Unmanned ship cluster remote control method and system

Abstract

The invention relates to the technical field of remote control and communication, and discloses an unmanned ship cluster remote control method and system, and the method comprises the following steps: building a communication link among a console, a remote control server and unmanned ships based on a communication technology; according to task requirements, the cluster console or the single-boat console is matched with the unmanned boat through the remote control server, and connection between the cluster console or the single-boat console and the predetermined unmanned boat is determined; dividing the task into a plurality of sub-tasks through a hierarchical planning method, and allocating the divided sub-tasks to the corresponding unmanned surface vehicles by using a graph neural network algorithm to generate corresponding control instructions; the cluster console or the single boat console sends a control instruction to the corresponding unmanned boat through the remote control server, and the unmanned boat receives the control instruction and executes a corresponding control action. According to the invention, flexible and efficient control of the unmanned ship cluster is realized, so that the working efficiency and task coordination can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of remote control and communication technology, and in particular to a method and system for remotely controlling a cluster of unmanned boats. Background Art

[0002] Remote control of unmanned boat swarms has become an important research direction in the current development of intelligent and automated technologies. Traditional control methods face great challenges, especially in complex scenarios where multiple unmanned boats need to be managed simultaneously. Although the traditional single console has certain practicality in the early small-scale operation of unmanned boats, it is difficult to cope with the situation where the number of unmanned boats increases and the complexity of tasks increases. Its main limitations are that it is difficult for the console to monitor and manage the operating status of each unmanned boat in real time, the operation response speed is not fast enough, and it is easy to cause the operator to be overloaded. In addition, although the use of multiple consoles can disperse the workload to a certain extent, it is easy to bring about problems such as information overload, uncoordinated operation and uneven task distribution, which may lead to operational errors, waste of resources, and even affect the execution effect of the entire task.

[0003] Although the existing remote control solutions for unmanned boat swarms have made some progress, there are still some obvious deficiencies in practical applications. For example, high communication delay is a common problem, especially in long-distance operations or marine environments with complex signals. This delay may directly affect the efficiency of task execution. In terms of data processing, the current processing architecture cannot effectively cope with massive real-time data, resulting in low information processing efficiency and delayed decision-making, which limits the practical application of the system. In addition, the existing solutions still have deficiencies in system robustness and scalability, making it difficult to cope with complex practical application scenarios and emergencies.

[0004] Unmanned boat technology has broad application prospects in fields such as ocean survey, underwater detection, rescue operations, environmental monitoring, etc. In order to overcome the shortcomings of the prior art, the present invention proposes a remote control method and system for an unmanned boat cluster. Summary of the invention

[0005] In view of the problems in the related technology, the present invention proposes a remote control method and system for an unmanned boat cluster to overcome the above-mentioned technical problems existing in the existing related technology.

[0006] To this end, the specific technical solution adopted by the present invention is as follows:

[0007] According to one aspect of the present invention, a method for remotely controlling a swarm of unmanned boats is provided, comprising the following steps:

[0008] S1. Based on communication technology, establish a communication link between the control console, remote control server and unmanned boat, wherein the control console includes a cluster control console and a single boat control console;

[0009] S2. According to the mission requirements, the cluster control console or the single boat control console matches the unmanned boat through the remote control server to establish the connection between the cluster control console or the single boat control console and the predetermined unmanned boat;

[0010] S3, divide the task into several subtasks through hierarchical planning method, and use graph neural network algorithm to assign the divided subtasks to the corresponding unmanned boats to generate corresponding control instructions;

[0011] S4. The cluster control console or the single boat control console sends a control command to the corresponding unmanned boat through the remote control server. The unmanned boat receives the control command and executes the corresponding control action.

[0012] Furthermore, the communication link between the control console, the remote control server and the unmanned boat is established based on the communication technology, including the following steps:

[0013] S11, the cluster console and the single boat console are connected to a single or multiple remote control servers through a wired or wireless network to establish a communication link between the console and the server;

[0014] S12. The unmanned boat establishes a connection with one or more remote control servers via a radio or satellite to establish a communication link between the unmanned boat and the remote control server.

[0015] Furthermore, the task is divided into several subtasks by the hierarchical planning method, and the divided subtasks are assigned to the corresponding unmanned boats by the graph neural network algorithm, and the corresponding control instructions are generated, including the following steps:

[0016] S31. Clarify the overall task that the unmanned boat swarm needs to complete, analyze the specific requirements of the task, and divide the overall task into several subtasks using the hierarchical planning method;

[0017] S32. Based on the graph neural network algorithm, the divided subtasks are assigned to the corresponding unmanned boats, and the corresponding control instructions are generated.

