Cloud platform intelligent scheduling system applied to unmanned fleet management

Abstract

The invention relates to the technical field of cloud platform intelligent scheduling, in particular to a cloud platform intelligent scheduling system applied to unmanned fleet management, which comprises the steps of identifying ship propulsion, energy and communication states to mark availability, matching a task sea area risk and precision requirement output list, and adjusting ship distribution according to task density. When the ship is abnormal, adjacent ships vote to take over the ship, and finally the ships with high overlapping characteristics are analyzed and dispatched marshalling is reconstructed. According to the method, a state label is generated by extracting multi-dimensional data of a propeller, energy and communication to quantify scheduling availability, and task matching, risk assessment and fault-tolerant boundary demarcation are carried out in combination with sea area storm, an obstacle layer and a positioning precision requirement. Ship scheduling distribution is dynamically adjusted according to operation coherence and task density, task substitution is responded through an abnormal voting mechanism, a fleet is recombined according to path overlapping and navigation collaborative characteristics, and a closed-loop scheduling control system covering the whole process of recognition, matching, adjustment, substitution and marshalling is constructed.

Description

Technical Field

[0001] This invention relates to the field of cloud platform intelligent scheduling technology, and in particular to a cloud platform intelligent scheduling system for the management of unmanned ship fleets. Background Technology

[0002] The field of intelligent scheduling technology in cloud platforms encompasses resource scheduling, task allocation, and optimized management based on cloud computing platforms. Its core content mainly includes how to centrally manage various resources through a cloud platform and coordinate the scheduling of different tasks or devices, thereby improving overall efficiency and response speed. Intelligent scheduling technology in cloud platforms typically involves the optimization of algorithm models, scheduling strategies, data transmission, and processing, aiming to achieve optimal resource allocation and efficient task execution through real-time data analysis and decision support. Intelligent scheduling is increasingly being applied across various industries, especially in large-scale distributed systems, where ensuring reasonable task allocation and effective resource scheduling is a key focus of technological research.

[0003] Among them, the cloud platform intelligent scheduling system for unmanned vessel fleet management mainly addresses the management needs of unmanned vessel fleets, using a cloud platform for fleet scheduling and task allocation. By employing a cloud computing platform, it acquires real-time fleet status information and, combined with vessel positions, navigation tasks, and environmental factors, performs intelligent scheduling and optimization. Through data processing and analysis, it determines the navigation paths and task sequences of each vessel to achieve efficient management and collaborative operation of the unmanned vessel fleet. Specifically, this includes real-time monitoring of vessel positions, dynamic task allocation, and optimized resource allocation based on navigation conditions. It solves the complex task allocation problem faced in unmanned vessel fleet scheduling through intelligent means and utilizes the resource scheduling capabilities of the cloud platform to achieve centralized management and optimized scheduling of the unmanned vessel fleet.

[0004] Existing technologies rely on ship position, mission information, and environmental factors for scheduling decisions, lacking the ability to process the deep characteristics of ship operating status. They cannot accurately identify scheduling suitability in the face of multi-factor interference, and the matching relationship between missions and ships is prone to deviation. The scheduling mechanism does not have a distributed and rapid response method for abnormal missions, resulting in a lack of flexible switching capability for scheduling paths. During mission execution, ship collaboration lacks an optimization mechanism based on trajectory overlap and temporal characteristics, making it difficult to achieve efficient grouping and resource scheduling coordination in areas with high mission density. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies and propose a cloud platform intelligent scheduling system for unmanned ship fleet management.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a cloud platform intelligent scheduling system for unmanned vessel fleet management, the system comprising:

[0007] The ship status identification module acquires thruster vibration records, energy temperature changes and communication stability logs, extracts fluctuation frequency, heating duration and signal interruption status, completes status labeling in sequence, classifies ship scheduling level based on labeling results, and generates ship status scheduling labels.

[0008] The risk task matching module calls the vessel status scheduling tag, processes wind and wave level, obstacle layer and positioning requirements, judges the degree of interference of wind and waves on propulsion stability, assesses the impact of communication interruption on operation in obstacle area, matches tolerance boundary, and outputs task matching scheduling list.

[0009] The energy efficiency density switching module calls the ship task relationship in the task matching and scheduling list, adjusts high-coherence ships to dense areas and low-coherence ships out according to the continuity of operations and the distribution of task-intensive areas, reconstructs task affiliation, and outputs a task density scheduling distribution map.

[0010] The abnormal voting takeover module reads the task execution records in the task density scheduling distribution map, identifies propulsion imbalance, energy anomalies, and communication interruptions, triggers a takeover request, broadcasts to neighboring ships, records feedback and identifies and confirms the response object, and generates an abnormal task replacement assignment record after the replacement is completed.

[0011] As a further embodiment of the present invention, the vessel status scheduling label includes an availability level identifier, a status classification code, and a scheduling adaptation level; the task matching scheduling list includes an environmental adaptability coefficient, a path passability index, and a positioning error tolerance; the task density scheduling distribution map includes high-density operation blocks, low-density scheduling zones, and continuity distribution parameters; and the abnormal task replacement assignment record includes a replacement task number, a takeover response identifier, and a feedback confirmation time stamp.

[0012] As a further aspect of the present invention, the ship status identification module includes a vibration identification submodule, a temperature fluctuation extraction submodule, and a scheduling tag generation submodule;

[0013] The vibration recognition submodule acquires vibration records of the unmanned vessel propulsion system from the cloud platform, divides the continuous recording segments in chronological order, identifies the arrangement pattern of peaks and troughs in the segments, filters the segments based on the number of abrupt changes in waveform and frequency variation, marks the recording segments with vibration characteristic changes, and generates vibration anomaly segment annotation results.

[0014] The temperature fluctuation extraction submodule, based on the time segment corresponding to the vibration anomaly segment annotation result, calls the same segment record during the temperature change process of the energy cabin, sorts out the upward trend and change rhythm of the temperature data, filters out the segments that show rapid temperature rise or fluctuation at the same time as the vibration anomaly segment, summarizes them to form an abnormal behavior chain of temperature response, and generates a set of temperature fluctuation related segments.

[0015] The scheduling tag generation submodule calls the temperature fluctuation associated segment set and vibration anomaly segment annotation results, compares the time marker of the interruption point recorded in the communication antenna communication log, screens the time intersection in the three dimensions of records, identifies the ship number with multiple abnormal behaviors, and classifies the abnormal combination according to the preset scheduling availability standard to generate ship status scheduling tags.

[0016] As a further aspect of the present invention, the risk task matching module includes an operation interference identification submodule, a path obstruction judgment submodule, and a scheduling list generation submodule;

[0017] The operation interference identification submodule obtains wind and wave records of the sea area where the mission is located. Based on the time period and geographical location marked in each record, it calls the tag content related to the propulsion status in the ship status scheduling tag of the corresponding ship, establishes a one-to-one correspondence between wind and wave records and propulsion status, matches them according to the combination of wind and wave intensity level and adaptive description in propulsion status tag, marks the interference-affected segments and organizes their ship numbers, and generates propulsion adaptive interference matching results.

