A smart IoT control system for unmanned ships

Abstract

This invention relates to the field of Internet of Things (IoT) technology, specifically to an intelligent IoT control system for unmanned surface vessels (USVs). The system includes collecting navigation environment data to identify wind and wave interference and generate disturbance markers. Based on these markers, it analyzes communication channel interruptions and generates a priority list. Next, it reads priority channel data, filters urgent and unique tasks to generate an information index set, and then compares the task direction with the real-time heading. It filters and sorts deflection and change-of-direction tasks to generate execution groups. Finally, it analyzes the control behaviors in the execution groups, extracts key parameters, and encapsulates them into ship control commands. This invention, by accurately collecting navigation environment data to mark wind and wave interference, adjusts communication channels to ensure priority transmission of critical data, filters high-priority tasks to ensure their timeliness and sequence, and then combines this with the ship's heading to precisely arrange tasks and generate precise control commands, comprehensively improving the USV's responsiveness, control stability, and intelligence level.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and in particular to an intelligent IoT control system for unmanned vessels. Background Technology

[0002] The Internet of Things (IoT) technology field encompasses an application system based on sensor acquisition, communication transmission, data storage, and remote control. Its core functionality involves acquiring real-time information about the environment, devices, and objects through sensing terminals, transmitting this data to a processing center via wired or wireless communication networks, and then integrating and scheduling the data. The overall IoT technology field covers the collaborative operation of the sensing layer, network layer, and application layer. The sensing layer is responsible for the acquisition and identification of physical quantities; the network layer enables data transmission and exchange under different communication protocols; and the application layer provides control and management methods for specific scenarios. It has wide applications in transportation, energy, marine monitoring, and unmanned systems, and supports intelligent management needs across multiple scenarios through a combination of hardware, communication protocols, and system platforms.

[0003] One type of intelligent IoT control system for unmanned vessels refers to a control system based on IoT architecture designed for unmanned vessel operation scenarios. It addresses control matters related to navigation, operation, and remote interaction of unmanned vessels, covering environmental parameter acquisition, communication link establishment, data exchange, and remote command execution. Specifically, it uses sensors to collect water environment and vessel operation parameters, establishes a data channel between the land-based terminal and the unmanned vessel through wireless communication, and completes the uploading of status information and the issuance of remote control commands through data interaction. At the same time, it uses actuators to convert control commands into navigation or operation operations, thus forming a complete IoT control architecture.

[0004] In existing technologies, the ability to handle the impact of environmental changes on unmanned surface vessels (USVs) is relatively weak. Especially when wind speed, water flow direction, and wave height change drastically, the vessel's behavior cannot be adjusted in time, resulting in inconsistent vessel responses. At the communication level, there is a lack of effective monitoring and processing mechanisms for sudden signal interruptions, which can easily cause communication delays or data loss, affecting the timely transmission and processing of mission data. In the process of mission scheduling, the deviation between the vessel's orientation and the target behavior is not fully considered, which may lead to inaccurate control commands during mission execution, reducing operational efficiency. In complex environments, the control precision and intelligence level of mission scheduling of USVs are insufficient, affecting their overall operational capabilities and efficiency. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies and propose an intelligent Internet of Things control system for unmanned vessels.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent Internet of Things control system for unmanned vessels, the system comprising:

[0007] The environmental perception module collects wind speed, water flow and wave height data during the unmanned vessel's navigation, extracts the changing trends, and, combined with the difference between the vessel's orientation and direction of movement, identifies the periods of inconsistent movement, marks the time points of wind and wave interference and the direction of disturbance to the vessel, and generates navigation environment disturbance labels.

[0008] The channel adjustment module analyzes the connection change sequence of each communication channel during the period of sudden wind and wave changes based on the wind direction shift and wave height change trend in the navigation environment disturbance indicator, counts the number of interruptions of each channel, and sorts them from low to high according to the number of interruptions to generate a primary channel sorting list.

[0009] The data filtering module reads the priority channel number from the primary channel sorting list, retrieves the task data to be transmitted from the channel, filters urgent and non-delayed tasks, removes duplicate records, keeps the order unchanged and writes them into the number sequence, and generates an instant task information index set.

[0010] The task scheduling module calls the task number in the real-time task information index set, compares the task target with the ship's real-time direction of travel, filters out tasks with excessive deflection, marks the change-of-direction tasks related to the course turning node, summarizes and optimizes the order of the marked tasks, and generates a task execution order group.

[0011] As a further aspect of the present invention, the navigation environment disturbance identifiers include wind speed direction shift, water flow deflection, wave height change trend, wind and wave interference time point, hull disturbance direction, and time period of inconsistent motion characteristics. The primary channel sorting list includes communication channel number, signal connection change record, wind direction change period, wave height jump period, and interruption number sorting. The real-time task information index set includes high-priority task data, emergency level marker, generation time marker, task data number, and valid data sorting. The task execution order group includes target behavior direction, current travel direction shift, target position and navigation path turning node, shift task number, and change-of-direction section task marker.

[0012] As a further aspect of the present invention, the environment perception module includes:

[0013] The data acquisition submodule collects real-time data on wind speed, water flow direction, and wave height during the unmanned vessel's navigation, records the changes in wind speed, water flow direction, and wave height in each time period, and updates the data in real time according to the set time intervals during data acquisition, generating a dataset of wind speed, water flow direction, and wave height.

[0014] The feature extraction submodule extracts the changing trends of wind speed direction, water flow deflection, and wave height jumps based on the wind speed, water flow direction, and wave height dataset. By analyzing the fluctuation characteristics of the data, it identifies the changes in wind speed direction, the magnitude of water flow deflection, and the jumps in wave height, and generates wind and wave change trends.

[0015] The disturbance identification submodule identifies whether there are time periods that are out of sync with environmental changes during the movement process based on the difference between the trend of wind and waves and the orientation and direction of movement of the unmanned vessel. It analyzes the disturbance situation of the vessel, marks the time of wind and wave interference, and marks the specific location according to the direction of the disturbance, and generates a navigation environment disturbance identification.

[0016] As a further aspect of the present invention, the channel adjustment module includes:

[0017] The signal extraction submodule extracts the communication status record of the unmanned vessel in the current period based on the wind direction shift and wave height jump change trend marked in the navigation environment disturbance identifier, obtains the signal connection change sequence of each communication channel, records the signal status of each channel in a specified time period, and generates a communication status record.

[0018] The connection judgment submodule, based on the communication status record, determines whether a communication interruption occurred during the period of sudden wind direction change and wave height jump, analyzes the connection status of each channel in different time periods, marks the time period of signal interruption, records the interruption frequency of each communication channel, and generates an interruption frequency set.