[0018] Furthermore, the overall task that the unmanned boat cluster needs to complete is clarified, the specific requirements of the task are analyzed, and the overall task is divided into several subtasks in combination with the hierarchical planning method, including the following steps:

[0019] S311. Clarify the overall tasks that the unmanned boat swarm needs to complete, and analyze the mission area, time limit, resource constraints and environmental conditions;

[0020] S312, decomposing the overall task into several main tasks, and prioritizing the main tasks according to their importance and urgency;

[0021] S313, subdividing the main task into specific subtasks, evaluating the resources and capabilities required for each subtask, and setting a specific execution plan for each subtask, wherein the execution plan includes the driving route, navigation speed and communication frequency.

[0022] Furthermore, the graph neural network algorithm is used to assign the divided subtasks to the corresponding unmanned boats and generate corresponding control instructions, including the following steps:

[0023] S321, constructing a graph structure based on the unmanned boat cluster and the subtasks, wherein each node in the graph structure represents an unmanned boat and a subtask, and the edges between the nodes represent the association between the unmanned boat and the subtask;

[0024] S322, initializing a feature vector for each node, and initializing the weight of each edge according to the suitability of the unmanned boat to perform the subtask;

[0025] S323. In each layer of the graph convolutional network model, the unmanned boat and the task node propagate and update information according to the characteristics of the neighboring nodes, and calculate the fitness score of the unmanned boat to perform the subtask through the scoring function combined with the updated feature representation;

[0026] S324, training a graph convolutional network model using historical allocation data of the unmanned boat and the subtasks to obtain a trained graph convolutional network model;

[0027] S325. Based on the trained graph convolutional network model, combined with the features of the unmanned boat node and the task node, predict the fitness score of each unmanned boat to perform each subtask;

[0028] S326. Select the best unmanned boat to perform each subtask according to the fitness score, and generate corresponding control instructions according to the task allocation result.

[0029] Furthermore, the expression of information update is:

[0030]

[0031] The expression for fitness score is:

[0032]

[0033] In the formula, represents the feature representation of the unmanned boat node i in the k+1th layer; represents the feature representation of the unmanned boat node i in the kth layer; σ represents the nonlinear activation function; N(i) represents the neighbor node set of the unmanned boat node i; di ,d j Respectively represent the degrees of unmanned boat node i and subtask node j; W (k) represents the weight matrix of the kth layer; S(i,j) represents the fitness score; h i 、h j represent the feature vectors of the unmanned boat node i and the subtask node j respectively; || represents the concatenation operation of the feature vectors; e ij Represents the characteristics of the edge between the unmanned boat node i and the subtask node j.

[0034] Furthermore, the cluster control console or the single boat control console sends a control instruction to the corresponding unmanned boat through the remote control server, and the unmanned boat receives the control instruction and executes the corresponding control action, including the following steps:

[0035] S41, determining whether the remote control server is connected to the unmanned boat for communication. If so, the cluster control console or the single boat control console sends a control command to the corresponding unmanned boat through the remote control server. If not, the optimal relay unmanned boat is selected through the unmanned boat cluster information integration service, and the cluster control console or the single boat control console sends a control command to the corresponding unmanned boat through the remote control server and the optimal relay unmanned boat.

[0036] S42. The unmanned boat receives the control command, automatically adjusts the navigation speed and communication frequency according to its own energy status, mission urgency and environmental conditions, and performs corresponding control actions based on the adjusted navigation speed and communication frequency.

[0037] Furthermore, the selecting the optimal relay unmanned boat through the unmanned boat cluster information integration service includes the following steps:

[0038] Obtain the location information and network status information of each unmanned boat in the cluster in real time, and monitor the mission load and mission priority of the unmanned boat in real time;

[0039] Based on the location information of the unmanned boat, the candidate unmanned boats whose distance to the target unmanned boat meets the preset threshold are screened out, and the candidate unmanned boats are further screened according to the network status information, task load and task priority to obtain a potential unmanned boat set;

[0040] Weights are assigned to the location, network status, task priority and task load of the unmanned boat respectively, and the comprehensive score of each potential unmanned boat in the potential unmanned boat set is calculated based on the weights, and the optimal relay unmanned boat is selected according to the comprehensive score.

[0041] Furthermore, the calculation formula of the sailing speed is:

[0042] V=V max ×(α v ×E+β v ×T+ωv ×C)

[0043] The calculation formula for the communication frequency is:

[0044] F=F max ×(α f ×E+β f ×T+ω f ×C)

[0045] Where, V represents the sailing speed; V max represents the maximum navigation speed; E represents the energy status score; T represents the mission urgency score; C represents the environmental condition score; α v , β v ,ω v represents the speed weights of energy state, mission urgency and environmental conditions respectively; F represents the communication frequency; F max Indicates the maximum communication frequency; α f , β f ,ω f The frequency weights represent energy status, task urgency and environmental conditions respectively.