[0018] The path obstruction judgment submodule calls the ship number involved in the propulsion adaptation interference matching result, obtains the communication status label of the position segment on the mission target path, and compares it with the path node data in the obstacle distribution layer in the mission area to identify the intersection range of unstable communication nodes and obstacle points, sorts and classifies the set of nodes with dual restrictions in the path, and generates a path restriction node distribution list.

[0019] The scheduling list generation submodule obtains the positioning control capability level content in the scheduling tag of each vessel number corresponding to each node in the path restriction node distribution list, and filters out the vessel numbers and corresponding task names that can operate within the accuracy tolerance range in the restricted area according to the positioning control accuracy requirements set by the task. The task numbers and executable vessels are mapped one by one to generate a task matching scheduling list.

[0020] As a further aspect of the present invention, the matching based on the combination of the wind and wave intensity level and the adaptive description in the propulsion status label specifically involves: comparing the real-time effective wave height data and real-time average wind speed data in the wind and wave record with the preset propulsion system wave resistance level threshold and propulsion system wind resistance level threshold in the propulsion status label; when the real-time effective wave height data exceeds the propulsion system wave resistance level threshold or the real-time average wind speed data exceeds the propulsion system wind resistance level threshold, the matching is completed.

[0021] The identification of nodes with unstable communication status specifically involves identifying path nodes whose signal packet loss rate is higher than a preset packet loss rate threshold or whose communication delay is higher than a preset delay threshold, as nodes with unstable communication status, as recorded in the communication status tags.

[0022] The process of identifying the intersection range between unstable communication nodes and obstacle points specifically involves: using each obstacle point in the obstacle distribution layer as the geometric center, calculating and generating a dynamic safety buffer zone based on the ship length, maximum speed, and minimum turning radius data of the vessel performing the task, and identifying the unstable communication nodes that fall within the dynamic safety buffer zone as nodes within the intersection range.

[0023] As a further aspect of the present invention, the energy efficiency density switching module includes a continuity extraction submodule, a task-intensive association submodule, and a regional scheduling update submodule.

[0024] The continuity extraction submodule calls the task and vessel combination information in the task matching and scheduling list to obtain the continuous operation records of each vessel. According to the frequency of continuous repetition of task numbers in the operation records and the time order, it marks the operation segments with close task connections, and numbers and groups the combinations with interval jumps between operation segments, establishes the vessel operation continuity level sorting, and generates operation continuity level distribution results.

[0025] The task-dense association submodule extracts the spatial density data of the current corresponding task based on the ship number in the operation continuity level distribution result. According to the spatial overlap area of ​​the task number and the task cluster density classification rules, it forms a combination structure of high continuity level ships and high density task areas. At the same time, it marks the degree of separation between the task corresponding to low continuity level ships and high density areas in spatial location, and generates a ship task matching combination list.

[0026] The regional scheduling update submodule calls the scheduling mapping information in the ship task matching combination list, redirects high-coherence-level ships to the task set area path nodes, repositions low-coherence-level ships to the task periphery area nodes, and maps and updates the adjustment trajectory of all task numbers in spatial distribution, completing the latest correspondence between task numbers and ship numbers in all regions, and generating a task density scheduling distribution map.

[0027] As a further embodiment of the present invention, the abnormal voting takeover module includes an abnormal identification submodule, a response feedback recording submodule, and a task replacement generation submodule;

[0028] The anomaly identification submodule reads the task execution process of each ship in the task density scheduling distribution map, extracts the propulsion anomaly signal, energy imbalance behavior and communication loss record in the task stage in chronological order, sorts them according to the order in which the records appear, determines whether there are two or more consecutive anomaly tag combinations, marks the corresponding task number and ship number, and generates an anomaly combination task label table.

[0029] The response feedback recording submodule collects the scheduling tags of other ships in the same scheduling area based on the task numbers of the ships marked in the abnormal combination task label table, broadcasts the task takeover request to these ships and records the feedback content, extracts the acceptance field and reception time from the feedback response, identifies the ship numbers that have confirmed the takeover intention and sorts them in ascending order of response time, and generates the task takeover response sorting result.

[0030] The task substitution generation submodule calls the first responding vessel number and the corresponding task number in the task takeover response sorting result, updates the executing vessel field of the task in the task density scheduling distribution diagram, records the adjustment nodes of the target task and the takeover vessel in the scheduling configuration, summarizes all completed substitution task information, and generates an abnormal task substitution assignment record.

[0031] As a further aspect of the present invention, the extraction of propulsion anomaly signals during the mission phase specifically involves: comparing the commanded propulsion power of the vessel with the actual propulsion power fed back by the sensor in real time; and generating the propulsion anomaly signal when the deviation between the actual propulsion power and the commanded propulsion power is greater than a first power threshold for a preset duration.

[0032] The energy imbalance behavior during the extraction task phase is specifically as follows: obtain the current remaining power of the ship and calculate the power consumption rate. When the power consumption rate is continuously higher than the expected power consumption rate generated based on task planning and environmental data, and the difference between the two exceeds a preset energy consumption rate threshold, the energy imbalance behavior is generated.

[0033] The process of extracting communication loss records during the task phase specifically involves: determining whether the network communication heartbeat packets between the ship and the cloud platform are received within a preset time period; and generating the communication loss record if the number of consecutively lost network communication heartbeat packets exceeds a preset packet loss threshold.

[0034] As a further aspect of the present invention, the system further includes:

[0035] The fleet scheduling and grouping module reads the abnormal task replacement assignment records, screens the overlapping trajectories of tasks and the departure synchronization records, identifies and groups cooperative vessels, rearranges their task paths and sequences, integrates them into the cloud for unified scheduling, and outputs the unmanned fleet scheduling and grouping structure diagram.

[0036] The unmanned vessel fleet scheduling and grouping structure diagram includes collaborative operation group numbers, path overlap measurement indicators, and joint scheduling time axis.

[0037] As a further aspect of the present invention, the fleet scheduling and grouping module includes a trajectory integration submodule, a collaborative screening module, and a scheduling reconstruction submodule;

[0038] The trajectory integration submodule reads the access order and corresponding operation trajectory of each task execution vessel in the abnormal task replacement assignment record, extracts the frequency of occurrence of the vessel task number, the overlapping position points and time segments of the task area in the trajectory path, sorts out the number of times each vessel task accesses and the number of path intersection nodes, and summarizes the vessel numbers with the same departure time in the same time period to generate a task trajectory intersection feature list.

[0039] The collaborative screening module determines whether a ship has both multiple overlapping records and departure time overlaps based on the ship number and path overlap count recorded in the task trajectory intersection feature list. Ship numbers that meet this condition are grouped and classified, marked as grouping objects, and a corresponding task number set structure is established for each group to generate a ship task collaborative grouping structure.

[0040] The scheduling reconfiguration submodule calls the task set corresponding to each group in the ship task collaborative grouping structure, reorders the task numbers in each group according to the original task order, and uniformly specifies the corresponding execution ship number as the scheduling entry in the sorting structure, updates the corresponding task node order in the platform scheduling arrangement structure, and maps the updated content to the regional task distribution layer to generate an unmanned fleet scheduling grouping structure diagram.

