Intelligent unmanned ship cluster health management and intelligent operation and maintenance system and method

Abstract

The invention discloses an intelligent unmanned ship cluster health management and intelligent operation and maintenance system and method, and belongs to the technical field of intelligent operation and maintenance of an offshore unmanned system cluster, and the system comprises a data obtaining module which is used for obtaining the position data, the communication quality and the task progress of each node in an unmanned ship cluster; the cooperative network construction module is used for constructing a cooperative relationship network between the unmanned ships based on the position data, the communication quality and the task progress; the community detection module is used for performing community division on the cooperative relationship network among the unmanned ships to obtain sub-communities; and the anomaly detection module is used for judging health states of the sub-communities and identifying and positioning regions or nodes with abnormal cooperation intensity. The invention aims to analyze the cooperation strength among cluster members to carry out the abnormal monitoring of the whole cluster, thereby achieving the comprehensive evaluation of the health state of the cluster, and guaranteeing that the task failure of the whole cluster is not caused by the influence of part of naval vessels.

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance technology for marine unmanned system clusters, and more specifically to an intelligent unmanned surface vessel cluster health management and intelligent operation and maintenance system and method. Background Technology

[0002] In modern naval warfare, intelligent unmanned surface vessel (USV) swarm operations have become a crucial component of naval combat power. Through collaborative operations, information sharing, and autonomous decision-making, intelligent USV swarms can accomplish various missions at sea, such as patrol, surveillance, reconnaissance, and strike, making their tactical and strategic value self-evident. However, with the expansion of swarm size and the increase in mission complexity, higher demands are placed on USV health management and intelligent operation and maintenance technologies. First, intelligent USV swarms need to operate stably for extended periods in harsh marine environments, posing a severe challenge to the health status of each USV subsystem. If a USV malfunctions and leaves the swarm, it will directly impact the combat effectiveness of the entire swarm. Health management technology can monitor and assess the status of each USV in real time, providing early warnings of potential faults and thus improving the overall reliability of the swarm. Second, intelligent operation and maintenance technology can quickly locate fault points and provide optimal repair solutions when USVs require maintenance, reducing non-combat time. Combining big data analytics and artificial intelligence algorithms, intelligent operation and maintenance technology can provide data support for USV maintenance decisions, optimize maintenance processes, and improve maintenance efficiency. Furthermore, during prolonged, high-intensity combat missions, the operational effectiveness of intelligent unmanned surface vessel (USV) swarms heavily relies on maintaining their good health and timely operational and maintenance capabilities. Therefore, health management and intelligent operation and maintenance technologies can not only ensure the efficient operation of intelligent USV swarms but also significantly enhance their combat effectiveness and survivability.

[0003] While health management and intelligent operation and maintenance technologies play a crucial role in intelligent unmanned surface vessel (USV) swarm operations, numerous problems and shortcomings remain. The most prominent issue is the continued reliance on traditional anomaly monitoring and fault diagnosis techniques targeting individual USVs, failing to manage the swarm as a whole. In reality, with significant advancements in materials science and manufacturing processes, the probability of mechanical or electronic failures in individual USVs is extremely low, already sufficient for long-endurance, long-range autonomous missions. The real problem with USV swarms lies in the decreased mission efficiency and failures caused by swarm collaboration anomalies.

[0004] Therefore, how to analyze the collaboration strength among cluster members and carry out overall cluster anomaly monitoring is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of this, the present invention provides an intelligent unmanned surface vessel (USV) swarm health management and intelligent operation and maintenance system and method, which aims to analyze the collaboration intensity among swarm members to carry out anomaly monitoring of the swarm as a whole, thereby achieving a comprehensive assessment of the swarm's health status and ensuring that the entire swarm will not fail due to the influence of some ships.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A smart unmanned surface vessel (USV) swarm health management and intelligent operation and maintenance system includes:

[0008] The data acquisition module is used to acquire the location data, communication quality, and mission progress of each node in the unmanned surface vessel cluster;

[0009] The collaborative network construction module is used to build a collaborative relationship network among unmanned surface vessels based on location data, communication quality, and mission progress.

[0010] The community detection module is used to divide the network of cooperative relationships among unmanned surface vessels into communities, thus obtaining sub-communities;

[0011] The anomaly detection module is used to determine the health status of sub-communities and identify and locate areas and nodes with abnormal collaboration intensity.

[0012] Preferably, the specific implementation process of the collaborative network construction module is as follows:

[0013] Based on communication quality, determine whether there is a cooperative relationship between nodes, construct a connection edge for two nodes that have a cooperative relationship, and build a cooperative relationship network between unmanned surface vessels;

[0014] Calculate the task progress similarity of edges based on task progress;

[0015] The communication signal strength of the edge is calculated based on the communication quality.