[0019] The channel sorting submodule sorts all communication channels according to the interruption frequency set, placing channels with lower interruption frequencies at the top with higher priority, and vice versa, generating a primary channel sorting list.

[0020] As a further aspect of the present invention, the data screening module includes:

[0021] The task filtering submodule retrieves the task information to be transmitted in the channel based on the highest priority channel number in the main channel sorting list, analyzes each task item in the task, filters out the task items marked with the highest urgency level and whose generation time has not been delayed, extracts the task items that meet these conditions, and generates high-priority task items.

[0022] The data cleaning submodule, based on the high-priority task items, identifies and removes duplicate records to ensure the uniqueness of each task item, while maintaining the original order of valid tasks. After removing redundant content, it sorts out the task items that meet the conditions and generates valid task items.

[0023] The index generation submodule assigns a number to each task according to the extraction order based on the valid task items, and writes the task numbers into the number sequence in the extraction order to ensure the correctness of the task execution order, thereby generating an instant task information index set.

[0024] As a further aspect of the present invention, the task scheduling module includes:

[0025] The behavior comparison submodule reads the target behavior direction corresponding to each task number according to the task number in the real-time task information index set, and compares it with the real-time travel direction determined by the current unmanned vessel's trajectory record. It determines whether there is an offset between the target behavior direction and the real-time travel direction, and further determines whether the offset exceeds the allowable deflection range. If there is an offset and it exceeds the set threshold, it is marked as a task that needs to be adjusted, and an offset detection result is generated.

[0026] The offset filtering submodule, based on the offset detection results, filters out all task numbers whose offset between the target behavior direction and the actual travel direction exceeds the allowable range, classifies the tasks, identifies tasks with larger offsets, marks them, and generates offset task numbers.

[0027] The order adjustment submodule compares the target position pointed to by each task with the turning nodes in the current navigation path diagram based on the offset task number, identifies those tasks that need to change direction, and adjusts their execution order for tasks entering the turning section so that the tasks are reasonably sorted according to the turning nodes in the predetermined navigation path, generating a task execution order group.

[0028] As a further aspect of the present invention, the system further includes:

[0029] The instruction generation module extracts each task number and corresponding control behavior content from the task execution sequence group, determines whether the task involves thruster action and directional control operation, extracts the propulsion direction, rudder angle offset and control continuity description from the content, encapsulates the task with control behavior characteristics, and generates hull control instruction results.

[0030] The results of the ship control commands include propulsion direction, rudder angle offset, control duration, and mission control set.

[0031] As a further aspect of the present invention, the instruction generation module includes:

[0032] The task reading submodule reads each task number and its corresponding control behavior content according to the task execution order, extracts the description of thruster action or direction control operation in the task, obtains the detailed control requirements of each task, and generates task control information by traversing the task numbers and summarizing the relevant control information of the task.

[0033] The behavior extraction submodule, based on the task control information, determines whether the task involves control behaviors such as thruster direction or rudder angle deviation, extracts the thrust direction, rudder angle deviation and control duration from the task, classifies these control behaviors, filters out task items containing control behavior features, and generates control behavior content.

[0034] The instruction encapsulation submodule encapsulates tasks with control behavior characteristics according to the execution order based on the control behavior content, organizes the control operation set corresponding to each task, arranges the operation set according to the task order, and generates the hull control instruction result.

[0035] As a further aspect of the present invention, after obtaining the real-time task information index set, the priority and time window of the task items are classified and identified. Task items with similar priority characteristics, urgency requirements and task time constraints are grouped into the same category and assigned corresponding task priority tags. In the subsequent task scheduling and execution process, a differentiated scheduling strategy is executed based on the task priority tags.

[0036] As a further aspect of the present invention, after obtaining the task execution order group, the execution order of different tasks is re-evaluated and optimized. Tasks with similar target locations, task types, and variable requirements are grouped into similar groups, and a corresponding priority execution strategy is assigned to each task. The execution order is adjusted in a differentiated manner according to the variable requirements and execution time of the task.

[0037] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0038] In this invention, by accurately collecting data on wind speed, water flow direction, and wave height changes in the navigation environment, the timing of wind and wave interference and the direction of disturbance to the hull can be marked in a timely manner, optimizing the hull's response capability in complex environments. Communication channel adjustment is based on the identification of signal interruption periods according to environmental changes, making the monitoring of interruptions more accurate. By sorting the number of interruptions, the channel selection for data transmission is optimized to ensure the priority transmission of critical task data. During the task screening process, high-priority tasks are accurately identified to ensure the timely transmission of emergency tasks and to maintain the data order, thereby improving the stability and efficiency of the communication process. Task arrangement is adjusted based on the offset between the hull's direction of travel and the target behavior to ensure the accuracy of task execution. Through the precise generation and execution of control commands, the precise control of the hull's propulsion and rudder angle is achieved, improving the control stability and intelligence level of the unmanned vessel in changing environments. Attached Figure Description

[0039] Figure 1 This is a system flowchart of the present invention;

[0040] Figure 2 This is a flowchart illustrating the acquisition process of the environmental perception module of the present invention.

[0041] Figure 3 This is a flowchart illustrating the acquisition process of the channel adjustment module of the present invention.

[0042] Figure 4 This is a flowchart illustrating the data acquisition process of the data screening module of the present invention.

[0043] Figure 5 This is a flowchart illustrating the acquisition process of the task scheduling module of the present invention.

[0044] Figure 6 This is a flowchart illustrating the process of obtaining instructions from the instruction generation module of this 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: an intelligent Internet of Things control system for unmanned vessels, the system comprising:

[0051] The environmental perception module collects instantaneous wind speed, water flow direction and wave height change data during the unmanned vessel's navigation process, extracts the changing trends of wind speed direction, water flow deflection and wave height jump, matches them according to the synchronous difference between the unmanned vessel's hull orientation and motion direction, identifies time periods with inconsistent motion characteristics, marks the time points of wind and wave interference and the corresponding positions of the hull's disturbed direction, and generates navigation environment disturbance labels.

[0052] The channel adjustment module extracts the communication status record of the unmanned vessel in the current period based on the wind direction shift and wave height jump trend marked in the navigation environment disturbance label, identifies the signal connection change sequence of each communication channel, and determines whether an interruption occurs during the wind direction change and wave height jump period. For continuously connected channels, it records the complete communication period; for interrupted channels, it records the number of interruptions. Based on the number of interruptions, it sorts all channels in order from low to high and generates a sorted list of primary channels.