[0041] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0042] In this invention, multi-dimensional status labels are generated by extracting the vibration frequency of the thruster, the temperature change of the energy compartment, and the communication interruption interval. A scheduling availability quantification mechanism is established. Combined with the characteristics of sea waves, obstacle layers, and positioning accuracy requirements, task interference assessment, path passage identification, and fault tolerance boundary delineation are achieved. The scheduling distribution structure is adjusted based on the continuity of ship operations and the spatial density of tasks. Task substitution and assignment response are realized through an abnormal voting mechanism. Fleet reorganization scheduling is completed based on path overlap and departure coordination time. Thus, a closed-loop scheduling control system covering the entire process of identification, matching, adjustment, substitution, and grouping is constructed. Attached Figure Description

[0043] Figure 1 This is a flowchart illustrating the overall system flow of the present invention.

[0044] Figure 2This is a flowchart of the system modules of the present invention. Detailed Implementation

[0045] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0046] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0047] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0048] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0049] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0050] Please see Figure 1 This invention provides a technical solution: a cloud platform intelligent scheduling system for unmanned vessel fleet management, the system comprising:

[0051] The vessel status identification module acquires the vibration records of the propellers of each vessel in the unmanned fleet, the temperature change process of the energy compartment, and the communication logs of the communication antenna in the cloud platform. It identifies the frequency of propulsion structure fluctuations, the duration of energy overheating, and the interval of signal interruption, and classifies and labels the three dimensions of status in turn. Based on the labeling results, it identifies the current scheduling availability of the vessel in the platform and generates a vessel status scheduling label.

[0052] The risk task matching module calls the vessel status scheduling tag, processes the wind and wave records, obstacle distribution layers and positioning control accuracy requirements of the sea area where the task is located, matches the wind and wave records with the propulsion tags one by one to determine the degree of operational interference, and uses the obstacle layer and communication stability tag to jointly identify the path obstruction area. Combined with the positioning accuracy requirements, it determines the fault tolerance boundary of task allocation and outputs the task matching scheduling list.

[0053] The energy efficiency density switching module calls the task and ship combination in the task matching and scheduling list, extracts the continuous operation records of ships and the spatial density data of corresponding tasks in sequence, compares the ship operation continuity and regional task distribution characteristics, adjusts the scheduling direction of high continuity ships in the task concentration area, and repositions low continuity ships to the outward distribution area. After completing the scheduling configuration, it outputs the task density scheduling distribution map.

[0054] The abnormal voting takeover module reads the task execution process of each ship in the task density scheduling distribution map, marks the propulsion abnormal signal, energy imbalance behavior and communication loss record, activates the takeover response mechanism according to the abnormal triggering order, broadcasts the signal to the nearby scheduled ships and records the feedback response whether the task is accepted, the platform counts the valid takeover responses and filters the first confirmed object, and generates an abnormal task replacement assignment record after completing the scheduling replacement.

[0055] The fleet scheduling and grouping module reads the access order and cross-operation trajectory of each task-executing vessel in the abnormal task replacement assignment record, sorts out the task access frequency, path overlap number and collaborative departure time period, screens out vessels with high overlap characteristics as grouping objects, reconstructs the task assignment of each group according to the task order, and incorporates them into the platform scheduling and arrangement process, outputting the unmanned fleet scheduling and grouping structure diagram.

[0056] The vessel status scheduling label includes availability level identifier, status classification code and scheduling adaptation level; the task matching scheduling list includes environmental adaptability coefficient, path passability index and positioning error tolerance; the task density scheduling distribution map includes high-density operation blocks, low-density scheduling zones and coherence distribution parameters; the abnormal task replacement assignment record includes replacement task number, takeover response identifier and feedback confirmation time stamp; and the unmanned vessel fleet scheduling grouping structure diagram includes collaborative operation group number, path overlap metric index and joint scheduling time axis.

[0057] Please see Figure 2 The ship status identification module includes a vibration identification submodule, a temperature fluctuation extraction submodule, and a scheduling tag generation submodule;

[0058] The vibration recognition submodule acquires vibration records of the unmanned vessel propulsion system from the cloud platform, divides the continuous recording segments in chronological order, identifies the arrangement pattern of peaks and troughs in the segments, filters the segments based on the number of abrupt changes in waveform and frequency variation, marks the recording segments with vibration characteristic changes, and generates vibration anomaly segment annotation results.

[0059] The propeller vibration record of the unmanned surface vessel USV-007 from 14:20:00 to 14:30:00 was obtained from the cloud platform. This record is a time series data set with a sampling frequency of 100Hz, specifically [20.1, 20.3, ..., 24.5, 28.9, 35.1, ..., 22.5]Hz. This 10-minute continuous record was divided into 60 continuous recording segments with 10-second intervals. For each segment, the maximum and minimum vibration frequencies were first extracted, and the average amplitude within the segment was calculated. For example, for the timestamp 14:25:10 to 14:25:20:00... Recording segment 32 of segment 0 has vibration frequency data of [24.5, 24.6, ..., 35.1, 34.8, ...] Hz and amplitude data of [0.5, 0.5, ..., 1.8, 1.7, ...] mm. The arrangement pattern of peaks and troughs in the segment is identified by calculating the Fast Fourier Transform (FFT) of the vibration signal within the segment to obtain the dominant frequency, and counting the number of peaks with amplitudes exceeding 1.5 times the mean. The segment is then filtered based on the number of abrupt changes in waveform and the frequency variation pattern. The filtering criterion for the number of abrupt changes is: within a recording segment, the instantaneous amplitude value exceeds 3 times the average amplitude of that segment. The frequency variation was determined by statistically analyzing vibration data from 10 unmanned surface vessels of the same model operating normally for 1000 hours in Beaufort scale 2 sea state. The sum of three times the standard deviation of the amplitude and the mean was taken as 1.6 mm. The selection criterion for frequency variation was the slope of the dominant frequency change within a segment, i.e., the change in frequency per unit time. This slope threshold was set at 5 Hz / s. This threshold was derived from the analysis of historical normal operation data. The average slope of frequency change during normal propeller start-up, shutdown, or speed adjustment was 3 Hz / s, and 1.65 times this was taken as the threshold. In the 32nd recording segment, the amplitude reached 1.8 mm at 14:25:15, exceeding the 1.6 mm threshold, and was counted as one mutation. The dominant frequency in this segment increased linearly from 25 Hz to 35 Hz over a period of 2 seconds, with a slope of (35-25) / 2 = 5 Hz / s, reaching the frequency change threshold. Since this segment simultaneously met the screening criteria of having more than or equal to one mutation and a frequency change slope of 5 Hz / s, it was marked as a recording segment with vibration characteristic changes. The timestamp, ship number, and marker of this segment were combined to generate the vibration anomaly segment labeling results.

[0060] The temperature fluctuation extraction submodule, based on the time segment corresponding to the vibration anomaly segment annotation results, calls the same segment record during the temperature change process of the energy cabin, sorts out the upward trend and change rhythm of the temperature data, filters out the segments that show rapid temperature rise or fluctuation at the same time as the vibration anomaly segment, summarizes them to form an abnormal behavior chain of temperature response, and generates a set of temperature fluctuation related segments.

