Decentralized networking method and system for digital twin unmanned site

Abstract

The application relates to the technical field of site networking, and particularly discloses a digital twin unmanned site decentralized networking method and system. After being modeled by digital twinning, the unmanned site serves as a self-organizing network node, forms a decentralized network, and is initialized to elect a leader node. After the leader node publishes a task, network nodes vote through a multi-dimensional verification and consensus mechanism. After the task is claimed, a message is written into a metadata extension field and is synchronized across the network, reducing multi-source heterogeneous data islands and repeated claims. Based on the task position and the basic environmental data of the node geographic grid, an optimal routing cost function under multi-dimensional constraints is constructed, the nearest neighbor node is determined through the minimum cost value, the lowest delay connection graph is obtained through spatial analysis, and the optimal path is determined, so that the multi-unmanned system device executes the task according to the optimal path, the task scheduling and resource coordination accuracy are improved, the decision support capability under comprehensive situation awareness is improved, and the multi-device collaborative efficiency and task dispatching accuracy are further improved.

Description

Technical Field

[0001] This invention relates to the field of site networking technology, and in particular to a decentralized networking method and system for digital twin unmanned sites. Background Technology

[0002] With the rapid development of intelligent manufacturing, smart cities, and unmanned operations, unmanned systems (such as drones, unmanned vehicles, unmanned ships, robots, and robotic dogs) are increasingly being used in logistics, inspection, security, agriculture, and other fields. Furthermore, with the rapid development of digital twin technology, it is gradually extending from industrial manufacturing and traffic management to unmanned systems, achieving state monitoring and prediction by constructing virtual mappings of physical entities.

[0003] Currently, a search reveals patent document (CN113284341B) entitled "An Intelligent Linkage System for Unmanned Aerial Vehicles Based on Road Internet of Things Monitoring." This invention utilizes an intelligent linkage system for unmanned aerial vehicles (UAVs) based on road Internet of Things monitoring. This intelligent linkage system includes: intelligent road sensing, a road-use acceleration sensing system, an intelligent microprocessor, wireless communication equipment, a road monitoring camera system, an intelligent gateway, a big data processing module, a central management platform, a dispatching and scheduling system for personnel, and unmanned aerial vehicles (UAVs). The intelligent road sensing system comprises multiple road-use acceleration sensing nodes embedded beneath the road surface. The road-use acceleration sensing system acquires monitoring information from these nodes, enabling rapid and continuous monitoring of road surface vibrations. This achieves intelligent linkage between the UAVs and the Internet of Things sensing information, ensuring comprehensive and timely reporting of accident scenes.

[0004] However, practice has shown that this patent document merely maps the objective world to the digital world, resulting in low accuracy in task scheduling and resource coordination among multiple unmanned system devices. This, in turn, limits decision support capabilities under comprehensive situational awareness, and leads to low system autonomy, reliability, and collaborative efficiency. Therefore, there is an urgent need to propose a new technical solution that can improve the accuracy of task scheduling and resource coordination among multiple unmanned system devices, thereby enhancing decision support capabilities under comprehensive situational awareness, and ultimately improving the system's autonomy, reliability, and collaborative efficiency. Summary of the Invention

[0005] This invention provides a decentralized networking method and system for digital twin unmanned sites, which can improve the accuracy of task scheduling and resource coordination among multiple unmanned system devices, thereby improving decision support capabilities under comprehensive situational awareness, and further improving the system's autonomy, reliability and collaborative efficiency.

[0006] To address the aforementioned technical problems, the first aspect of this invention discloses a decentralized networking method for digital twin unmanned sites, the method comprising:

[0007] After obtaining the corresponding digital twin site model by performing digital twin modeling on the unmanned site, the digital twin site model is determined as the node of the self-organizing network, and all the nodes form a network; and an initialization operation is performed on the network to obtain a distributed architecture network, wherein the leader node of the network is elected during the initialization process;

[0008] The leader node publishes the current task to the network, wherein, upon receiving the current task, the network nodes perform multi-dimensional verification of the current task and execute a voting operation according to consensus rules to obtain a current node that claims the current task; wherein, after claiming the current task, the current node writes the message that the current task has been claimed into the extended field of the current node's metadata and synchronizes it to all other nodes in the network; wherein, the current node includes the leader node or other nodes in the network;

[0009] The system determines that the location of the current task belongs to a geographic grid within the radiation range of a node in the network. Based on the basic environmental data between the geographic grid and the geographic grid of the current node, it constructs an optimal routing cost function under multi-dimensional constraints. Based on the minimum cost value of the optimal routing cost function, it determines the nearest neighbor node of the current node under the multi-dimensional constraints. Through spatial analysis of the nearest neighbor node, it dynamically constructs a minimum latency connectivity graph of the current node. Based on the minimum latency connectivity graph, it determines the optimal path of the current node, wherein the optimal path of the current node is used as the basis for the current node to execute the current task.