[0046] According to another aspect of the present invention, there is provided an unmanned boat swarm remote control system, the unmanned boat swarm remote control system comprising: a communication link establishment module, a matching connection module, a control instruction generation module and a control action execution module;

[0047] The communication link establishment module is used to establish a communication link between the control console, the remote control server and the unmanned boat based on the communication technology, wherein the control console includes a cluster control console and a single boat control console;

[0048] The matching and connecting module is used to match the cluster control console or the single boat control console with the unmanned boat through the remote control server according to the mission requirements, and establish the connection between the cluster control console or the single boat control console and the predetermined unmanned boat;

[0049] The control instruction generation module is used to divide the task into several subtasks by using a hierarchical planning method, and to assign the divided subtasks to corresponding unmanned boats by using a distributed algorithm based on game theory, and to generate corresponding control instructions;

[0050] The control action execution module is used for the cluster control console or the single boat control console to send control instructions to the corresponding unmanned boat through the remote control server, and the unmanned boat receives the control instructions and executes the corresponding control action.

[0051] The beneficial effects of the present invention are:

[0052] 1) The present invention realizes flexible and efficient control of the unmanned boat cluster through wired or wireless network connection between the cluster console, single boat console, remote control server and unmanned boat, and uses radio and (or) Beidou to establish connection, thereby effectively improving work efficiency and task coordination. At the same time, through matching connection, one console can control one or more unmanned boats at the same time, greatly improving the control efficiency.

[0053] 2) The present invention introduces the unmanned boat relay service, selects the appropriate relay unmanned boat through the unmanned boat cluster information integration service, and realizes the transmission of control commands and working data through appropriate routes. The introduction of this service effectively solves the problem of being unable to directly connect to the remote control server, realizes the flexible control of the unmanned boat that cannot be directly connected to the remote control server, and improves the efficiency and reliability of the remote control of the unmanned boat.

[0054] 3) The present invention divides the task into several subtasks by using a hierarchical planning method, and assigns the subtasks to the most suitable unmanned boats in combination with a graph neural network algorithm, which can maximize the use of the resources and capabilities of the unmanned boat cluster. The graph neural network model can automatically match the unmanned boat that is most suitable for performing a specific task according to the state of the unmanned boat, task requirements and environmental conditions, thereby improving the rationality and efficiency of task allocation and ensuring that the task can be completed quickly and accurately.

[0055] 4) The present invention can dynamically adjust the navigation speed and communication frequency of the unmanned boat according to the energy status of the unmanned boat, the urgency of the mission and the environmental conditions. According to the importance of the mission, the current environment and the remaining energy of the unmanned boat, the optimal navigation speed and communication frequency can be automatically selected, thereby optimizing resource utilization, extending the endurance of the unmanned boat, and ensuring the quality of mission completion. This mechanism ensures the smooth execution of the mission in a complex and changing environment, while extending the service life of the unmanned boat. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 The present invention is a flowchart of a method for remotely controlling a swarm of unmanned boats according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0059] According to the technical solution of the present invention, a method and system for remotely controlling a cluster of unmanned boats are provided.

[0060] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a method for remotely controlling a swarm of unmanned boats is provided, comprising the following steps:

[0061] S1. Based on communication technology, establish a communication link between the control console, remote control server and unmanned boat, wherein the control console includes a cluster control console and a single boat control console;

[0062] Specifically, the cluster console and the single boat console are connected to a single or multiple remote control servers through a wired or wireless network to establish a communication link between the console and the server. At the same time, the unmanned boat is connected to one or more remote control servers through a radio and (or) Beidou to establish a communication link between the unmanned boat and the server. Through this connection method, remote communication between the console and the unmanned boat is achieved.

[0063] S2. According to the mission requirements, the cluster control console or the single boat control console matches the unmanned boat through the remote control server to establish the connection between the cluster control console or the single boat control console and the predetermined unmanned boat;

[0064] Specifically, the cluster or single boat console matches the unmanned boat through the remote control server to establish the connection between the cluster console or single boat console and the specific unmanned boat. This matching connection method allows a console to connect to one or more unmanned boats through matching, thereby achieving flexible control of multiple unmanned boats.

[0065] S3, divide the task into several subtasks through hierarchical planning method, and use graph neural network algorithm to assign the divided subtasks to the corresponding unmanned boats to generate corresponding control instructions;

[0066] The process of dividing the task into several subtasks by using a hierarchical planning method, and assigning the divided subtasks to corresponding unmanned boats by using a graph neural network algorithm to generate corresponding control instructions includes the following steps:

[0067] S31. Clarify the overall task that the unmanned boat swarm needs to complete, analyze the specific requirements of the task, and divide the overall task into several subtasks using the hierarchical planning method;

[0068] Specifically, the above-mentioned steps of clarifying the overall task that the unmanned boat cluster needs to complete, analyzing the specific requirements of the task, and dividing the overall task into several subtasks in combination with the hierarchical planning method include the following steps:

[0069] S311. Clarify the overall tasks that the unmanned boat swarm needs to complete, and analyze the mission area, time limit, resource constraints and environmental conditions;

[0070] S312, decomposing the overall task into several main tasks, and prioritizing the main tasks according to their importance and urgency;

[0071] S313, subdividing the main task into specific subtasks, evaluating the resources and capabilities required for each subtask, and setting a specific execution plan for each subtask, wherein the execution plan includes the driving route, navigation speed and communication frequency.