[0016] Calculate the proximity of edges using location data;

[0017] The comprehensive edge weight is calculated based on the similarity of task progress, the strength of communication signals, and the proximity of locations. The comprehensive edge weight is used to quantify the degree of cooperation between different nodes in the collaborative network relationship.

[0018] Preferably, the formula for calculating task progress similarity is:

[0019]

[0020] in, and unmanned surface vessel nodes and unmanned surface vessel nodes Task progress, This is a collection of progress data for all node tasks. The weights are the similarity scores of the task progress.

[0021] The formula for calculating communication signal strength is:

[0022] ;

[0023] in, For unmanned surface vessel nodes and unmanned surface vessel nodes Communication quality, and These are the maximum and minimum values ​​in the communication quality data set among all nodes. For communication signal strength weights;

[0024] The formula for calculating location proximity is:

[0025] ;

[0026] in, For location scale parameters, For unmanned surface vessel nodes Location data, For unmanned surface vessel nodes Location data, The weight is the proximity of the location;

[0027] The formula for calculating the overall edge weight is:

[0028] ;

[0029] in, , and These are weighting coefficients; the sum of the three must be 1. This is for the comprehensive edge weights.

[0030] Preferably, the specific implementation process of the community detection module is as follows:

[0031] Step 1: Treat each unmanned surface vessel node in the collaborative relationship network as a community and calculate the initial value of modularity;

[0032] Step 2: Calculate the modularity gain of the unmanned surface vessel node after it moves from the current community to the target community;

[0033] Step 3: Determine if the modularity gain is greater than 0. If it is, update the community tag to which the unmanned surface vessel node belongs; otherwise, do not update.

[0034] Step 4: After the update, treat each community as a supernode, calculate the weighted adjacency matrix of the supernode network, and iterate through Step 2 and Step 3 until the modularity converges to obtain the sub-community.

[0035] Preferably, the formula for calculating the initial value of modularity is:

[0036] ;

[0037] in, , For unmanned surface vessel nodes The degree, This is the initial value for modularity. For indicator functions, when the community and community The indicator function is 1 if the user belongs to a community, and 0 otherwise.

[0038] The formula for calculating modularity gain is:

[0039] ;

[0040] in, For unmanned surface vessel nodes The current boundaries and rights within the community, For unmanned surface vessel nodes Boundary rights and interests with the target community For the target community Total degree, This represents the modularity gain.

[0041] The preferred formula for calculating the weighted adjacency matrix is:

[0042] ;

[0043] in, For a weighted adjacency matrix, Represents a node and The combined edge weights between them.

[0044] Preferably, the specific implementation process of the anomaly detection module is as follows:

[0045] Calculate the intensity of cooperation among sub-communities;

[0046] Determining whether a sub-community is abnormal based on the strength of sub-community collaboration:

[0047] like If the condition is met, the community is deemed abnormal, and T represents the abnormality threshold.

[0048] The preferred formula for calculating the sub-community collaboration intensity is:

[0049] ;

[0050] in, To enhance the intensity of cooperation among sub-communities, For community tags, This represents the number of nodes within the community.

[0051] Preferably, the existence of a cooperative relationship between nodes is determined based on communication quality. The specific determination method is as follows:

[0052] Determine the strength of communication signals between nodes Is it not lower than the strength threshold? If the conditions are met, then the unmanned surface vessel node... and nodes They have a collaborative relationship.

[0053] A method for health management and intelligent operation and maintenance of intelligent unmanned surface vessel (USV) swarms includes:

[0054] Acquire the location data, communication quality, and mission progress of each node in the unmanned surface vessel (USV) swarm.

[0055] A collaborative network among unmanned surface vessels is constructed based on location data, communication quality, and mission progress.

[0056] The network of collaborative relationships among unmanned surface vessels is divided into communities to obtain sub-communities;

[0057] Determine the health status of sub-communities and identify and locate areas and nodes with abnormal collaboration intensity.

[0058] As can be seen from the above technical solutions, compared with the prior art, the present invention discloses an intelligent unmanned surface vessel (USV) swarm health management and intelligent operation and maintenance system and method, which can conduct health status assessment based on the differences of swarm nodes under similar working conditions, thereby ensuring that the entire swarm will not fail due to the influence of individual vessels. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0060] Figure 1 This invention provides a structural block diagram of an intelligent unmanned surface vessel (USV) cluster health management and intelligent operation and maintenance system.