[0053] The data filtering module reads the highest priority channel number from the main channel sorting list, retrieves the task data to be transmitted in the channel, filters the task data with the highest urgency level and whose generation time has not been delayed, classifies the filtered task items that meet the conditions into high-priority task data, removes the content that belongs to the retransmission record, keeps the original order of the valid data unchanged, writes it into the number sequence according to the extraction order, and generates an instant task information index set.

[0054] The task scheduling module calls the task number in the real-time task information index set, reads the target behavior direction corresponding to each task number, compares it with the real-time travel direction of the current unmanned vessel hull determined by the track record, determines whether there is a deviation between the target behavior direction and the real-time travel direction and exceeds the allowable deflection range, filters out the deviated task numbers, compares the target position pointed to by each task number with the turning node marked in the current navigation path icon, marks the tasks that enter the change of direction section, summarizes all marked tasks and adjusts their order, and generates a task execution order group.

[0055] The instruction generation module reads each task number and corresponding control behavior content from the task execution order group, determines whether the task involves thruster action or directional control operation, extracts the description of propulsion direction, rudder angle offset and control continuity from the content, writes the tasks with control behavior characteristics into the control buffer, encapsulates the operation set according to the task order, and generates the hull control instruction result.

[0056] The navigation environment disturbance indicators include wind speed direction deviation, water flow deflection, wave height change trend, wind and wave interference time point, hull disturbance direction, and time period of inconsistent motion characteristics. The primary channel sorting list includes communication channel number, signal connection change record, wind direction change period, wave height jump period, and interruption number sorting. The real-time mission information index set includes high priority mission data, emergency level mark, generation time mark, mission data number, and valid data sorting. The mission execution order group includes target behavior direction, current travel direction deviation, target position and navigation path turning node, deviation mission number, and change section mission mark. The hull control command results include propulsion direction, rudder angle deviation, control duration, and mission control set.

[0057] Please see Figure 2 The environmental perception module includes:

[0058] The data acquisition submodule collects real-time data on wind speed, water flow direction, and wave height during the unmanned vessel's navigation, records the changes in wind speed, water flow direction, and wave height in each time period, and updates the data in real time according to the set time intervals during data acquisition, generating a dataset of wind speed, water flow direction, and wave height.

[0059] The system collects raw sensor signals from the wind speed propeller cover on the foredeck, the side wave measurement plate, and the wake monitoring camera in the aft cabin at set time intervals. It performs Fast Fourier Transform (FFT) on the propeller cover's rotation frequency signal to accurately extract the dominant frequency and convert it into wind speed. It performs spatial difference and time series analysis on the pressure sensor matrix data below the wave measurement plate to calculate the effective wave height and period. Local flow field data is verified using particle image velocimetry (PIV) via the wake monitoring camera in the aft cabin. During a specific navigation period, for example, from 03:00:00 to 03:05:00 UTC, the system continuously collects navigation environment parameters at a frequency of 1 Hz. Instantaneous wind speed recorded by the wind speed sensor, water flow profile data recorded by the Acoustic Doppler Current Profiler (ADCP), and wave height data recorded by the wave sensor are collected synchronously. The process combines the wind speed value (unit: m / s), surface water flow direction (unit: degrees), significant wave height value (unit: meters), corresponding geographical coordinates (latitude and longitude), and sensor self-check status code (e.g., 0x00 indicates normal, 0x01 indicates data out of range) for each timestamp into a complete data record. For example, at 03:01:10 UTC, the collected wind speed is 12.5 m / s, the water flow direction is 215 degrees, and the wave height is 1.8 m, with a status code of 0x00. All collected data records are stored in a fixed-size circular buffer in memory to achieve efficient "first-in, first-out" data coverage. At the end of each collection cycle, for example, every 5 minutes, this dataset undergoes integrity verification and packaging integration. After removing bad frames with status codes other than 0x00, a preliminarily verified dataset of wind speed, water flow direction, and wave height is generated, as shown in Table 1.

[0060] Table 1. Examples of Raw Data Collection for Unmanned Vessel Navigation Environment

[0061] Timestamp (UTC) Wind speed (m / s) Surface water flow direction (°) Significant wave height (m) Geographic coordinates (E, N) status codes 03:01:09 12.2 216 1.7 115°10.1',22°05.4' 0x00 03:01:10 12.5 215 1.8 115°10.2',22°05.5' 0x00 03:01:11 12.6 215 1.9 115°10.2',22°05.5' 0x00 03:01:12 11.9 214 1.8 115°10.3',22°05.6' 0x00

[0062] Table 1 shows a portion of the raw data records continuously collected within the circular buffer. Each row constitutes a complete record containing information about the environment, time and space, and equipment status.

[0063] The feature extraction submodule, based on the dataset of wind speed, water flow direction and wave height, extracts the changing trends of wind speed direction, water flow deflection and wave height jumps. By analyzing the fluctuation characteristics of the data, it identifies the changes in wind speed direction, the magnitude of water flow deflection and the jumps in wave height, and generates wind and wave change trends.

[0064] Based on the datasets of wind speed, water flow direction, and wave height, vector decomposition is performed on the wind speed data to calculate the east-west wind component over the most recent 10 seconds (10 data points). Components of north-south winds The arithmetic mean of the wind direction is calculated, and then the average wind speed and direction are inversely calculated using the arctangent function. The difference between the currently calculated average wind direction and the average wind direction of the previous 10-second period is used to determine the magnitude of the wind direction change, thus smoothing out the interference of instantaneous gusts. For example, if the average wind direction in the previous period was 45 degrees, the calculated average wind direction in this period is 62 degrees, with a change of 17 degrees. Next, the water flow deflection is processed. The average heading value (e.g., 90.5 degrees) provided by the ship's fiber optic gyroscope and smoothed by the Kalman filter algorithm is retrieved as a high-precision reference baseline. The angle between the water flow direction (e.g., 115 degrees) at each moment and this reference baseline is calculated. The absolute value of the angle was used to obtain a water flow deflection angle of 24.5 degrees. Then, wave height jumps were analyzed, calculating the difference in wave height readings every 2 seconds (i.e., between the current point and the previous two data points). This difference was compared to a dynamically adjusted wave height jump threshold. This threshold was determined based on the standard deviation of wave height fluctuations within a 95% confidence interval under Force 7 winds, statistically analyzed from historical navigation data in the current sea area. Specifically, wave height data was collected for 1000 hours of navigation, and the standard deviation of wave height at Force 7 was calculated to be 0.4 meters. 1.96 times this standard deviation, or 0.784 meters, was used as the base threshold for wave height jumps. And combined with the hull size correction factor (For ships less than 20 meters in length, use 1.1) Final threshold When the calculated wave height difference within 2 seconds is greater than 0.86 meters, it is determined as a wave height jump. For example, if the wave height is 1.5 meters at 03:02:30 and 2.4 meters at 03:02:32, the difference is 0.9 meters, which is greater than 0.86 meters, so a wave height jump is recorded. The wind direction change amplitude, water flow deflection angle and wave height jump status (Boolean value) calculated at all time points are combined to generate the wind and wave change trend.