[0072] S32. Based on the graph neural network algorithm, the divided subtasks are assigned to the corresponding unmanned boats, and the corresponding control instructions are generated.

[0073] Specifically, the graph neural network algorithm is used to assign the divided subtasks to the corresponding unmanned boats and generate corresponding control instructions, including the following steps:

[0074] S321. Construct a graph structure based on the unmanned boat cluster and subtasks, where each node in the graph structure represents an unmanned boat and a subtask, and the edges between the nodes represent the association between the unmanned boat and the subtask. The strength of the association can be set with an initial weight based on the capabilities, resources, and requirements of the unmanned boat and the subtask. For example, if the location of a certain unmanned boat is close to a certain subtask, the weight of the edge is larger; if the resources of a certain unmanned boat are insufficient, the weight of the edge is smaller;

[0075] S322, initializing a feature vector for each node, and initializing a weight of each edge according to the suitability of the unmanned boat to perform the subtask. Usually, the weight can be set based on the distance and capability of the unmanned boat and the degree of matching with the task requirements. The larger the weight, the more suitable it is.

[0076] Specifically, initialize the feature vector for each node:

[0077] Unmanned boat node characteristics: including the capabilities of the unmanned boat (speed, energy status, communication capabilities, etc.), current location, mission load, mission priority, etc.

[0078] Subtask node characteristics: including task requirements, task priority, task execution time, etc.

[0079] S323. In each layer of the graph convolutional network model, the unmanned boat and the task node propagate and update information according to the characteristics of the neighboring nodes, and calculate the fitness score of the unmanned boat to perform the subtask through the scoring function combined with the updated feature representation;

[0080] The expression for information update is:

[0081]

[0082] The expression for fitness score is:

[0083]

[0084] In the formula, represents the feature representation of the unmanned boat node i in the k+1th layer; represents the feature representation of the unmanned boat node i in the kth layer; σ represents the nonlinear activation function; N(i) represents the neighbor node set of the unmanned boat node i; d i d j Respectively represent the degrees of unmanned boat node i and subtask node j; W (k) represents the weight matrix of the kth layer; S(i,j) represents the fitness score; h i 、h j represent the feature vectors of the unmanned boat node i and the subtask node j respectively; || represents the concatenation operation of the feature vectors; e ij Represents the characteristics of the edge between the unmanned boat node i and the subtask node j.

[0085] S324, training a graph convolutional network model using historical allocation data of the unmanned boat and the subtasks to obtain a trained graph convolutional network model;

[0086] S325. Based on the trained graph convolutional network model, combined with the features of the unmanned boat node and the task node, predict the fitness score of each unmanned boat to perform each subtask;

[0087] S326. Select the best unmanned boat to perform each subtask according to the fitness score, and generate corresponding control instructions according to the task allocation result.

[0088] S4. The cluster control console or the single boat control console sends a control command to the corresponding unmanned boat through the remote control server. The unmanned boat receives the control command and executes the corresponding control action.

[0089] Specifically, the cluster control console or the single boat control console sends a control instruction to the corresponding unmanned boat through the remote control server, and the unmanned boat receives the control instruction and performs the corresponding control action, including the following steps:

[0090] S41, determining whether the remote control server is connected to the unmanned boat for communication. If so, the cluster control console or the single boat control console sends a control command to the corresponding unmanned boat through the remote control server. If not, the optimal relay unmanned boat is selected through the unmanned boat cluster information integration service, and the cluster control console or the single boat control console sends a control command to the corresponding unmanned boat through the remote control server and the optimal relay unmanned boat.

[0091] The unmanned boat swarm information integration service is an important component. It integrates and analyzes the information of the unmanned boat swarm, determines which unmanned boats cannot directly connect to the remote control server, and selects appropriate relay unmanned boats to assist in control. The relay unmanned boat should have a stable network connection and good communication capabilities to ensure reliable data transmission. The implementation of this service relies on real-time data transmission and analysis to ensure accurate understanding of the status of the unmanned boat.

[0092] Specifically, the method of selecting the optimal relay unmanned boat through the unmanned boat cluster information integration service includes the following steps:

[0093] Obtain the location information of each unmanned boat in the cluster in real time (usually using satellite navigation systems such as GPS or Beidou, and regularly uploading location data to the remote control server to ensure a comprehensive understanding of the distribution of unmanned boats) and network status information (monitor the communication status of each unmanned boat with the remote control server through indicators such as signal strength, delay, and packet loss rate of the communication link, and determine which unmanned boats have unstable communications or cannot communicate directly with the remote server), and monitor the task load and task priority of the unmanned boats in real time, giving priority to unmanned boats with lower current task importance and lighter load as relay nodes;

[0094] Based on the location information of the unmanned boats, unmanned boats with a moderate distance from the target unmanned boats are selected as candidates. Unmanned boats at a longer distance may cause the communication link signal to weaken or increase the delay. The candidate unmanned boats are further screened according to the network status information, task load and task priority to obtain a potential unmanned boat set.