[0061] Based on the time segment corresponding to the vibration anomaly segment annotation results, namely the period from 14:25:10 to 14:25:20 for USV-007, the same segment recording of the ship's energy compartment temperature change process stored in the cloud platform was retrieved. The obtained temperature sensor data is a time series, for example [45.1, 45.1, 45.2, ..., 46.8, 47.5]℃, with a sampling interval of 1 second. The upward trend and change rhythm of the temperature data were analyzed. Specifically, the difference calculation was performed on the 10-second temperature data sequence to obtain a temperature change rate sequence consisting of 9 data points, i.e. The result is [0, 0.1, ..., 0.7]℃ / s. Segments exhibiting rapid temperature rise or fluctuation simultaneously with the abnormal vibration segment are selected. The selection criterion is that the temperature change rate exceeds a preset rapid rise threshold for more than 2 seconds. This threshold is based on the following: Testing the propulsion system of this type of unmanned vessel under full power output conditions, the normal maximum temperature rise rate of its energy compartment was recorded as 0.2℃ / s. Considering sensor errors and environmental influences, the threshold is set to 1.5 times this value, i.e., 0.3℃ / s. In the above example, 14: The temperature data from 25:18 to 14:25:20 were 46.1℃, 46.8℃, and 47.5℃, with corresponding change rates of 0.7℃ / s and 0.7℃ / s, respectively. Both exceeded the threshold of 0.3℃ / s and lasted for 2 seconds, meeting the screening criteria. Therefore, this time period [14:25:18, 14:25:20] was marked as a temperature anomaly and associated with the vibration anomaly segment [14:25:10, 14:25:20] to form an abnormal behavior chain of temperature response, generating a set of temperature fluctuation associated segments.

[0062] The scheduling tag generation submodule calls the temperature fluctuation associated fragment set and vibration abnormal fragment annotation results, compares the time marker of the interruption point recorded in the communication antenna communication log, screens the time intersection in the three dimensions of records, identifies the ship number with multiple abnormal behaviors, and classifies the abnormal combination according to the preset scheduling availability standard to generate ship status scheduling tags.

[0063] The system retrieved the temperature fluctuation correlation fragment set and vibration anomaly fragment annotation results from USV-007 between 14:25:10 and 14:25:20, and compared them with the communication antenna communication log for the same time period. This log recorded the satellite link status with the cloud platform, showing a communication interruption at 14:25:17, recorded as {timestamp:14:25:17, event:'loss_of_signal'}. The system then screened for time intersections across the three dimensions of the records. Specifically, it compared the vibration anomaly time period [14:25:10, 14:25:20], the temperature anomaly time period [14:25:18, 14:25:20], and the communication interruption time of 14:25:17, finding that all three anomalies occurred between 14:25:10 and 14:25:20. Within the 10-second window of 25:20, the vessel with multiple anomalies was identified as USV-007. The anomaly combination was then categorized according to a preset scheduling availability standard, which classifies anomaly combinations into three levels: Level 1 is a single-dimensional anomaly, allowing the vessel to continue its mission; Level 2 is any combination of two-dimensional anomalies, restricting the vessel from performing high-risk missions; Level 3 is concurrent three-dimensional anomalies, rendering the vessel unavailable. Since USV-007 exhibited anomalies in vibration, temperature, and communication simultaneously, this combination was classified as Level 3, generating a vessel status scheduling label with the following content: {Vessel Number: 'USV-007', Timestamp: 'T14:25:20Z', Status Level: 3, Anomaly Description: 'Anomalous thruster vibration, rapid rise in power bay temperature, communication interruption'}.

[0064] The risk task matching module includes an operation interference identification submodule, a path obstruction judgment submodule, and a schedule list generation submodule;

[0065] The operation interference identification submodule obtains wind and wave records of the sea area where the mission is located. Based on the time period and geographical location marked in each record, it calls the tag content related to the propulsion status in the ship status scheduling tag of the corresponding ship, establishes a one-to-one correspondence between wind and wave records and propulsion status, matches them according to the combination of wind and wave intensity level and adaptive description in propulsion status tag, marks the interference-affected segments and organizes their ship numbers, and generates propulsion adaptive interference matching results.

[0066] Obtain the wind and wave records at 15:00 in the sea area where mission T-082 is located, i.e., within 5 nautical miles of coordinates (N30.5, E122.1). These records include a real-time significant wave height of 2.8 meters and a real-time average wind speed of 18 m / s. Based on the time period and geographical location indicated in the records, retrieve the vessel status scheduling tag for the corresponding vessel, USV-008. This tag contains propulsion status-related tags: {propulsion system wave resistance threshold: 2.5 meters, propulsion system wind resistance threshold: 20 m / s}. Establish a one-to-one correspondence between the wind and wave records and the propulsion status. Match the wind and wave intensity level with the adaptive description in the propulsion status tag. Specifically, match the real-time significant wave height data of 2.8 meters and the real-time average wind speed data of 18 m / s in the wind and wave records with the preset propulsion status in the propulsion status tag. The wave resistance threshold of 2.5 meters for the propulsion system and the wind resistance threshold of 20 m / s for the propulsion system were numerically compared. The wave resistance threshold of the propulsion system was set based on the structural strength analysis and hydrodynamic performance simulation of the hull. By performing finite element analysis on the hull stress under different wave height conditions, it was determined that at a significant wave height of 2.5 meters, the stress of the key structural components of the hull reaches 80% of the allowable design stress. This value was set as the threshold. When the real-time significant wave height data of 2.8 meters exceeded the wave resistance threshold of 2.5 meters for the propulsion system, the matching was completed. Although the real-time average wind speed data of 18 m / s did not exceed the wind resistance threshold of 20 m / s for the propulsion system, the matching was still valid because the wave height condition was met. The time period of 15:00 and the ship number USV-008 were marked as the interference segment, and the ship numbers to which they belonged were organized to generate the propulsion adaptation interference matching results.

[0067] The path obstruction judgment submodule calls the ship number involved in the propulsion adaptation interference matching result, obtains the communication status label of the position segment on the mission target path, and compares it with the path node data in the obstacle distribution layer in the mission area to identify the intersection range of unstable communication nodes and obstacle points. It then organizes and classifies the set of nodes with dual restrictions in the path to generate a path restriction node distribution list.

[0068] The system retrieves the vessel number USV-008 from the propulsion adaptation interference matching results and obtains its location segment on the mission target path P-082, specifically the communication status label from node N30.52-E122.11 to N30.54-E122.15. This label records the signal packet loss rate and communication delay of each path node along the route. For example, the packet loss rate of path node PN-01 is 3%, and the delay is 450ms. Referring to the path node data in the obstacle distribution layer within the mission area, which marks obstacles such as reefs and shipwrecks, one obstacle point, OBS-03, has coordinates N30.53-E122.13. Unstable communication nodes are identified by: path nodes whose signal packet loss rate or communication delay recorded in the communication status label exceeds a preset packet loss rate threshold or a preset delay threshold, and these are identified as unstable communication nodes. The packet loss rate threshold is based on historical data. Statistically, taking the upper limit of the 95% confidence interval under normal communication conditions, we set it to 2%. The delay threshold, based on the real-time control requirements of the task, was set to 500ms. The experimental verification process involved gradually increasing the packet loss rate and delay in a simulated environment. When the packet loss rate exceeded 2% or the delay exceeded 500ms, the success rate of remote manual control dropped below 90%. Therefore, this threshold was determined. The packet loss rate of path node PN-01 was 3%, which is higher than the 2% threshold. Therefore, PN-01 was identified as a node with unstable communication. Next, we identified the intersection range between the unstable communication node and the obstacle point. Specifically, taking each obstacle point OBS-03 in the obstacle distribution layer as the geometric center, a dynamic safety buffer zone was calculated based on the hull length of the vessel USV-008 performing the task (10 meters), maximum speed (8 m / s), and minimum turning radius (20 meters). The radius of this zone was calculated as the minimum turning radius plus one times the ship length. The nodes with unstable communication states that fall within a circular area with a radius of 30 meters centered on OBS-03 are identified as nodes within the intersection range. If the coordinates of PN-01 are located within this area, they are classified into the set of nodes subject to dual restrictions. All such sets of nodes are sorted and categorized to generate a list of path-restricted node distributions.