[0061] Figure 2 The present invention provides a flowchart of a method for health management and intelligent operation and maintenance of an intelligent unmanned surface vessel (USV) swarm. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] This invention discloses an intelligent unmanned surface vessel (USV) swarm health management and intelligent operation and maintenance system, such as... Figure 1 As shown, it includes:

[0064] The data acquisition module is used to acquire the location data, communication quality, and mission progress of each node in the unmanned surface vessel cluster;

[0065] The collaborative network construction module is used to build a collaborative relationship network among unmanned surface vessels based on location data, communication quality, and mission progress; by adjusting the weight coefficients, it ensures that the impact of different features on the strength of collaboration is reasonably quantified.

[0066] The community detection module is used to divide the cooperative relationship network among unmanned surface vessels into communities based on the Louvain algorithm, and obtain sub-communities.

[0067] The anomaly detection module is used to determine the health status of sub-communities and identify and locate areas or nodes with abnormal collaboration intensity.

[0068] As a preferred embodiment of the present invention, the data acquisition module acquires the following data from each cluster node through the communication link with the unmanned surface vessel cluster node.

[0069] Table 1 Data for each cluster node

[0070]

[0071] As a preferred embodiment of the present invention, the cooperative network construction module constructs a cooperative relationship network among unmanned surface vessels according to the following steps:

[0072] 1) Determining the existence of cooperative relationships between nodes based on communication quality. This invention uses the communication signal strength between nodes. Is it not lower than the strength threshold? Determine unmanned surface vessel nodes and Do they have a collaborative relationship? If the unmanned surface vessel nodes... and nodes If a cooperative relationship exists, a connecting line (edge) is constructed between the two nodes, ultimately forming a cooperative relationship network. Then, the following steps are performed on the edges in the cooperative relationship network.

[0073] 2) Calculate the task progress similarity weight using the following formula.

[0074] ;

[0075] in, and unmanned surface vessel nodes and unmanned surface vessel nodes Task progress, This is a collection of progress data for all node tasks. The weights represent the similarity of task progress.

[0076] 3) Calculate the strength weight of the communication signal using the following formula:

[0077] ;

[0078] in, For unmanned surface vessel nodes and unmanned surface vessel nodes Communication quality, and These are the maximum and minimum values ​​in the communication quality data set among all nodes. This represents the signal strength weights for communication.

[0079] 4) Calculate the proximity weight using the following formula:

[0080]

[0081] in, The location scale parameter determines the sensitivity of distance to the assessment of cooperation intensity. Its selection needs to be considered in conjunction with the distribution characteristics of the unmanned surface vessel (USV) swarm and mission requirements in the specific application scenario. Typically, for small swarms (relatively densely distributed, with a range of less than 1 kilometer), [the parameter is used]. For 50 to 100 meters; medium-sized clusters (covering 1 to 10 kilometers) For distances of 500 to 1000 meters; large clusters (covering 10 kilometers) The distance is 2 to 5 kilometers.

[0082] 5) Calculate the combined edge weights using the following formula:

[0083] ;

[0084] in, , and These are weighting coefficients, and the sum of the three must be 1.

[0085] As a preferred embodiment of the present invention, the community detection module performs the following steps to divide the cooperative relationship network into several sub-communities, thereby characterizing the distribution characteristics of the cooperative relationships between unmanned surface vessel nodes:

[0086] Step 1: Initialize nodes and calculate initial modularity values.

[0087] The collaborative relationship network of the intelligent unmanned surface vessel (USV) swarm is constructed based on the collaborative network building module. Each node in the network is initially treated as a separate community (each community is represented as...). That is, the initial number of communities equals the number of nodes. The modularity is calculated using the following formula. The initial value is recorded, and the maximum value is also recorded. ;

[0088]

[0089] in , For unmanned surface vessel nodes The degree, This is the initial value for modularity. For indicator functions, when the community and community The indicator function has a value of 1 if it belongs to a community, and 0 otherwise.

[0090] Step 2: Calculate the modularity gain.

[0091] Modularity gain Indicates the unmanned surface vessel node From the current community Move to target community The subsequent change in modularity is calculated using the following formula:

[0092]

[0093] in, For unmanned surface vessel nodes The current boundaries and rights within the community, For unmanned surface vessel nodes Boundary rights and interests with the target community For the target community Total degree, This represents the modularity gain.

[0094] For each node To perform local optimization, the specific process is to first traverse the communities to which the neighboring nodes belong. Then calculate the nodes Move to Modularity gain ;

[0095] Step 3: Determine if the modularity gain is greater than 0. Then the node Move to and update its community tags (i.e. ).