[0095] Network status screening: Among the candidate unmanned boats screened by location, they will be further screened according to network status. Unmanned boats with good communication quality will be given priority to ensure that they can assume stable relay functions. Specifically, the following indicators can be evaluated: high signal strength, low latency, fast data transmission, low packet loss rate, and ensuring data integrity.

[0096] Mission priority screening: Based on the importance and urgency of the current mission of the unmanned boat, evaluate whether it can be used for relay. If some unmanned boats are performing high-priority or urgent missions, they will be excluded from the relay candidate range.

[0097] Mission load screening: Evaluate the mission load of the unmanned boat to ensure that the selected relay unmanned boat has sufficient resources and capabilities to undertake the relay mission without interfering with the execution of its current mission. Unmanned boats with lighter loads or idle ones are more suitable as relay nodes.

[0098] The weights are assigned to the location, network status, task priority and task load of the unmanned boat respectively. Specifically, in this embodiment, the preferred location weight is 0.3, the network status weight is 0.4, the task priority weight is 0.2, and the task load weight is 0.1. The comprehensive score of each potential unmanned boat in the potential unmanned boat set is calculated based on the weights, and the optimal relay unmanned boat is selected according to the comprehensive score.

[0099] In addition, in some complex environments, if a single relay unmanned boat cannot ensure the stability of the communication link, multiple relay unmanned boats can be selected to form a multi-level relay link.

[0100] In this embodiment, the calculation formula of the comprehensive score is: Total score = A1·S1+A2·S2+A3·S3+A4·S4, where A1, A2, A3, and A4 respectively represent the position, network status, task priority, and task load weights of the unmanned boat. S1 represents the position score, which is scored according to the distance between the unmanned boat and the target (the closer the distance, the higher the score). S2 represents the network status score, which is scored based on the communication signal quality, delay, packet loss rate, etc. (the better the communication quality, the higher the score). S3 represents the task priority score, which is scored based on the priority of the current task (the lower the task priority, the higher the score). S4 represents the task load score, which is scored based on the current workload of the unmanned boat (the lighter the load, the higher the score).

[0101] In this embodiment, the introduction of the unmanned boat relay service solves the control problem of unmanned boats that cannot be directly connected to the remote control server, greatly improving the efficiency and reliability of unmanned boat remote control. The unmanned boat cluster information integration service selects a suitable relay unmanned boat, and the transmission of control commands and working data is achieved through a suitable route, so that the control instructions can be transmitted to the target unmanned boat in a timely manner, while ensuring the security and reliability of data transmission.

[0102] S42. The unmanned boat receives the control command, automatically adjusts the navigation speed and communication frequency according to its own energy status, mission urgency and environmental conditions, and performs corresponding control actions based on the adjusted navigation speed and communication frequency.

[0103] Specifically, in order to automatically adjust the navigation speed and communication frequency according to the energy status of the unmanned boat, the urgency of the mission and the environmental conditions, an adaptive control mechanism can be adopted. The following are the specific steps:

[0104] 1) Real-time monitoring of the status of the unmanned boat

[0105] Before adjusting the navigation speed and communication frequency, it is necessary to first monitor the status of the unmanned boat in real time, including the following key aspects:

[0106] 1.1. Energy status monitoring: Detect the remaining power or fuel status of the unmanned boat, usually expressed as a percentage. For example, when the power is below a certain threshold, energy consumption needs to be reduced.

[0107] 1.2. Mission urgency monitoring: Evaluate the mission urgency based on the mission priority and deadline. Urgent missions require faster navigation speeds and more frequent communications to ensure rapid execution of missions.

[0108] 1.3. Environmental condition monitoring: Use sensors and environmental sensing equipment to detect current sea conditions, wind speed, weather and other factors. Harsh environments may limit navigation speed and affect communication quality.

[0109] 2) Adjust navigation speed and communication frequency according to energy status

[0110] 2.1 Energy state classification

[0111] According to the energy status, it can be divided into multiple levels, and different levels correspond to different adjustment strategies:

[0112] High energy state (>80%): The unmanned boat can perform tasks at a higher navigation speed and normal communication frequency.

[0113] Medium energy state (50%-80%): The unmanned boat maintains a moderate sailing speed and reduces the communication frequency to save energy.

[0114] Low energy state (<50%): The unmanned boat needs to significantly reduce its sailing speed and communication frequency to extend the use time of the remaining energy.