[0069] The scheduling list generation submodule obtains the positioning control capability level content in the scheduling tag of each vessel number corresponding to each node in the path restriction node distribution list, and filters out the vessel numbers and corresponding task names that can maintain the accuracy tolerance range within the restricted area according to the positioning control accuracy requirements set by the task. The task numbers and executable vessels are combined and mapped one by one to generate a task matching scheduling list.

[0070] Based on the vessel number USV-008 corresponding to each node in the path restriction node distribution list, the positioning control capability level content in its scheduling tag is obtained, which is "Level A". Then, the positioning control accuracy requirements set in task T-082 are compared with the vessel number and corresponding task name that can maintain the position deviation within the work point of less than 0.5 meters. This screening is completed by comparing the positioning control capability level of the vessel with the accuracy requirements of the task. The preset correspondence between the vessel positioning control capability level and accuracy is shown in Table 1.

[0071] Table 1: Explanation of Ship Positioning and Control Capability Levels

[0072] grade Positioning accuracy (meters) Applicable task types Grade A ≤0.5 High-precision surveying and mapping, facility maintenance Level B 0.5 < precision ≤ 2.0 Water sampling and routine inspection Grade C ＞2.0 Extensive search and vigilance

[0073] As shown in Table 1, the table details the specific positioning accuracy range and applicable task types corresponding to different positioning control capability levels. USV-008 has a positioning control capability of level A, with a corresponding positioning accuracy of less than or equal to 0.5 meters. This value meets the accuracy requirement of less than 0.5 meters for task T-082. Therefore, vessel USV-008 is selected. The task number T-082 is mapped one-to-one with the executable vessel USV-008. Meanwhile, another vessel, USV-009, has a positioning control capability of level B (accuracy 0.5-2.0 meters), which does not meet the task requirements and is therefore not selected. Finally, all the selected combinations are summarized to generate a task matching and scheduling list.

[0074] The energy efficiency density switching module includes a continuity extraction submodule, a task-intensive association submodule, and a regional scheduling update submodule;

[0075] The continuity extraction submodule calls the task and vessel combination information in the task matching and scheduling list to obtain the continuous operation records of each vessel. According to the frequency of continuous repetition of task numbers in the operation records and the time order, it marks the operation segments with close task connections, and numbers and groups the combinations with interval jumps between operation segments, establishes the vessel operation continuity level sorting, and generates operation continuity level distribution results.

[0076] The task matching and scheduling list is accessed to retrieve task and vessel combination information, such as {T-082:USV-008}, {T-083:USV-008}, {T-085:USV-008}, {T-086:USV-010}. The continuous operation records for each vessel are obtained. Taking USV-008 as an example, its operation records are: T-082 (08:00-10:00), T-083 (10:00-12:00), 1-hour interval, T-085 (13:00-15:00). Based on the frequency of consecutive repetitions and time sequence of task numbers in the operation records, closely connected operation segments are marked, where "task sequence" is used. "Closely connected" is defined as an interval of less than or equal to 30 minutes between two consecutive tasks. For USV-008, the interval between tasks T-082 and T-083 is 0 minutes, which is considered closely connected. However, the interval between T-083 and T-085 is 1 hour, indicating a jump in connection. Combinations with jumps in connection between work segments are numbered and grouped, with {T-082, T-083} as one group and {T-085} as another. A ranking of vessel operation continuity is established, based on the following rule: calculating the proportion of the total duration of all closely connected tasks within 24 hours to the total operation time. The continuity level is divided into three grades: High (proportion > 80%), Medium (50% < proportion). 80%), low (proportion) (50%), assuming the total operation time of USV-008 on that day was 6 hours, of which 4 hours were closely connected tasks, calculate its continuity ratio as follows: The first one is classified as medium level, while USV-010 performed 5 consecutive missions that day, totaling 5 hours, without any intervals, with a continuity rate of 100%, which is classified as high level. The levels of all ships are summarized to generate the operation continuity level distribution results.

[0077] The task-dense association submodule extracts the spatial density data of the current corresponding task based on the ship number in the operation continuity level distribution results. According to the spatial overlap area of ​​the task number and the task cluster density classification rules, it forms a combination structure of high continuity level ships and high density task areas. At the same time, it marks the degree of separation between the task corresponding to low continuity level ships and high density areas in spatial location, and generates a ship task matching combination list.

[0078] Based on the vessel numbers in the operational continuity level distribution results, such as USV-010 (high continuity level) and USV-008 (medium continuity level), the spatial density data of their current corresponding tasks is extracted. This data is calculated by analyzing the geographical distribution of tasks. According to the spatial overlap area of ​​task numbers and the task cluster density classification rules, high continuity level vessels are combined with high-density task areas. The task cluster density classification rules are as follows: within a circular area with a radius of 10 kilometers, if the number of tasks is greater than 10, it is defined as a high-density area; 5 to 10 tasks are medium-density; and less than 5 tasks are low-density. Assuming that USV-010's current mission T-086 is located in a high-density area containing 12 missions, USV-010 is grouped with this high-density area. At the same time, the spatial separation degree of the missions corresponding to low-coherence-level vessels from the high-density area is marked. USV-008 has a medium coherence level, and its mission T-085 is located in a low-density area. The distance between the center of this area and the center of the nearest high-density area is 25 kilometers. This distance is recorded as the separation degree. All vessels are matched with the density and separation degree of their mission areas to generate a list of vessel mission matching combinations.

[0079] The regional scheduling update submodule calls the scheduling mapping information in the ship task matching combination list, redirects high-coherence-level ships to the task set area path nodes, repositions low-coherence-level ships to the task periphery area nodes, and maps and updates the adjustment trajectory of all task numbers in spatial distribution, completes the latest correspondence between task numbers and ship numbers in all regions, and generates a task density scheduling distribution map.

[0080] The scheduling mapping information from the ship task matching combination list is retrieved, such as {Ship: USV-010, Consistency Level: High, Task Area: High Density}, {Ship: USV-008, Consistency Level: Medium, Task Area: Low Density, Separation: 25 km}. The high-consistency-level ship USV-010 is reassigned to a path node in the task concentration area, ensuring it continues to execute sequential tasks within the high-density area. The low-consistency-level ship USV-008 is repositioned to a node in the task periphery area, and assigned a node located at its current position and with high density. The new mission T-090 in the medium-density region between the density zones changes its original mission plan. The adjustment trajectory of all mission numbers in spatial distribution is mapped and updated. Specifically, on the global mission map, the execution vessel field of mission T-085 is changed from USV-008 to pending assignment, and the execution vessel field of mission T-090 is updated to USV-008. At the same time, the scheduled route of USV-008 is updated to complete the latest correspondence between mission numbers and vessel numbers in all regions and generate a mission density scheduling distribution map.