[0096] Step 4: Perform community merging. Treat each community as a supernode, and construct the weighted adjacency matrix of the supernode network using the following formula. Based on this, treat the supernode as a separate community again, and repeat steps 2 and 3 until the modularity converges, i.e., reaches its maximum value. It will no longer change.

[0097]

[0098] That is, the community detection process ends when the modularity no longer increases or reaches the set maximum number of iterations, and the community division label and modularity are output. .

[0099] As a preferred embodiment of the present invention, the anomaly detection module performs the following steps to identify abnormal regions or nodes:

[0100] Step 1: Evaluate the intensity of sub-community collaboration using the following formula:

[0101]

[0102] in, To enhance the intensity of cooperation among sub-communities, For community tags, This represents the number of nodes within the community.

[0103] Step 2: Identify abnormal communities.

[0104] This step will determine abnormal communities based on an anomaly threshold T. If the community is deemed abnormal, it indicates that the nodes within the sub-community are exhibiting an abnormal health status in terms of cooperation strength, thus requiring further scheduling and dispatch.

[0105] The invention will be further explained below with reference to specific data:

[0106] A smart unmanned surface vessel (USV) swarm health management and intelligent operation and maintenance system includes a data acquisition module, a collaborative network construction module, a community detection module, and an anomaly detection module.

[0107] The data acquisition module utilizes the communication link between the intelligent unmanned surface vessel (USV) cluster health management and intelligent operation and maintenance system and the cluster nodes to acquire the following data for each USV node in a cluster.

[0108] Table 2 Location Data and Task Progress Data

[0109]

[0110] Table 3 Communication signal strength data

[0111]

[0112] The collaborative network building module will perform the following steps to evaluate the health status of each node in the cluster. First, The value is -85dBm, used to determine the node. and Whether there is a cooperative relationship between them is determined as follows (the row and column numbers of matrix A from left to right and from top to bottom are 1 to 10 respectively).

[0113]

[0114]

[0115] Next, for each node, the similarity weights for task progress, the strength weights for communication signals, and the proximity weights are calculated. These include relevant parameters. It is 100; , and The values ​​are 0.4, 0.4, and 0.2. The final combined edge weight calculation results are as follows:

[0116] Table 4. Statistics of Comprehensive Edge Weights

[0117]

[0118] Based on the above results, the community detection module performs relevant steps to divide the collaborative network into communities. Ultimately, the module degree Q converges to 0.567, and the community labeling results are as follows: nodes 1, 2, 3, and 4 are labeled as community 1; nodes 5, 6, and 7 are labeled as community 2; and nodes 8, 9, and 10 are labeled as community 3.

[0119] Subsequently, the anomaly detection module will assess the collaboration strength of communities based on the community division results and determine anomalies. Through the execution of relevant steps, the following collaboration strength calculation results are obtained. Based on historical data, the anomaly threshold T is 0.3. Therefore, the collaboration strength of community 3 is abnormal; hence, the health status of nodes 8, 9, and 10 is determined to be abnormal, requiring node rescheduling.

[0120] Table 5. Results of Cooperation Intensity Assessment

[0121]

[0122] This invention discloses a method for health management and intelligent operation and maintenance of intelligent unmanned surface vessel (USV) swarms, such as... Figure 2 As shown, it includes:

[0123] Acquire the location data, communication quality, and mission progress of each node in the unmanned surface vessel (USV) swarm.

[0124] A collaborative network among unmanned surface vessels is constructed based on location data, communication quality, and mission progress.

[0125] The network of collaborative relationships among unmanned surface vessels is divided into communities to obtain sub-communities;

[0126] Determine the health status of sub-communities and identify and locate areas or nodes with abnormal collaboration intensity.

[0127] The specific implementation process of each step in the method part of this invention is the same as that in the system part, and will not be repeated here.

