersection of the navigation paths of each unmanned vessel, the marine environmental conditions, and the current location and available fuel inventory of the unmanned refueling vessel, the final replenishment point location and replenishment time window are determined through a spatiotemporal conflict resolution algorithm, resulting in a preliminary fuel replenishment scheduling plan. By conducting multi-objective optimization and evaluation of the preliminary fuel replenishment scheduling plan, and comprehensively considering the timeliness of replenishment, navigation interference and fuel type matching, a final fuel replenishment scheduling plan is generated through a dynamic weight adjustment mechanism. The final fuel replenishment scheduling plan includes the receiver vessel identifier, replenishment point coordinates, fuel type, replenishment quantity, estimated arrival time and priority parameters.

7. The unmanned surface vessel mission planning and refueling coordination scheduling system according to claim 6, characterized in that, Through a fuel replenishment scheduling plan, unmanned refueling vessels are dispatched to corresponding refueling points, and their robotic arms dock with the receiving vessel's fuel inlet to perform precise fuel refueling, including: By analyzing the oil receiving vessel identification information, refueling point location, fuel type and quantity parameters in the fuel supply scheduling scheme, the mission instructions of the unmanned refueling vessel are obtained. Based on the mission execution instructions, the designated unmanned refueling vessel is scheduled to obtain a dedicated route to the refueling point from its current position using a segmented path planning method, and the navigation status and estimated arrival time of the unmanned refueling vessel are monitored in real time through a satellite communication network. When the unmanned refueling vessel reaches the preset safe distance range of the refueling point, the relative position and attitude information of the unmanned refueling vessel and the receiving vessel are obtained, and high-precision berthing control commands are obtained. Based on the high-precision berthing control command, the robot arm's motion trajectory is planned by identifying the type of oil port on the receiving vessel in real time, determining the oil port specifications and spatial coordinates. By planning the motion trajectory of the robotic arm, the end effector of the dual redundant six-degree-of-freedom robotic arm is controlled to move towards the target oil port. By sensing the contact force and visual feedback information in real time, the motion posture of the robotic arm is dynamically adjusted to complete the precise docking of the refueling nozzle and the oil port. After confirming successful docking, the corresponding fuel tank is activated according to the fuel type and quantity requirements in the fuel supply scheduling plan. The fuel flow is monitored in real time by a high-precision mass flow meter, and a precise fuel refueling operation is performed until the preset refueling amount is reached.

8. The unmanned surface vessel mission planning and refueling coordination scheduling system according to claim 7, characterized in that, During navigation and refueling, the system monitors the real-time status of all vessels, providing early warnings of malfunctions and comprehensive status assessments. If a sudden emergency is detected, it autonomously executes emergency procedures based on the event type, including: The system receives real-time ship operation status data uploaded by each unmanned vessel and unmanned refueling vessel through a satellite communication network, and generates a real-time ship status vector set based on the ship operation status data. By extracting multi-dimensional features from the real-time status vector set of ships and performing pattern matching in combination with a pre-set fault feature library, the fault risk index of each ship is calculated. When the fault risk index exceeds a pre-set threshold, a fault warning signal is obtained and the abnormal ship is marked. Based on fault warning signals and real-time ship status vector sets, a comprehensive status assessment is conducted on abnormal ships to calculate their navigation safety level. At the same time, the collaborative operation capability of normal ships is dynamically assessed, generating a ship status assessment report that includes both safety level and operational capability. Based on the ship condition assessment report, identify the types of sudden emergency events, match the pre-set emergency operation strategy library based on the event type, and obtain the corresponding emergency operation instruction set; The emergency operation instruction set is issued to the abnormal vessel and its cooperating vessels, controlling the abnormal vessel to perform emergency shutdown, safe anchoring and return to port operations. At the same time, cooperating vessels are dispatched to perform auxiliary rescue tasks. The effectiveness of the emergency operation is verified by real-time feedback data, and the emergency strategy is dynamically adjusted according to the effectiveness until the vessel's status returns to normal.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to perform the system as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, performs the system as described in any one of claims 1 to 8.

Images