[0065] The disturbance identification submodule identifies whether there are time periods that are out of sync with environmental changes during the movement process based on the difference between the trend of wind and waves and the orientation and direction of movement of the unmanned vessel. It analyzes the disturbance situation of the vessel, marks the time of wind and wave interference, and marks the specific location according to the direction of the disturbance, and generates a navigation environment disturbance identification.

[0066] Based on the differences between the changing wind and wave trends and the unmanned surface vessel's orientation and direction of motion, the ship's orientation data (HDG) recorded by the fiber optic gyro inertial navigation system (INS) and the ground trajectory data (COG) recorded by the differential global positioning system (RTK-GPS) with centimeter-level accuracy are retrieved. Before comparison, the rudder angle sensor readings are checked. If the rudder angle readings are within... If the rudder angle reading is outside the zero dead zone, it is determined that the unmanned vessel is executing active control commands, and the disturbance judgment for this cycle will be skipped to avoid misjudgment. If the rudder angle reading is 0 degrees, the instantaneous difference between the two at the same timestamp is calculated. At the same time, an adaptive threshold for judging motion inconsistency is set. The threshold was set with reference to the International Maritime Organization (IMO) requirements for course stability in ship maneuverability standards, and was verified experimentally. The experiment selected 50 sets of navigation data under different sea states, and statistically showed that when the wind force is less than 4, the wave height is less than 1 meter, and there is no active maneuvering, The average value is 2.1 degrees, and the standard deviation is 0.8 degrees. The base threshold is calculated by adding three times the average value to the standard deviation. Degrees, when the rudder angle is 0 and If the angle exceeds 4.5 degrees for more than 3 consecutive seconds, this time period is identified as an inconsistency in motion caused by the external environment. For example, between 03:02:30 and 03:02:35, the wind direction suddenly changes to 62 degrees, the wave height jumps to 0.9 meters, and the rudder angle reading is 0 degrees. At the same time, the calculated HDG is 90 degrees and COG is 96 degrees during this period. If the wind and wave angle is 6 degrees and remains greater than 4.5 degrees for 5 seconds, then 03:02:30 is marked as the time of wind and wave interference. Further, based on the wind direction (62 degrees) and current direction (115 degrees) at that time, the direction and magnitude of the resultant force of the wind and current vectors relative to the ship's bow direction (90 degrees) are calculated. The dominant disturbance direction is determined to be approximately 30 degrees to the starboard side of the bow. This direction and the interference level (based on...) are then... The size classification, timestamp of the interference time point, and geographical location (e.g., 115°10.2' E, 22°05.5' N) are recorded together to generate a navigation environment disturbance identifier.

[0067] Please see Figure 3 The channel adjustment module includes:

[0068] The signal extraction submodule extracts the communication status record of the unmanned vessel in the current period based on the wind direction shift and wave height jump change trend marked in the navigation environment disturbance mark, obtains the signal connection change sequence of each communication channel, records the signal status of each channel in a specified time period, and generates a communication status record.

[0069] Based on the wind direction shift and wave height jump trends marked in the navigation environment disturbance indicators, the connection status information of the high-frequency radio wave transmitter, the 5G long-range beamforming substrate, and the synchronous satellite waveguide antenna within the current period is read to pinpoint the wind and wave interference time point, i.e., 03:02:30. Using this time point as the center, a dynamic time window of 60 seconds is formed by extending 30 seconds before and after it (03:02:00 to 03:03:00). Within this time window, the system concurrently retrieves four sets of heterogeneous communication equipment on the unmanned vessel (Channel 1: Beidou satellite communication, ...). The system collects communication logs and real-time diagnostic interface data for Channel 2 (nearshore 4G / LTE network), Channel 3 (offshore 5G millimeter-wave network), and Channel 4 (Iridium global satellite communication). The logs detail the received signal strength (RSRP), signal-to-noise ratio (SINR), packet round-trip delay (Latency), and bit error rate (BER) per second for each channel. The system uses a multi-dimensional connectivity criterion: RSRP greater than -110dBm, SINR greater than 3dB, Latency less than 800ms, and BER less than [missing value]. Only when the condition is met is it considered "valid connection" and assigned a value of 1. If any condition is not met, it is considered "interrupted" and assigned a value of 0. The signal connection status of each channel within the 60-second window is extracted to form a binary sequence composed of 0s and 1s. For example, the sequence for channel 2 is [1,1,…,1,0,0,0,1,…,1]. Consecutive 0s indicate that a signal interruption occurred during the interference period. The binary sequences of these four channels and all corresponding diagnostic data are organized to generate a communication status record, as shown in Table 2.

[0070] Table 2. Communication log excerpts during periods of wind and wave interference.

[0071] Timestamp (UTC) aisle RSRP (dBm) SINR (dB) Delay (ms) BER Connectivity 03:02:31 2(4G) -108 5.2 120 5e-6 1 03:02:32 2(4G) -112 2.5 350 9e-5 0 03:02:33 2(4G) -115 1.8 410 1e-4 0 03:02:32 3(5G) -105 6.0 80 2e-6 1 03:02:33 3(5G) -111 2.1 500 2e-4 0 03:02:34 3(5G) -106 5.5 95 4e-6 1

[0072] As shown in Table 2, the table details the changes in key performance indicators of some communication channels during the core period of wind and wave interference (around 03:02:30). Among them, the RSRP, SINR and other indicators of channels 2 and 3 deteriorated instantly, causing the connectivity status to be judged as interrupted (0).

[0073] The connection judgment submodule, based on the communication status record, determines whether a communication interruption occurred during the period of sudden wind direction change and wave height jump, analyzes the connection status of each channel in different time periods, marks the time period of signal interruption, records the interruption frequency of each communication channel, and generates an interruption frequency set.