[0010] As an optional implementation, in the first aspect of the present invention, after digital twin modeling, both the unmanned station and the current task are expressed as geographic entities, wherein each geographic entity corresponding to the unmanned station has a globally unique identifier, a geohash code and metadata, wherein the metadata corresponding to each unmanned station has extensible attributes;

[0011] The metadata corresponding to each unmanned station is stored in a distributed manner in the network and is tamper-proof;

[0012] The extended fields of the metadata of the current node are used to record the resources of the current node, wherein the resources of the current node include one or more of computing power, storage, load, carrier, and other schedulable nodes. The extended fields of the metadata of the current node are also used to record the historical service parameters of the current node, wherein the historical service parameters of the current node include the historical service quality score and the number of historical services.

[0013] The extended fields of the metadata of the current task are used to record the resources required, including computing power, storage, payload, carrier, and other nodes that need to be scheduled.

[0014] As an optional implementation, in the first aspect of the present invention, the method further includes:

[0015] After the current node claims the current task, it determines whether subtask division is necessary based on the resources it can provide and the resources required by the current task.

[0016] When it is determined that subtasks need to be divided, the current task is recursively decomposed into subtasks through the current node to obtain one or more indivisible atomic tasks.

[0017] Based on the resources required for each atomic task and the resources that other nodes in the network can provide, nodes in the network that have the ability to complete the atomic task are selected as candidate nodes. According to the consensus rules of the atomic task, the current node that claims the atomic task is determined from all the candidate nodes. After claiming the atomic task, the current node writes the message that the atomic task has been claimed into the extended field of the current node's metadata and synchronizes it to all other nodes in the network.

[0018] For any of the atomic tasks, update the atomic task to the current task and trigger the operation of determining that the location of the current task belongs to the geographic grid within the radiation range of the nodes in the network.

[0019] As an optional implementation, in the first aspect of the present invention, the geographic grid within the radiation range of the nodes in the network where the current task is located is obtained in the following way:

[0020] Determine that the current task's location belongs to a node in the network, and obtain the geographic grid within the node's radiation range from the digital twin model corresponding to that node;

[0021] The geographic grid of the current node is obtained in the following way:

[0022] Determine the digital twin model corresponding to the current node, and obtain the geographic grid of the current node from the digital twin model;

[0023] The multi-dimensional verification includes at least two of the following: task reachability verification, task legality verification, and task priority verification. The task reachability includes one or more of the following: avoiding or overcoming obstacles along the route, weather conditions meeting the deployment requirements, and the carrier's energy support to reach the destination.

[0024] The legitimacy of the mission includes one or more of the following: having airspace resources, having an approved mission plan, and the mission path not conflicting with prohibited / restricted areas.

[0025] The task priority includes one or more of the following: priority determined by different task types, priority determined by time sequence in the queue of tasks to be executed, and priority determined by site task traffic optimization strategies.

[0026] As an optional implementation, in the first aspect of the present invention, the method further includes:

[0027] When it is detected that the current conditions of the network meet the predetermined reconstruction conditions, the reconstruction type corresponding to the network is determined;

[0028] Based on the reconstruction type corresponding to the network, a reconstruction operation matching the reconstruction type is performed on the network, wherein the reconstructed network is used as the basis for the reconstruction of the task to be executed;

[0029] The step of performing a reconstruction operation on the network that matches the reconstruction type according to the network's corresponding reconstruction type includes:

[0030] When the network's corresponding reconstruction type includes a node reconstruction type, a reconstruction operation is performed on the network based on the globally unique identifiers and / or geohash codes of other schedulable nodes recorded in the extended fields of their metadata for nodes that are currently online. When a node in the network changes, other nodes currently online perceive the change based on the globally unique identifiers and / or geohash codes of the node whose status has changed, obtaining a change perception result. Specifically, when the node whose status has changed is an offline node or a faulty node in the network, the change perception result is an offline perception result or a fault perception result; when the node whose status has changed is an online node in the network, the change perception result is an online perception result.

[0031] When the reconstruction type corresponding to the network includes the resource reconstruction type, the task execution status of the current node in the network is obtained, and the resource reconstruction operation of the network is performed according to the task execution status of the current node.

[0032] As an optional implementation, in the first aspect of the present invention, the method further includes:

[0033] When the node whose state has changed is the current node, it is determined whether the current node is the leader node. If it is determined that the current node is not the leader node, based on the leader node and according to the current node's execution status of the current task, the current task is re-published to the network so that the current node in the network can claim the current task.