[0081] The exception voting takeover module includes an exception identification submodule, a response feedback recording submodule, and a task replacement generation submodule;

[0082] The anomaly identification submodule reads the task execution process of each ship in the task density scheduling distribution map, extracts the propulsion anomaly signals, energy imbalance behavior and communication loss records in the task phase in chronological order, sorts them according to the order in which the records appear, determines whether there are two or more consecutive anomaly tag combinations, marks the corresponding task number and ship number, and generates an anomaly combination task label table.

[0083] The updated task density scheduling distribution map is read to record the task execution process of each ship. For example, if ship USV-007 is reassigned to task T-091, propulsion anomaly signals, energy imbalance behaviors, and communication loss records within the task phase are extracted in chronological order. First, propulsion anomaly signals within the task phase are extracted. Specifically, the commanded propulsion power of USV-007 (e.g., 5kW) is compared in real time with the actual propulsion power fed back by the sensors (e.g., 3.5kW). When the deviation between the actual propulsion power and the commanded propulsion power is... If the power output kW continuously exceeds a first power threshold for a preset duration, such as 10 seconds, a propulsion anomaly signal is generated. The first power threshold is set based on the thruster efficiency curve and is set to 20% of the rated power, i.e., 1kW. In this example, 1.5kW > 1kW and this exceeds 1kW for 10 seconds, hence the propulsion anomaly signal is generated. Next, the energy imbalance behavior during the mission phase is extracted, specifically: the current remaining power of USV-007 is obtained as 60%, and the power consumption rate over the past 5 minutes is calculated as 1.5% / min. When the power consumption rate continuously exceeds the expected power consumption rate of 1.0% / min generated based on mission planning and environmental data, and the difference between the two is greater than 10%, the propulsion anomaly signal is generated. When the energy consumption rate exceeds the preset energy consumption rate threshold of 0.4% / min, the energy imbalance behavior is generated. Finally, the communication disconnection record during the task phase is extracted. Specifically, it is determined whether the network communication heartbeat packet between USV-007 and the cloud platform is received within a preset time period of 2 seconds. If the number of consecutive loss of the network communication heartbeat packet exceeds the preset packet loss threshold of 5 times (i.e., 10 seconds of disconnection), the communication disconnection record is generated. The judgment criteria for each abnormal event are shown in Table 2.

[0084] Table 2: Thresholds for Judging Abnormal Events

[0085] Exception types Monitoring parameters Judgment conditions Duration requirement Progress Anomaly Deviation between commanded power and actual power >1.0 kW ≥10≥10 s Energy Imbalance Deviation between actual consumption rate and expected rate >0.4% / min continuous Communication lost Number of consecutive heartbeat packets lost >5 times not applicable

[0086] As shown in Table 2, this table summarizes the specific thresholds and conditions used to determine propulsion anomalies, energy imbalances, and communication failures. The records are sorted according to their order of appearance to determine whether there are two or more consecutive anomaly tag combinations. If a propulsion anomaly is generated at 16:10:05 and an energy imbalance is generated at 16:10:15, then an anomaly combination is formed, and the corresponding task number T-091 and vessel number USV-007 are marked, generating an anomaly combination task labeling table.

[0087] The response feedback recording submodule collects the scheduling tags of other ships in the same scheduling area based on the task numbers of the ships marked in the abnormal combination task label table, broadcasts the task takeover request to these ships and records the feedback content, extracts the acceptance field and reception time from the feedback response, identifies the ship numbers that have confirmed the takeover intention and sorts them in ascending order of response time, and generates the task takeover response sorting result.

[0088] Based on the mission number T-091 of vessel USV-007 already marked in the abnormal combination mission labeling table, the scheduling tags of other vessels USV-010, USV-011, and USV-012 within the same scheduling area (a circular area with a radius of 20 nautical miles centered on mission point T-091) are collected. A mission takeover request is broadcast to these vessels. This request includes the mission number, location, type, and estimated duration, and the feedback content is recorded. Specifically, the feedback content is a data structure containing {vessel number, whether to accept, response timestamp, current mission priority, remaining range}. The "accept" field and the reception time are extracted from the feedback response. For example, if USV-010 responds at 16:11:05, the response is {USV-010, true, 16:11:05}. At 16:11:10, USV-011 responded with {USV-011, false, 16:11:10, 2, 6h}, and at 16:11:08, USV-012 responded with {USV-012, true, 16:11:08, 3, 9h}. The vessel numbers USV-010 and USV-012, which have confirmed their willingness to take over, are identified and sorted in ascending order of response time. The response time of USV-010 (16:11:05) is compared with that of USV-012 (16:11:08). Since 16:11:05 is earlier than 16:11:08, USV-010 is ranked first. The task takeover response sorting result is generated, with the content being [USV-010, USV-012].

[0089] The task replacement generation submodule calls the first responding ship number and corresponding task number in the task takeover response sorting result, updates the executing ship field of the task in the task density scheduling distribution diagram, records the adjustment nodes of the target task and the takeover ship in the scheduling configuration, summarizes all completed replacement task information, and generates an abnormal task replacement assignment record.

[0090] Retrieve the first responding vessel number USV-010 and its corresponding task number T-091 from the task takeover response sorting results. Update the executing vessel field of task T-091 in the task density scheduling distribution diagram, changing its value from 'USV-007' to 'USV-010'. Record the adjustment node in the scheduling configuration between the target task T-091 and the receiving vessel USV-010 that took over the task. This node information includes the original vessel, the receiving vessel, the task number, the handover timestamp 16:12:00, and the handover geographical coordinates (N30). 8,E122.5), and send instructions to USV-007 to return or proceed to the designated maintenance point, while sending a new mission route plan to USV-010, and summarizing all completed replacement mission information, such as {Original vessel: USV-007, Replacement vessel: USV-010, Mission: T-091, Time: 16:12:00}, {Original vessel: USV-002, Replacement vessel: USV-005, Mission: T-056, Time: 11:45:00}, and generate an abnormal mission replacement assignment record.

[0091] The fleet scheduling and grouping module includes a trajectory integration submodule, a collaborative screening module, and a scheduling reconstruction submodule;

[0092] The trajectory integration submodule reads the access order and corresponding operation trajectory of each task execution vessel in the abnormal task replacement assignment record, extracts the frequency of the vessel task number, the overlapping position points and time segments of the task area in the trajectory path, sorts out the number of times each vessel task accesses and the number of path intersection nodes, and summarizes the vessel numbers with the same departure time in the same time period to generate a task trajectory intersection feature list.