[0074] Based on communication status records, this study focuses on analyzing the binary sequence values ​​of the connection status of each communication channel during a specific time period defined by sudden changes in wind direction and wave height, from 03:02:30 to 03:02:35. The sequence value for channel 1 (BeiDou) is [1,1,1,1,1], for channel 2 (4G) it is [1,0,0,0,1], for channel 3 (5G) it is [1,0,1,0,1], and for channel 4 (Iridium) it is [1,1,1,1,1]. Signal interruptions are marked by identifying the occurrence of "0" in the sequence, and the number of interruption events for each channel is precisely counted within the entire 60-second observation window. This counting process is implemented using a state machine, which detects the falling edge of the signal state from "1" to "0". To trigger the start of an interrupt event, for each falling edge detected, the interrupt frequency accumulator of the corresponding channel is incremented by 1, and the state machine is locked during a continuous "0" state to prevent double counting. The counting trigger is reset only when a rising edge from "0" to "1" is detected, preparing for the next count. According to this logic, channel 1 has no falling edges in this window, so the interrupt frequency is 0; channel 2 has 1 falling edge in the interference period, so the total interrupt frequency is 1; channel 3 has 2 non-contiguous falling edges in the interference period, so the total interrupt frequency is 2; and channel 4 has no falling edges, so the interrupt frequency is 0. The channels and their corresponding interrupt frequencies are organized into a key-value pair set: {channel 1:0, channel 2:1, channel 3:2, channel 4:0}, generating the interrupt frequency set.

[0075] The channel sorting submodule sorts all communication channels according to the interrupt frequency set, placing channels with lower interrupt frequencies at the top with higher priority, and vice versa, generating a primary channel sorting list.

[0076] Based on the interruption frequency set, data from {Channel 1:0, Channel 2:1, Channel 3:2, Channel 4:0} is retrieved. A multi-level sorting algorithm is used to dynamically prioritize all communication channels, arranging them in ascending order of interruption frequency from low to high. When interruption frequencies are the same (such as Channel 1 and Channel 4 in this example), a second sorting criterion is activated: the average signal-to-noise ratio (SINR) of each channel within the observation window is retrieved from the communication status record. This average SINR is the arithmetic mean of the SINR values ​​of all data points with a status of "1" (effective connectivity) within the window, ensuring that the evaluation reflects true communication quality rather than noise. Channels with higher noise ratios are assigned higher priorities. For example, if the interrupt frequency of channels 1 and 4 is 0, the average SINR is calculated. Suppose the average SINR of channel 1 is 18dB and the average SINR of channel 4 is 15dB. Since 18dB is greater than 15dB, the priority of channel 1 is higher than that of channel 4. The interrupt frequency of channel 2 is 1 and that of channel 3 is 2. Combining the two levels of sorting, the final priority result is: channel 1, channel 4, channel 2, channel 3. This sorting result is stored in a dedicated priority queue register in the form of an ordered list and broadcast to the task scheduling module to generate the primary channel sorting list.

[0077] Please see Figure 4 The data screening module includes:

[0078] The task filtering submodule retrieves the task information to be transmitted in the channel with the highest priority channel number in the primary channel sorting list, analyzes each task item in the task, filters out the task items marked with the highest urgency level and whose generation time has not been delayed, extracts the task items that meet these conditions, and generates high-priority task items.

[0079] Based on the highest priority channel number in the primary channel sorting list, namely channel 1 (BeiDou satellite communication), the currently bound FIFO (First-In-First-Out) queue for the pending transmission tasks is immediately locked. This queue is named "TX-Buffer-CH1" at the system level. This queue currently contains 5 task items, each of which is a data structure containing multiple fields, such as: urgency level, generation timestamp, task hash value, and whether it is delayed. The data structures of these 5 task items are filtered one by one using bitwise AND logic. The filtering conditions are "urgency level is marked as the highest" and "generation time is not marked as delayed". The urgency level is divided into three levels: the highest (value 3, corresponding to flag 0x03), the second highest (value 2, corresponding to 0x02), and so on. Generally (value 1, corresponding to 0x01), the delay flag is a Boolean value (no / yes, corresponding to bit 0 / 1). The filtering logic expression is (task item.urgency level == 0x03) AND (task item.delay flag == 0b0). Task item 1 has an urgency level of 0x03 and a delay flag of 0, which meets the condition. Task item 2 has an urgency level of 0x02, which does not meet the condition. Task item 3 has an urgency level of 0x03 but a delay flag of 1, which does not meet the condition. Task item 4 has an urgency level of 0x03 and a delay flag of 0, which meets the condition. Task item 5 has an urgency level of 0x01, which does not meet the condition. The memory address pointers of the tasks that meet the condition, namely task item 1 and task item 4, are extracted and stored in a temporary array to generate high-priority task items, as shown in Table 3.

[0080] Table 3. Information on Tasks to be Transmitted

[0081] Task item number Emergency Level Generation time Delayed marking 1 3 03:01:50 no 2 2 03:02:10 no 3 3 02:58:30 yes 4 3 03:02:40 no 5 1 03:02:55 no

[0082] As shown in Table 3, logical judgments are made based on the "urgency level" and "delayed" fields of the table to filter out task item 1 and task item 4.

[0083] The data cleaning submodule identifies and removes duplicate records based on high-priority task items, ensuring the uniqueness of each task item while maintaining the original order of valid tasks. After removing redundant content, it sorts out the task items that meet the conditions and generates valid task items.

[0084] Based on the high-priority task items (containing pointers to task items 1 and 4), a deduplication and verification process is initiated. To prevent duplicate task execution, the module performs standardized string concatenation on the "target location coordinates" and "behavior type code" fields in the data structure of each task item. Then, it calculates the SHA-256 hash value of the concatenated string and stores this hash value as a unique identifier in a temporary set (HashSet) with fast lookup functionality. Assuming there is a resend record in the system log that is completely identical to task item 4 (i.e., the target location and behavior type code are the same), but with the task number 4-retry, this record is also incorrectly classified as a high-priority task item. When calculating the hash value of task item 4-retry... When the system finds a matching hash value from task item 4 in the temporary set, a conflict comparison mechanism is triggered. The system retrieves the timestamps of both items and, following the principle of "keeping the earliest generated," retains task item 4 (timestamp 03:02:40). The system then safely removes the pointer to task item 4-retry from the pointer set. In this example, there are no duplicate records in the high-priority task items. After processing, the task item pointer list remains [Ptr_Task1, Ptr_Task4]. The original order of this list is maintained, that is, the order in which it was selected in the queue to be transmitted, without reordering it to respect the temporal order of task generation. The sorted list is the unique set of task items that meet the conditions, and a valid task item is generated.

[0085] The index generation submodule assigns a number to each task according to the extraction order based on the valid task items, and writes the task numbers into the number sequence in the order of extraction to ensure the correctness of the task execution order and generate an instant task information index set.