[0115] 2.2 Adjustment strategy

[0116] At high energy: maintain maximum speed (depending on mission requirements) and higher communication frequency (usually conventional communication frequency).

[0117] At medium energy: reduce speed and communication frequency to important mission nodes or as needed.

[0118] When energy is low: maintain the minimum safe speed and set the communication frequency to the minimum, communicating only when the mission or status changes.

[0119] 3) Adjust according to the urgency of the task

[0120] 3.1. Task urgency classification

[0121] Task urgency can be divided into three categories:

[0122] Emergency missions: such as rescue and time-sensitive missions. Requires maximum navigation speed and real-time high-frequency communication.

[0123] Normal mission: Routine missions, such as cruising and monitoring. The speed and communication frequency are maintained at normal levels.

[0124] Non-urgent tasks: such as low-priority tasks, can reduce navigation speed and communication frequency to save resources.

[0125] 3.2 Adjustment strategy

[0126] In case of emergency: adjust to the highest sailing speed and implement real-time high-frequency communication (to ensure communication quality through a stronger communication link).

[0127] During normal missions: maintain normal speed and communication frequency.

[0128] In non-urgent tasks: reduce navigation speed and communication frequency to focus resources on other more urgent tasks.

[0129] 4) Adjust according to environmental conditions

[0130] 4.1 Classification of environmental conditions

[0131] Environmental factors mainly include sea conditions, wind speed, weather, etc. Based on these conditions, the speed and communication frequency can be adjusted as follows:

[0132] Good environment: no wind or light wind, calm sea surface, good communication conditions. The unmanned boat can maintain normal speed and communication frequency.

[0133] Medium environment: medium wind speed, normal sea conditions, and changeable weather. The unmanned boat needs to reduce the speed and communication frequency appropriately to avoid excessive energy consumption or signal interference.

[0134] Harsh environment: high wind speed, bad sea conditions, unstable communication signals. The unmanned boat needs to significantly reduce the speed and minimize the communication frequency to save energy and resources and prevent equipment damage.

[0135] 4.2 Adjustment strategy

[0136] In good environment: the unmanned boat is allowed to sail and communicate at normal speed.

[0137] In medium conditions: reduce speed and communication frequency to avoid signal interference and navigation risks.

[0138] In harsh environments: the unmanned boat needs to reduce to a safe speed, reduce the communication frequency to a minimum, and communicate only when necessary.

[0139] 5) Comprehensive Adjustment

[0140] 5.1 Comprehensive weight

[0141] Different weights are assigned to energy status, mission urgency, and environmental conditions, and based on these weights, the navigation speed and communication frequency can be dynamically adjusted through a weighted decision model.

[0142] 5.2. Adjustment formula

[0143] The calculation formula for sailing speed is:

[0144] V=V max ×(α v ×E+β v ×T+ω v ×C)

[0145] The calculation formula for the communication frequency is:

[0146] F=F max ×(α f ×E+β f ×T+ω f ×C)

[0147] Where, V represents the sailing speed; V max represents the maximum navigation speed; E represents the energy status score (high energy is 1, low energy is 0.5, and extremely low is 0.2); T represents the mission urgency score (urgent is 1, normal is 0.7, and non-urgent is 0.3); C represents the environmental condition score (good is 1, bad is 0.5, and extreme is 0.2); α v , β v ,ω v They represent the speed weights of energy state, task urgency and environmental conditions respectively. In this embodiment, α v , β v ,ω v The values ​​of are 0.4, 0.4, and 0.2 respectively; F represents the communication frequency; F max Indicates the maximum communication frequency; α f , β f ,ω f Respectively represent the frequency weights of energy state, task urgency and environmental conditions. In this embodiment, α f , β f ,ω f The values ​​are 0.4, 0.4, and 0.2 respectively.

[0148] According to another embodiment of the present invention, there is provided an unmanned boat swarm remote control system, the unmanned boat swarm remote control system comprising: a communication link establishment module, a matching connection module, a control instruction generation module and a control action execution module;

[0149] The communication link establishment module is used to establish a communication link between the control console, the remote control server and the unmanned boat based on the communication technology, wherein the control console includes a cluster control console and a single boat control console;

[0150] The matching and connecting module is used to match the cluster control console or the single boat control console with the unmanned boat through the remote control server according to the mission requirements, and establish the connection between the cluster control console or the single boat control console and the predetermined unmanned boat;

[0151] The control instruction generation module is used to divide the task into several subtasks by using a hierarchical planning method, and to assign the divided subtasks to corresponding unmanned boats by using a distributed algorithm based on game theory, and to generate corresponding control instructions;

[0152] The control action execution module is used for the cluster console or the single boat console to send control instructions to the corresponding unmanned boat through the remote control server, and the unmanned boat receives the control instructions and executes the corresponding control action.