[0093] Read the access order and corresponding operation trajectory of each task-executing vessel in the abnormal task replacement assignment record. For example, after completing its own task T-088, vessel USV-010 accessed and executed task T-091, which originally belonged to USV-007, at 16:12:00. Extract the frequency of the vessel's task number, the overlapping locations of task areas in the trajectory path, and the time segment. It is found that USV-010 accessed the tasks of other vessels twice within 24 hours. Its trajectory for executing T-088 and its trajectory for executing T-091 overlapped by 2 nautical miles in the area (N30.7, E122.4). Organize the number of times each vessel accessed the task and the number of path intersection nodes, and summarize the vessel numbers whose departure time was in the same time period to generate a task trajectory intersection feature list, as shown in Table 3.

[0094] Table 3: List of Task Trajectory Cross Features

[0095] Ship number Task access count Number of path intersection nodes Departure time period USV-010 2 5 08:00-20:00 USV-012 1 3 08:00-20:00 USV-005 0 1 09:00-21:00

[0096] See Table 3, which lists some of the cross-feature data of the mission trajectories of the vessels. This data will be used for subsequent collaborative screening.

[0097] The collaborative screening module determines whether a ship has both overlapping records of multiple tasks and the same departure time based on the ship number and the number of overlapping paths recorded in the task trajectory intersection feature list. Ship numbers that meet the conditions are grouped and classified, marked as grouping objects, and a corresponding task number set structure is established for each group to generate a ship task collaborative grouping structure.

[0098] Based on the number of overlaps between vessel numbers USV-010 and USV-012 and their paths recorded in the task trajectory intersection feature list, it is determined whether they simultaneously possess the characteristics of multiple task overlap records and departure time coincidence. The criteria for "multiple task overlap records" is a path intersection node count greater than or equal to 3, and the criteria for "departure time coincidence" is a departure time window overlap rate exceeding 50%. USV-010 has 5 path intersection nodes, greater than 3, and USV-012 has 3 path intersection nodes, equal to 3. Both have a departure time period of 0. From 8:00 to 20:00, the overlap rate is 100%, which meets the judgment condition. The vessel numbers USV-010 and USV-012 that meet the condition are grouped and classified, assigned group number G-01, and marked as group objects. A corresponding task number set structure is established for each group. All task numbers executed by USV-010 and USV-012 on the same day, such as {T-088, T-091, T-095, ...}, are integrated into the task set of group G-01 to generate a vessel task collaborative grouping structure.

[0099] The scheduling reconfiguration submodule calls the task set corresponding to each group in the ship task collaborative grouping structure, reorders the task numbers in each group according to the original task order, and uniformly specifies the corresponding execution ship number as the scheduling entry in the sorting structure, updates the corresponding task node order in the platform scheduling arrangement structure, and maps the updated content to the regional task distribution layer to generate an unmanned ship fleet scheduling grouping structure diagram.

[0100] The task set {T-088, T-095, T-091} corresponding to group G-01 in the ship task coordination grouping structure is invoked. The task numbers within the group are reordered according to the original task order, i.e., according to the planned start time, to obtain the sorted task sequence [T-088, T-091, T-095]. The corresponding execution ship number is uniformly designated as the scheduling entry point within the sorting structure. That is, the executor of the entire task sequence [T-088, T-091, T-095] is designated as group G-01, instead of a single ship. The order of the corresponding task nodes in the platform scheduling arrangement structure is updated. Specifically, the original independent task nodes T-088, T-091, and T-095 are merged into a group task node G-01, which contains ordered sub-tasks. The updated content is mapped to the regional task distribution layer, and the icons of these three tasks are aggregated and displayed on the map, marked as the operation range of group G-01, generating an unmanned fleet scheduling grouping structure diagram.

[0101] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A cloud platform intelligent scheduling system for unmanned vessel fleet management, characterized in that, The system includes: The ship status identification module acquires thruster vibration records, energy temperature changes and communication stability logs, extracts fluctuation frequency, heating duration and signal interruption status, completes status labeling in sequence, classifies ship scheduling level based on labeling results, and generates ship status scheduling labels. The risk task matching module calls the vessel status scheduling tag, processes wind and wave level, obstacle layer and positioning requirements, judges the degree of interference of wind and waves on propulsion stability, assesses the impact of communication interruption on operation in obstacle area, matches tolerance boundary, and outputs task matching scheduling list. The energy efficiency density switching module calls the ship task relationship in the task matching and scheduling list, adjusts high-coherence ships to dense areas and low-coherence ships out according to the continuity of operations and the distribution of task-intensive areas, reconstructs task affiliation, and outputs a task density scheduling distribution map. The abnormal voting takeover module reads the task execution records in the task density scheduling distribution map, identifies propulsion imbalance, energy anomalies, and communication interruptions, triggers a takeover request, broadcasts to neighboring ships, records feedback and identifies and confirms the response object, and generates an abnormal task replacement assignment record after the replacement is completed.

2. The cloud platform intelligent scheduling system for unmanned vessel fleet management according to claim 1, characterized in that: The vessel status scheduling label includes an availability level identifier, a status classification code, and a scheduling adaptation level. The task matching scheduling list includes an environmental adaptability coefficient, a path passability index, and a positioning error tolerance. The task density scheduling distribution map includes high-density operation blocks, low-density scheduling zones, and continuity distribution parameters. The abnormal task replacement assignment record includes a replacement task number, a takeover response identifier, and a feedback confirmation time stamp.

3. The cloud platform intelligent scheduling system for unmanned vessel fleet management according to claim 1, characterized in that, The ship status identification module includes a vibration identification submodule, a temperature fluctuation extraction submodule, and a scheduling tag generation submodule; The vibration recognition submodule acquires vibration records of the unmanned vessel propulsion system from the cloud platform, divides the continuous recording segments in chronological order, identifies the arrangement pattern of peaks and troughs in the segments, filters the segments based on the number of abrupt changes in waveform and frequency variation, marks the recording segments with vibration characteristic changes, and generates vibration anomaly segment annotation results. The temperature fluctuation extraction submodule, based on the time segment corresponding to the vibration anomaly segment annotation result, calls the same segment record during the temperature change process of the energy cabin, sorts out the upward trend and change rhythm of the temperature data, filters out the segments that show rapid temperature rise or fluctuation at the same time as the vibration anomaly segment, summarizes them to form an abnormal behavior chain of temperature response, and generates a set of temperature fluctuation related segments. The scheduling tag generation submodule calls the temperature fluctuation associated segment set and vibration anomaly segment annotation results, compares the time marker of the interruption point recorded in the communication antenna communication log, screens the time intersection in the three dimensions of records, identifies the ship number with multiple abnormal behaviors, and classifies the abnormal combination according to the preset scheduling availability standard to generate ship status scheduling tags.

4. The cloud platform intelligent scheduling system for unmanned vessel fleet management according to claim 1, characterized in that, The risk task matching module includes an operation interference identification submodule, a path obstruction judgment submodule, and a scheduling list generation submodule. The operation interference identification submodule obtains wind and wave records of the sea area where the mission is located. Based on the time period and geographical location marked in each record, it calls the tag content related to the propulsion status in the ship status scheduling tag of the corresponding ship, establishes a one-to-one correspondence between wind and wave records and propulsion status, matches them according to the combination of wind and wave intensity level and adaptive description in propulsion status tag, marks the interference-affected segments and organizes their ship numbers, and generates propulsion adaptive interference matching results. The path obstruction judgment submodule calls the ship number involved in the propulsion adaptation interference matching result, obtains the communication status label of the position segment on the mission target path, and compares it with the path node data in the obstacle distribution layer in the mission area to identify the intersection range of unstable communication nodes and obstacle points, sorts and classifies the set of nodes with dual restrictions in the path, and generates a path restriction node distribution list. The scheduling list generation submodule obtains the positioning control capability level content in the scheduling tag of each vessel number corresponding to each node in the path restriction node distribution list, and filters out the vessel numbers and corresponding task names that can operate within the accuracy tolerance range in the restricted area according to the positioning control accuracy requirements set by the task. The task numbers and executable vessels are mapped one by one to generate a task matching scheduling list.