[0086] After obtaining the real-time task information index set, the priority and time window of the task items are classified and labeled. Task items with similar priority characteristics, urgency requirements and task time limits are grouped into the same category and assigned corresponding task priority labels. Differentiated scheduling strategies are executed based on task priority labels during subsequent task scheduling and execution.

[0087] Based on the valid task items, temporary execution numbers are assigned to the two task items [Ptr_Task1, Ptr_Task4]. This is done to facilitate subsequent referencing by the scheduling module. The assignment rule follows the extraction order in the list of valid task items, i.e., task item 1 first, task item 4 second, and consecutive integer numbers starting from 1 are assigned sequentially. Task item 1 is assigned the new number 1, and task item 4 is assigned the new number 2. These new numbers are written sequentially into an integer array or sequence to obtain the number sequence [1,2]. This sequence clarifies the basic order of task processing and generates an immediate task information index set. After obtaining this index set, the system continues to analyze the deeper attributes of task items 1 and 4. The objective of Task 1 (original number 1) is "emergency avoidance of channel obstacles," and its metadata contains the highest priority trait (value 10). The task time limit (TOL, TimeOnLine) is 60 seconds to complete. The objective of Task 4 (original number 4) is "update route planning points," with the second highest priority trait (value 8) and a task time limit of 120 seconds to complete. Therefore, the system classifies Task 1 into the "emergency avoidance group" and assigns it the tags "highest priority" and "time-sensitive," while classifying Task 4 into the "route management group" and assigning it the tags "second highest priority" and "non-time-sensitive." These tags will affect subsequent resource allocation and execution strategies.

[0088] Please see Figure 5 The task scheduling module includes:

[0089] The behavior comparison submodule reads the target behavior direction corresponding to each task number based on the task number in the real-time task information index set, and compares it with the real-time travel direction determined by the current unmanned vessel's trajectory record. It then determines whether there is an offset between the target behavior direction and the real-time travel direction, and further determines whether the offset exceeds the allowable deflection range. If there is an offset and it exceeds the set threshold, it is marked as a task that needs to be adjusted, and an offset detection result is generated.

[0090] Based on the task number in the real-time task information index set, task number 1 is read, whose corresponding target behavior is a 20-degree leftward deviation of the heading. This module calculates the target heading (Target Heading) to be 70 degrees based on the current planned heading of 90 degrees. The module then retrieves the unmanned surface vessel's real-time track record and determines the current real-time heading (COG) output by the RTK-GPS module, filtered by a moving average over the past 5 seconds, to be 96 degrees. This eliminates errors caused by fluctuations in single-point GPS data. Next, the absolute offset between the target heading and the real-time heading is calculated. Then, this offset is compared with a dynamic allowable deflection range threshold. For comparison, this threshold is set based on a mathematical model of the hull's turning performance, with the current speed (e.g., 10 knots) as the variable. Experimental data shows that at a speed of 10 knots, for every 5-degree change in the rudder angle, the maximum stable change in heading within 10 seconds is 15 degrees. To ensure hull attitude stability and avoid equipment overload, the system sets 80% of this value as the instantaneous allowable deflection threshold, i.e. Since the calculated offset of 26 degrees is greater than the threshold of 12 degrees, task number 1 is marked as a "high offset task" and needs to be adjusted. The same operation is performed on task number 2 (update course), with its target behavior direction being "maintain current planned heading," which is 90 degrees, and the offset is... If the degree is less than 12 degrees, it is not marked. The offset detection result is generated by combining the comparison results of all tasks.

[0091] The offset filtering submodule, based on the offset detection results, filters out all task numbers whose offset between the target behavior direction and the actual travel direction exceeds the allowable range, classifies the tasks, identifies tasks with larger offsets, marks them, and generates offset task numbers.

[0092] Based on the offset detection results, the module iterates through the marked states of all tasks and extracts all tasks marked as "high offset tasks" through logical filtering. These tasks share the common characteristic that the offset between their target direction and the UAV's current actual direction of travel exceeds the dynamic threshold of 12 degrees. According to the calculations of the previous module, only task number 1 has an offset of 26 degrees exceeding this threshold. Therefore, the filtered result is a list containing [task number 1]. To achieve more refined control, the module classifies the filtered tasks by offset level, assigning them to different levels based on the magnitude of the offset. The preset offset level classification standard is: offset within 12 degrees (i.e., ... Offsets between 26 and 25 degrees are classified as "Level 1 Offsets," meaning they can be adjusted using conventional rudder angles. Offsets greater than 25 degrees are classified as "Level 2 Offsets," which may require auxiliary steering in conjunction with thruster differentials. Since the offset of mission number 1 is 26 degrees, which is greater than 25 degrees, it is classified as a "Level 2 Offset" mission and marked. This mark will be written into the mission's attribute data structure as a critical state flag. All the filtered and categorized mission numbers will be aggregated to generate the offset mission number.

[0093] The order adjustment submodule compares the target position of each task with the turning nodes in the current navigation path diagram based on the offset task number, identifies those tasks that need to change direction, and adjusts their execution order for tasks entering the turning section so that the tasks are reasonably sorted according to the turning nodes in the predetermined navigation path, generating a task execution order group.

[0094] After obtaining the task execution order group, the execution order of different tasks is re-evaluated and optimized. Tasks with similar target locations, task types, and variable requirements are grouped into the same category, and a corresponding priority execution strategy is assigned to each task. The execution order is adjusted differently according to the variable requirements of the task and the execution time period.

[0095] Based on the offset task number [Task Number 1], the system retrieves the target location (a latitude and longitude point) of the task and the planned navigation path map containing the next 5 nautical miles loaded in the current task computer. This path map labels all turning points (TP, Turning Point) in data structure form. Using a geographic information algorithm, the module compares the target location coordinates of Task 1 with the coordinates of the turning point closest to the current location in the path map. It finds that the target location of Task 1 is 500 meters before the next turning point (TP-05), and the route planning of TP-05 requires the ship to turn from a 90-degree heading to a 60-degree heading. The system identifies that the behavioral direction of Task 1 (turning to 70 degrees for obstacle avoidance) and the upcoming planned route turn (turning to 60 degrees) are highly overlapping spatially and closely connected temporally. This change of direction requirement is related to and potentially conflicts with the behavioral direction of Task 1. Therefore, the system will... Task 1 is marked as "associated task entering the pre-turning section". At the same time, the same position comparison is performed on other tasks in the task queue that are not marked as offset (such as task number 2). The target position of task 2 is a data upload point on the main channel. Since it does not involve turning, it is not specially marked. All marked tasks (only task 1) are summarized, and the execution order is dynamically adjusted based on the rule base. The priority of the task marked as "associated task entering the pre-turning section" is raised to the highest and placed at the front of the task queue, while the order of other tasks (task 2) is postponed. The task execution order group [task 1, task 2] is generated. This group is re-evaluated. Since the target positions, task types and turning requirements of task 1 and task 2 are not similar, they cannot be merged into the same group. Therefore, the original adjustment order is maintained, and the strategy of "prioritizing execution and linking with the route turning controller" is assigned to task 1, and the strategy of "execution after the attitude of the main channel is stable" is assigned to task 2.