[0128] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0129] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent unmanned vehicle cluster health management and intelligent operation and maintenance system, characterized in that, The application relates to an unmanned ship cluster cooperation network health state detection method, which comprises the following steps: a data acquisition module is used for acquiring position data, communication quality and task progress of each node in an unmanned ship cluster; a cooperation network construction module is used for constructing a cooperation relationship network among unmanned ships based on the position data, the communication quality and the task progress; a community detection module is used for community division of the cooperation relationship network among the unmanned ships to obtain sub-communities; an anomaly detection module is used for judging the health state of the sub-communities, identifying and positioning areas and nodes with abnormal cooperation strength. 2.The intelligent unmanned vehicle cluster health management and intelligent operation and maintenance system of claim 1, wherein, The cooperation network construction module specifically realizes the process as follows: whether there is a cooperation relationship between nodes is judged based on the communication quality, edges of two nodes with the cooperation relationship are constructed, and the cooperation relationship network among the unmanned ships is constructed; task progress similarity of the edges is calculated based on the task progress; communication signal strength of the edges is calculated based on the communication quality; position proximity of the edges is calculated through the position data; comprehensive edge weight is calculated based on the task progress similarity, the communication signal strength and the position proximity, and the cooperation degree between different nodes in the cooperation network relationship is quantified through the comprehensive edge weight. 3.The intelligent unmanned vehicle cluster health management and intelligent operation and maintenance system of claim 2, wherein, The task progress similarity calculation formula is as follows: ； wherein, and are unmanned vehicle nodes and unmanned vehicle nodes task progress, is a collection of all node task progress data, is a task progress similarity weight; the communication signal strength calculation formula is as follows: ； wherein, is an unmanned boat node and an unmanned boat node a communication quality, and is a maximum and a minimum of a set of communication quality data between all nodes, is a communication signal strength weight; the position proximity calculation formula is as follows: ； wherein, is a position scale parameter, is position data of the USV node , is position data of the USV node , is a position proximity weight; the comprehensive edge weight calculation formula is as follows: ； wherein, , and are weight coefficients, the sum of which must be 1, is the integrated edge weight.

4. The intelligent unmanned vehicle cluster health management and intelligent operation and maintenance system according to claim 3, characterized in that, The community detection module specifically realizes the process as follows: step 1: each unmanned ship node in the cooperation relationship network is taken as a community, and an initial value of a modularity is calculated; step 2: a modularity gain after the unmanned ship node is moved from a current community to a target community is calculated; step 3: whether the modularity gain is greater than 0 is judged, if yes, the community label of the unmanned ship node is updated, otherwise, the update is not performed; step 4: after the update, each community is taken as a super node, a weighted adjacency matrix of the super node network is calculated, and steps 2 and 3 are iterated until the modularity converges, and the sub-communities are obtained.

5. The intelligent unmanned vehicle cluster health management and intelligent operation and maintenance system according to claim 4, characterized in that, The initial value of the modularity calculation formula is as follows: ； wherein, , is the degree of the unmanned boat node , is the initial value of the modularity, is an indicator function, which has a value of 1 when a community and a community belongs to the same community, and 0 otherwise; the modularity gain calculation formula is as follows: ； wherein, is the sum of edge weights within the current community, is the sum of edge weights between the current community and the target community, is the sum of edge weights within the target community, is the sum of edge weights between the current community and the target community, is the target community, is the total degree of the target community, is the modularity gain. 6.The intelligent unmanned vehicle cluster health management and intelligent operation and maintenance system of claim 4, wherein, the weighted adjacency matrix calculation formula is as follows: ； wherein, is a weighted adjacency matrix, denotes the integrated edge weight between nodes and .

7. The intelligent unmanned vehicle cluster health management and intelligent operation and maintenance system according to claim 5, characterized in that, The anomaly detection module specifically realizes the process as follows: sub-community cooperation strength is calculated; whether the sub-community is abnormal is judged based on the sub-community cooperation strength: If then the community is determined to be abnormal, T denotes an abnormality threshold.

8. The intelligent unmanned vehicle cluster health management and intelligent operation and maintenance system according to claim 7, characterized in that, the sub-community cooperation strength calculation formula is as follows: ； wherein, is the sub-community collaboration strength, is the community tag, is the number of nodes within the community. 9.The intelligent unmanned vehicle cluster health management and intelligent operation and maintenance system of claim 2, wherein, whether there is a cooperation relationship between nodes is judged based on the communication quality, and the specific judgment method is as follows: Determining inter-node communication signal strength Whether not lower than intensity threshold If satisfied, unmanned ship node And node Have a cooperative relationship.

10. An intelligent unmanned ship cluster health management and intelligent operation and maintenance method, characterized in that, The application further discloses an unmanned ship cluster cooperation network health state detection device, which comprises the following modules: a data acquisition module is used for acquiring position data, communication quality and task progress of each node in an unmanned ship cluster; a cooperation network construction module is used for constructing a cooperation relationship network among unmanned ships based on the position data, the communication quality and the task progress; a community detection module is used for community division of the cooperation relationship network among the unmanned ships to obtain sub-communities; a judgment module is used for judging the health state of the sub-communities, identifying and positioning areas and nodes with abnormal cooperation strength.

Images