5. The cloud platform intelligent scheduling system for unmanned vessel fleet management according to claim 4, characterized in that: The matching based on the combination of wind and wave intensity level and adaptive description in propulsion status label specifically involves: comparing the real-time effective wave height data and real-time average wind speed data in the wind and wave record with the preset propulsion system wave resistance level threshold and propulsion system wind resistance level threshold in the propulsion status label; when the real-time effective wave height data exceeds the propulsion system wave resistance level threshold or the real-time average wind speed data exceeds the propulsion system wind resistance level threshold, the matching is completed. The identification of nodes with unstable communication status specifically involves identifying path nodes whose signal packet loss rate is higher than a preset packet loss rate threshold or whose communication delay is higher than a preset delay threshold, as nodes with unstable communication status, as recorded in the communication status tags. The process of identifying the intersection range between unstable communication nodes and obstacle points specifically involves: using each obstacle point in the obstacle distribution layer as the geometric center, calculating and generating a dynamic safety buffer zone based on the ship length, maximum speed, and minimum turning radius data of the vessel performing the task, and identifying the unstable communication nodes that fall within the dynamic safety buffer zone as nodes within the intersection range.

6. The cloud platform intelligent scheduling system for unmanned vessel fleet management according to claim 1, characterized in that, The energy efficiency density switching module includes a continuity extraction submodule, a task-intensive association submodule, and a regional scheduling update submodule. The continuity extraction submodule calls the task and vessel combination information in the task matching and scheduling list to obtain the continuous operation records of each vessel. According to the frequency of continuous repetition of task numbers in the operation records and the time order, it marks the operation segments with close task connections, and numbers and groups the combinations with interval jumps between operation segments, establishes the vessel operation continuity level sorting, and generates operation continuity level distribution results. The task-dense association submodule extracts the spatial density data of the current corresponding task based on the ship number in the operation continuity level distribution result. According to the spatial overlap area of ​​the task number and the task cluster density classification rules, it forms a combination structure of high continuity level ships and high density task areas. At the same time, it marks the degree of separation between the task corresponding to low continuity level ships and high density areas in spatial location, and generates a ship task matching combination list. The regional scheduling update submodule calls the scheduling mapping information in the ship task matching combination list, redirects high-coherence-level ships to the task set area path nodes, repositions low-coherence-level ships to the task periphery area nodes, and maps and updates the adjustment trajectory of all task numbers in spatial distribution, completing the latest correspondence between task numbers and ship numbers in all regions, and generating a task density scheduling distribution map.

7. The cloud platform intelligent scheduling system for unmanned vessel fleet management according to claim 1, characterized in that, The abnormal voting takeover module includes an abnormal identification submodule, a response feedback recording submodule, and a task replacement generation submodule; The anomaly identification submodule reads the task execution process of each ship in the task density scheduling distribution map, extracts the propulsion anomaly signal, energy imbalance behavior and communication loss record in the task stage in chronological order, sorts them according to the order in which the records appear, determines whether there are two or more consecutive anomaly tag combinations, marks the corresponding task number and ship number, and generates an anomaly combination task label table. The response feedback recording submodule collects the scheduling tags of other ships in the same scheduling area based on the task numbers of the ships marked in the abnormal combination task label table, broadcasts the task takeover request to these ships and records the feedback content, extracts the acceptance field and reception time from the feedback response, identifies the ship numbers that have confirmed the takeover intention and sorts them in ascending order of response time, and generates the task takeover response sorting result. The task substitution generation submodule calls the first responding vessel number and the corresponding task number in the task takeover response sorting result, updates the executing vessel field of the task in the task density scheduling distribution diagram, records the adjustment nodes of the target task and the takeover vessel in the scheduling configuration, summarizes all completed substitution task information, and generates an abnormal task substitution assignment record.

8. The cloud platform intelligent scheduling system for unmanned vessel fleet management according to claim 7, characterized in that: The extraction of propulsion anomaly signals during the task phase specifically involves: comparing the commanded propulsion power of the vessel with the actual propulsion power fed back by the sensors in real time; and generating the propulsion anomaly signal when the deviation between the actual propulsion power and the commanded propulsion power is greater than a first power threshold for a preset duration. The energy imbalance behavior during the extraction task phase is specifically as follows: obtain the current remaining power of the ship and calculate the power consumption rate. When the power consumption rate is continuously higher than the expected power consumption rate generated based on task planning and environmental data, and the difference between the two exceeds a preset energy consumption rate threshold, the energy imbalance behavior is generated. The process of extracting communication loss records during the task phase specifically involves: determining whether the network communication heartbeat packets between the ship and the cloud platform are received within a preset time period; and generating the communication loss record if the number of consecutively lost network communication heartbeat packets exceeds a preset packet loss threshold.

9. The cloud platform intelligent scheduling system for unmanned vessel fleet management according to claim 1, characterized in that, The system also includes: The fleet scheduling and grouping module reads the abnormal task replacement assignment records, screens the overlapping trajectories of tasks and the departure synchronization records, identifies and groups cooperative vessels, rearranges their task paths and sequences, integrates them into the cloud for unified scheduling, and outputs the unmanned fleet scheduling and grouping structure diagram. The unmanned vessel fleet scheduling and grouping structure diagram includes collaborative operation group numbers, path overlap measurement indicators, and joint scheduling time axis.

10. The cloud platform intelligent scheduling system for unmanned vessel fleet management according to claim 9, characterized in that, The fleet scheduling and grouping module includes a trajectory integration submodule, a collaborative screening module, and a scheduling reconstruction submodule. The trajectory integration submodule reads the access order and corresponding operation trajectory of each task execution vessel in the abnormal task replacement assignment record, extracts the frequency of occurrence of the vessel task number, the overlapping position points and time segments of the task area in the trajectory path, sorts out the number of times each vessel task accesses and the number of path intersection nodes, and summarizes the vessel numbers with the same departure time in the same time period to generate a task trajectory intersection feature list. The collaborative screening module determines whether a ship has both multiple overlapping records and departure time overlaps based on the ship number and path overlap count recorded in the task trajectory intersection feature list. Ship numbers that meet this condition are grouped and classified, marked as grouping objects, and a corresponding task number set structure is established for each group to generate a ship task collaborative grouping structure. The scheduling reconfiguration submodule calls the task set corresponding to each group in the ship task collaborative grouping structure, reorders the task numbers in each group according to the original task order, and uniformly specifies the corresponding execution ship number as the scheduling entry in the sorting structure, updates the corresponding task node order in the platform scheduling arrangement structure, and maps the updated content to the regional task distribution layer to generate an unmanned fleet scheduling grouping structure diagram.

Images