[0096] Please see Figure 6 The instruction generation module includes:

[0097] The task reading submodule reads each task number and its corresponding control behavior content according to the task execution order, extracts the description of thruster action or direction control operation in the task, obtains the detailed control requirements of each task, and summarizes the relevant control information of the task by traversing the task number to generate task control information.

[0098] Based on the task execution order group [Task 1, Task 2], the first task number (1) is read from the head of the queue. Its memory pointer is used to retrieve the corresponding complete control behavior content from the task database. This content is stored as standardized structured data, not plain text, containing a unique action code and several associated parameters. Task 1's action code is 0x1A (predefined in the system as "turn to specified heading"), and its parameter set is {target heading: 70.0, maximum rudder angle: -35.0, angular velocity limit: 5 degrees / second, duration: dynamic feedback}. Next, task number 2 is read, with an action code of 0x5B (representing "data packet push") and a parameter set of {target module ID: master computer, data packet pointer: 0x4A8F00, transmission protocol: TCP}. Following the task execution order group, this module iterates through all task numbers, extracts the structured control requirements for each task, and loads this information sequentially into a dedicated execution buffer to prepare for subsequent instruction parsing and generation, thus generating task control information.

[0099] The behavior extraction submodule, based on task control information, determines whether the task involves control behaviors such as thruster direction or rudder angle deviation, extracts the thrust direction, rudder angle deviation and control duration from the task, classifies these control behaviors, filters out task items containing control behavior features, and generates control behavior content.

[0100] Based on the task control information flow, a preset "behavior code-action mapping table" stored in non-volatile memory is used to parse the tasks in the execution buffer one by one. For task 1, its behavior code 0x1A is judged in the mapping table to be a physical control operation involving the ship's attitude. Then, the submodule extracts the key control parameters according to the mapping rules: the propulsion direction (since it is not specified, it is maintained by the propulsion controller) is defined as "unchanged", the rudder angle offset is determined as "full left rudder" according to the parameter {maximum rudder angle: -35.0}, that is, -35 degrees (the system specifies that left rudder is negative and right rudder is positive). The dynamic process of control must simultaneously meet the constraint {angular velocity limit: 5 degrees / second}. The duration of control is determined according to the parameter {continuation mode: dynamic feedback}, that is, the servo will maintain -35 degrees or adjust within the limit until the real-time reading of the ship's heading sensor (such as a fiber optic compass) stabilizes at 70 degrees of the target heading. Within the tolerance range of degrees, for Task 2, its behavior code 0x5B is judged by the mapping table as "software-level data interaction operation", which does not involve direct physical control of the thruster or rudder. Therefore, physical control behavior parameters are not extracted. Task items containing specific physical control behavior characteristics (only Task 1) are selected and fully parameterized as {Task number: 1, thrust direction: 0, rudder angle offset: -35 degrees, angular velocity limit: 5, control duration: dynamic feedback, termination condition: HDGin[69,71]}, generating control behavior content.

[0101] The instruction encapsulation submodule encapsulates tasks with control behavior characteristics according to their execution order based on the content of the control behavior, organizes the control operation set corresponding to each task, arranges the operation set according to the task order, and generates the hull control instruction result.

[0102] Based on the control behavior content, the task with control behavior characteristics {Task Number: 1, Propulsion Direction: 0, Rudder Angle Offset: -35 degrees, Angular Velocity Limit: 5, Control Duration: Dynamic Feedback, Termination Condition: HDGin[69,71]} is retrieved. Following the task execution order [Task 1, Task 2], the control operation set of this task is placed at the beginning of the underlying instruction sequence. For Task 2, a software-level execution instruction is generated. The operation set is encapsulated into a standardized data frame structure conforming to the unmanned vessel's internal communication bus protocol (such as CAN-FD or Time-Sensitive Network TSN). This structure includes a frame header, priority identifier, instruction type code (e.g., 0x01 represents attitude control), instruction parameters (rudder angle value -35, angular velocity limit 5), CRC checksum, and frame tail. The encapsulated operation set is a single instruction sent to the servo controller.

[0103] "SET_RUDDER_ANGLE_WITH_RATE_LIMIT(angle:-35,rate:5)", and at the same time, a high-frequency monitoring condition is set in the motion control main loop. When the return value of the "CURRENT_HEADING()" function enters the range [69,71] and remains stable for more than a preset time (such as 2 seconds), the next instruction "SET_RUDDER_ANGLE(0)" is automatically triggered to return the rudder. Since task 2 is a data transmission task, its operation set is encapsulated into an instruction to the communication management module:

[0104] The command “SEND_DATA(destination:MainComputer,pointer:0x4A8F00,protocol:TCP)” arranges these encapsulated instruction sets, which have execution order and trigger conditions, in the order of [Task 1 (including termination and rudder return logic), Task 2] to generate the ship control instruction results.

[0105] 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. An intelligent Internet of Things control system for unmanned vessels, characterized in that, The system includes: The environmental perception module collects wind speed, water flow and wave height data during the unmanned vessel's navigation, extracts the changing trends, and, combined with the difference between the vessel's orientation and direction of movement, identifies the periods of inconsistent movement, marks the time points of wind and wave interference and the direction of disturbance to the vessel, and generates navigation environment disturbance labels. The channel adjustment module analyzes the connection change sequence of each communication channel during the period of sudden wind and wave changes based on the wind direction shift and wave height change trend in the navigation environment disturbance indicator, counts the number of interruptions of each channel, and sorts them from low to high according to the number of interruptions to generate a primary channel sorting list. The data filtering module reads the priority channel number from the primary channel sorting list, retrieves the task data to be transmitted from the channel, filters urgent and non-delayed tasks, removes duplicate records, keeps the order unchanged and writes them into the number sequence, and generates an instant task information index set. The task scheduling module calls the task number in the real-time task information index set, compares the task target with the ship's real-time direction of travel, filters out tasks with excessive deflection, marks the change-of-direction tasks related to the course turning node, summarizes and optimizes the order of the marked tasks, and generates a task execution order group.