[0153] In summary, with the help of the above technical solution of the present invention, the present invention realizes flexible and efficient control of the unmanned boat cluster through the wired or wireless network connection between the cluster console, the single boat console, the remote control server and the unmanned boat, and the use of the radio and (or) Beidou to establish a connection, thereby effectively improving work efficiency and task coordination. At the same time, through matching connections, one console can control one or more unmanned boats at the same time, greatly improving the control efficiency. In addition, the present invention introduces an unmanned boat relay service, selects a suitable relay unmanned boat through the unmanned boat cluster information integration service, and realizes the transmission of control commands and work data through a suitable route. The introduction of this service effectively solves the problem of being unable to directly connect to the remote control server, realizes flexible control of unmanned boats that cannot be directly connected to the remote control server, and improves the efficiency and reliability of unmanned boat remote control. In addition, the present invention divides the task into several subtasks by using a hierarchical planning method, and assigns the subtasks to the most suitable unmanned boat in combination with the graph neural network algorithm, which can maximize the use of the resources and capabilities of the unmanned boat cluster. The graph neural network model can automatically match the unmanned boat that is most suitable for performing a specific task according to the state of the unmanned boat, task requirements and environmental conditions, thereby improving the rationality and efficiency of task allocation and ensuring that the task can be completed quickly and accurately. In addition, the present invention can dynamically adjust the navigation speed and communication frequency of the unmanned boat according to the energy status of the unmanned boat, the urgency of the mission and the environmental conditions. According to the importance of the mission, the current environment and the remaining energy of the unmanned boat, the optimal navigation speed and communication frequency can be automatically selected, thereby optimizing resource utilization, extending the endurance of the unmanned boat, and ensuring the quality of mission completion. This mechanism ensures the smooth execution of the mission in a complex and changing environment, while extending the service life of the unmanned boat.

[0154] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A remote control method for an unmanned boat swarm, characterized in that: The remote control method of the unmanned boat cluster includes the following steps: S1. Based on communication technology, establish a communication link between the control console, remote control server and unmanned boat, wherein the control console includes a cluster control console and a single boat control console; S2. According to the mission requirements, the cluster control console or the single boat control console matches the unmanned boat through the remote control server to establish the connection between the cluster control console or the single boat control console and the predetermined unmanned boat; S3, divide the task into several subtasks through hierarchical planning method, and use graph neural network algorithm to assign the divided subtasks to the corresponding unmanned boats to generate corresponding control instructions; S4. The cluster control console or the single boat control console sends a control command to the corresponding unmanned boat through the remote control server. The unmanned boat receives the control command and executes the corresponding control action.

2. The method for remotely controlling a swarm of unmanned boats according to claim 1, characterized in that: The communication link between the control console, the remote control server and the unmanned boat is established based on the communication technology, including the following steps: S11, the cluster console and the single boat console are connected to a single or multiple remote control servers through a wired or wireless network to establish a communication link between the console and the server; S12. The unmanned boat establishes a connection with one or more remote control servers via a radio or satellite to establish a communication link between the unmanned boat and the remote control server.

3. The method for remotely controlling a swarm of unmanned boats according to claim 1, characterized in that: The task is divided into several subtasks by the hierarchical planning method, and the divided subtasks are assigned to the corresponding unmanned boats by the graph neural network algorithm to generate corresponding control instructions, including the following steps: S31. Clarify the overall task that the unmanned boat swarm needs to complete, analyze the specific requirements of the task, and divide the overall task into several subtasks using the hierarchical planning method; S32. Based on the graph neural network algorithm, the divided subtasks are assigned to the corresponding unmanned boats, and the corresponding control instructions are generated.

4. The method for remotely controlling a swarm of unmanned boats according to claim 3, characterized in that: The above-mentioned process of clarifying the overall task that the unmanned boat cluster needs to complete, analyzing the specific requirements of the task, and dividing the overall task into several subtasks in combination with the hierarchical planning method includes the following steps: S311. Clarify the overall tasks that the unmanned boat swarm needs to complete, and analyze the mission area, time limit, resource constraints and environmental conditions; S312, decomposing the overall task into several main tasks, and prioritizing the main tasks according to their importance and urgency; S313, subdividing the main task into specific subtasks, evaluating the resources and capabilities required for each subtask, and setting a specific execution plan for each subtask, wherein the execution plan includes the driving route, navigation speed and communication frequency.

5. The method for remotely controlling a swarm of unmanned boats according to claim 3, characterized in that: The method of allocating the divided subtasks to the corresponding unmanned boats based on the graph neural network algorithm and generating corresponding control instructions includes the following steps: S321, constructing a graph structure based on the unmanned boat cluster and the subtasks, wherein each node in the graph structure represents an unmanned boat and a subtask, and the edges between the nodes represent the association between the unmanned boat and the subtask; S322, initializing a feature vector for each node, and initializing the weight of each edge according to the suitability of the unmanned boat to perform the subtask; S323. In each layer of the graph convolutional network model, the unmanned boat and the task node propagate and update information according to the characteristics of the neighboring nodes, and calculate the fitness score of the unmanned boat to perform the subtask through the scoring function combined with the updated feature representation; S324, training a graph convolutional network model using historical allocation data of the unmanned boat and the subtasks to obtain a trained graph convolutional network model; S325. Based on the trained graph convolutional network model, combined with the features of the unmanned boat node and the task node, predict the fitness score of each unmanned boat to perform each subtask; S326. Select the best unmanned boat to perform each subtask according to the fitness score, and generate corresponding control instructions according to the task allocation result.

6. The method for remotely controlling a swarm of unmanned boats according to claim 5, characterized in that: The expression for information update is: The expression for fitness score is: S(i,j)=σ[W·(h i ||h j ||e ij )] In the formula, represents the feature representation of the unmanned boat node i in the k+1th layer; represents the feature representation of the unmanned boat node i in the kth layer; σ represents a nonlinear activation function; N(i) represents the neighbor node set of unmanned boat node i; d i ,d j They represent the degrees of the unmanned boat node i and the subtask node j respectively; W (k) represents the weight matrix of the kth layer; S(i,j) represents the fitness score; h i 、h j Represent the feature vectors of unmanned boat node i and subtask node j respectively; || represents the concatenation operation of feature vectors; e ij Represents the characteristics of the edge between the unmanned boat node i and the subtask node j.

7. The method for remotely controlling a swarm of unmanned boats according to claim 1, characterized in that: The cluster control console or the single boat control console sends a control instruction to the corresponding unmanned boat through the remote control server, and the unmanned boat receives the control instruction and executes the corresponding control action, including the following steps: S41, determining whether the remote control server is connected to the unmanned boat for communication. If so, the cluster control console or the single boat control console sends a control command to the corresponding unmanned boat through the remote control server. If not, the optimal relay unmanned boat is selected through the unmanned boat cluster information integration service, and the cluster control console or the single boat control console sends a control command to the corresponding unmanned boat through the remote control server and the optimal relay unmanned boat. S42. The unmanned boat receives the control command, automatically adjusts the navigation speed and communication frequency according to its own energy status, mission urgency and environmental conditions, and performs corresponding control actions based on the adjusted navigation speed and communication frequency.

8. The method for remotely controlling a swarm of unmanned boats according to claim 7, characterized in that: The method of selecting the optimal relay unmanned boat through the unmanned boat cluster information integration service includes the following steps: Obtain the location information and network status information of each unmanned boat in the cluster in real time, and monitor the mission load and mission priority of the unmanned boat in real time; Based on the location information of the unmanned boat, the candidate unmanned boats whose distance to the target unmanned boat meets the preset threshold are screened out, and the candidate unmanned boats are further screened according to the network status information, task load and task priority to obtain a potential unmanned boat set; Weights are assigned to the location, network status, task priority and task load of the unmanned boat respectively, and the comprehensive score of each potential unmanned boat in the potential unmanned boat set is calculated based on the weights, and the optimal relay unmanned boat is selected according to the comprehensive score.

9. The method for remotely controlling a swarm of unmanned boats according to claim 7, characterized in that: The calculation formula for sailing speed is: V=V max ×(a v ×E+β v ×T+ω v ×C) The calculation formula for the communication frequency is: F=F max ×(a f ×E+β f ×T+ω f ×C) Where, V represents the sailing speed; V max Indicates the maximum sailing speed; E represents the energy status score; T represents the task urgency score; C represents the environmental condition score; α v , β v ,ω v speed weights representing energy status, task urgency, and environmental conditions, respectively; F represents the communication frequency; F max Indicates the maximum communication frequency; α f , β f ,ω f The frequency weights represent energy status, task urgency and environmental conditions respectively.

10. An unmanned boat swarm remote control system, used to implement the steps of the unmanned boat swarm remote control method described in any one of claims 1 to 9, characterized in that: The unmanned boat cluster remote control system includes: a communication link establishment module, a matching connection module, a control instruction generation module and a control action execution module; The communication link establishment module is used to establish a communication link between the control console, the remote control server and the unmanned boat based on the communication technology, wherein the control console includes a cluster control console and a single boat control console; The matching and connecting module is used to match the cluster control console or the single boat control console with the unmanned boat through the remote control server according to the mission requirements, and establish the connection between the cluster control console or the single boat control console and the predetermined unmanned boat; The control instruction generation module is used to divide the task into several subtasks by using a hierarchical planning method, and to assign the divided subtasks to corresponding unmanned boats by using a distributed algorithm based on game theory, and to generate corresponding control instructions; The control action execution module is used for the cluster control console or the single boat control console to send control instructions to the corresponding unmanned boat through the remote control server, and the unmanned boat receives the control instructions and executes the corresponding control action.

Images