2. The intelligent IoT control system for unmanned vessels according to claim 1, characterized in that: The navigation environment disturbance indicators include wind speed direction deviation, water flow deflection, wave height change trend, wind and wave interference time point, hull disturbance direction, and time period of inconsistent motion characteristics. The primary channel sorting list includes communication channel number, signal connection change record, wind direction change period, wave height jump period, and interruption number sorting. The real-time task information index set includes high-priority task data, emergency level marker, generation time marker, task data number, and valid data sorting. The task execution order group includes target behavior direction, current travel direction deviation, target position and navigation path turning node, deviation task number, and change section task marker.

3. The intelligent IoT control system for unmanned vessels according to claim 1, characterized in that, The environment sensing module includes: The data acquisition submodule collects real-time data on wind speed, water flow direction, and wave height during the unmanned vessel's navigation, records the changes in wind speed, water flow direction, and wave height in each time period, and updates the data in real time according to the set time intervals during data acquisition, generating a dataset of wind speed, water flow direction, and wave height. The feature extraction submodule extracts the changing trends of wind speed direction, water flow deflection, and wave height jumps based on the wind speed, water flow direction, and wave height dataset. By analyzing the fluctuation characteristics of the data, it identifies the changes in wind speed direction, the magnitude of water flow deflection, and the jumps in wave height, and generates wind and wave change trends. The disturbance identification submodule identifies whether there are time periods that are out of sync with environmental changes during the movement process based on the difference between the trend of wind and waves and the orientation and direction of movement of the unmanned vessel. It analyzes the disturbance situation of the vessel, marks the time of wind and wave interference, and marks the specific location according to the direction of the disturbance, and generates a navigation environment disturbance identification.

4. The intelligent IoT control system for unmanned vessels according to claim 1, characterized in that, The channel adjustment module includes: The signal extraction submodule extracts the communication status record of the unmanned vessel in the current period based on the wind direction shift and wave height jump change trend marked in the navigation environment disturbance identifier, obtains the signal connection change sequence of each communication channel, records the signal status of each channel in a specified time period, and generates a communication status record. The connection judgment submodule, based on the communication status record, determines whether a communication interruption occurred during the period of sudden wind direction change and wave height jump, analyzes the connection status of each channel in different time periods, marks the time period of signal interruption, records the interruption frequency of each communication channel, and generates an interruption frequency set. The channel sorting submodule sorts all communication channels according to the interruption frequency set, placing channels with lower interruption frequencies at the top with higher priority, and vice versa, generating a primary channel sorting list.

5. The intelligent IoT control system for unmanned vessels according to claim 1, characterized in that, The data screening module includes: The task filtering submodule retrieves the task information to be transmitted in the channel based on the highest priority channel number in the main channel sorting list, analyzes each task item in the task, filters out the task items marked with the highest urgency level and whose generation time has not been delayed, extracts the task items that meet these conditions, and generates high-priority task items. The data cleaning submodule, based on the high-priority task items, identifies and removes duplicate records to ensure the uniqueness of each task item, while maintaining the original order of valid tasks. After removing redundant content, it sorts out the task items that meet the conditions and generates valid task items. The index generation submodule assigns a number to each task according to the extraction order based on the valid task items, and writes the task numbers into the number sequence in the extraction order to ensure the correctness of the task execution order, thereby generating an instant task information index set.

6. The intelligent IoT control system for unmanned vessels according to claim 1, characterized in that, The task scheduling module includes: The behavior comparison submodule reads the target behavior direction corresponding to each task number according to the task number in the real-time task information index set, and compares it with the real-time travel direction determined by the current unmanned vessel's trajectory record. It determines whether there is an offset between the target behavior direction and the real-time travel direction, and further determines whether the offset exceeds the allowable deflection range. If there is an offset and it exceeds the set threshold, it is marked as a task that needs to be adjusted, and an offset detection result is generated. The offset filtering submodule, based on the offset detection results, filters out all task numbers whose offset between the target behavior direction and the actual travel direction exceeds the allowable range, classifies the tasks, identifies tasks with larger offsets, marks them, and generates offset task numbers. The order adjustment submodule compares the target position pointed to by each task with the turning nodes in the current navigation path diagram based on the offset task number, identifies those tasks that need to change direction, and adjusts their execution order for tasks entering the turning section so that the tasks are reasonably sorted according to the turning nodes in the predetermined navigation path, generating a task execution order group.

7. The intelligent IoT control system for unmanned vessels according to claim 1, characterized in that, The system also includes: The instruction generation module extracts each task number and corresponding control behavior content from the task execution sequence group, determines whether the task involves thruster action and directional control operation, extracts the propulsion direction, rudder angle offset and control continuity description from the content, encapsulates the task with control behavior characteristics, and generates hull control instruction results. The results of the ship control commands include propulsion direction, rudder angle offset, control duration, and mission control set.

8. The intelligent IoT control system for unmanned vessels according to claim 7, characterized in that, The instruction generation module includes: The task reading submodule reads each task number and its corresponding control behavior content according to the task execution order, extracts the description of thruster action or direction control operation in the task, obtains the detailed control requirements of each task, and generates task control information by traversing the task numbers and summarizing the relevant control information of the task. The behavior extraction submodule, based on the task control information, determines whether the task involves control behaviors such as thruster direction or rudder angle deviation, extracts the thrust direction, rudder angle deviation and control duration from the task, classifies these control behaviors, filters out task items containing control behavior features, and generates control behavior content. The instruction encapsulation submodule encapsulates tasks with control behavior characteristics according to the execution order based on the control behavior content, organizes the control operation set corresponding to each task, arranges the operation set according to the task order, and generates the hull control instruction result.

9. The intelligent IoT control system for unmanned vessels according to claim 1, characterized in that: After obtaining the real-time task information index set, the priority and time window of the task items are classified and identified. Task items with similar priority characteristics, urgency requirements and task time constraints are grouped into the same category and assigned corresponding task priority labels. In the subsequent task scheduling and execution process, a differentiated scheduling strategy is executed based on the task priority labels.

10. The intelligent IoT control system for unmanned vessels according to claim 1, characterized in that: After obtaining the task execution order group, the execution order of different tasks is re-evaluated and optimized. Tasks with similar target locations, task types, and variable requirements are grouped into the same category, and a corresponding priority execution strategy is assigned to each task. The execution order is adjusted differently according to the variable requirements and execution time of the task.

Images