[0034] When the leader node is identified, other leader nodes that are online and match the current task are determined from the network based on the task execution status of the current task and the geographical grid of the leader node in the digital twin model corresponding to the network.

[0035] The other leader nodes are updated to become the leader node, and the operation of publishing the current task to the network is re-executed so that the current node in the network can claim the current task.

[0036] As an optional implementation, in the first aspect of the invention, before determining the optimal path of the current node based on the lowest latency connectivity graph, the method further includes:

[0037] Determine the number of the lowest latency connection graphs. When the number is equal to 1, perform the operation of determining the optimal path of the current node based on the lowest latency connection graphs.

[0038] When the number is greater than or equal to 2, obtain the historical quality of service score of each intermediate routing node in the lowest latency connection graph, and determine the reputation value of the intermediate routing node based on the historical quality of service score of each intermediate routing node; based on the reputation values ​​of all intermediate routing nodes, select the lowest latency connection graph corresponding to the intermediate routing node with the highest reputation value from all intermediate routing nodes, and perform the operation of determining the optimal path of the current node based on the lowest latency connection graph;

[0039] The method further includes:

[0040] When the reputation values ​​of all intermediate routing nodes are equal, the historical service frequency of each intermediate routing node is calculated, and based on the historical service frequency of all intermediate routing nodes, the lowest latency connection graph corresponding to the intermediate routing node with the lowest historical service frequency is selected.

[0041] The step of determining the optimal path for the current node based on the lowest latency connectivity graph includes:

[0042] The optimal path for the current node is determined based on the lowest latency connection graph corresponding to the intermediate routing node with the lowest historical service frequency.

[0043] A second aspect of this invention discloses a decentralized networking system for digital twin unmanned sites, the system comprising:

[0044] The networking module is used to determine the digital twin site model as a node of the self-organizing network after obtaining the corresponding digital twin site model by performing digital twin modeling on the unmanned site, and all the nodes form a network; and to perform an initialization operation on the network to obtain a distributed architecture network, wherein the leader node of the network is elected during the initialization process;

[0045] The task claiming module is used to publish the current task to the network through the leader node. Upon receiving the current task, the network nodes perform multi-dimensional verification and execute a voting operation according to consensus rules to determine the current node that claims the task. After claiming the task, the current node writes the message that it has claimed the task into an extended field of its metadata and synchronizes it to all other nodes in the network. The current node includes the leader node or other nodes in the network.

[0046] The routing module is used to determine that the location of the current task belongs to a geographic grid within the radiation range of a node in the network, and to construct an optimal routing cost function under multi-dimensional constraints based on the basic environmental data between the geographic grid and the geographic grid of the current node. Based on the minimum cost value of the optimal routing cost function, it determines the nearest neighbor node of the current node under the multi-dimensional constraints, and dynamically constructs a minimum latency connectivity graph of the current node through spatial analysis of the nearest neighbor node. Finally, it determines the optimal path of the current node based on the minimum latency connectivity graph, wherein the optimal path of the current node serves as the basis for the current node to execute the current task.

[0047] A third aspect of this invention discloses another decentralized networking system for digital twin unmanned sites, the system comprising:

[0048] Memory containing executable program code;

[0049] A processor coupled to the memory;

[0050] The processor calls the executable program code stored in the memory to execute some or all of the steps in any of the decentralized networking methods for digital twin unmanned sites disclosed in the first aspect of the present invention.

[0051] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0052] In this embodiment of the invention, after obtaining the corresponding digital twin model by performing digital twin modeling on unmanned sites, all the digital twin models of unmanned sites serve as nodes in a self-organizing network, forming a decentralized network. The network is initialized to form a distributed structure, and a leader node is elected. When a task needs to be executed, the leader node publishes the task to the network, enabling nodes to perform multi-dimensional verification of the task and vote based on a consensus mechanism. This ensures that the task is claimed by the current node in the network. After claiming the task, the current node writes the task claim message into an extended field of its metadata and synchronizes it to the network, allowing other nodes to be aware of the message and understand the task claim status. This reduces the risk of multi-source heterogeneous data silos and tasks being repeatedly claimed by other nodes in the network, thus ensuring that when other... When a task is in progress or changes, other nodes can coordinate in a timely and accurate manner. Furthermore, based on the basic environmental data between the geographic grid within the radiation range of nodes in the network where the current task is located and the geographic grid of the current node, an optimal routing cost function under multi-dimensional constraints is constructed. Then, based on the minimum cost of the optimal routing cost function, the nearest neighbor node of the current node is determined. Through spatial analysis of the nearest neighbor node, the lowest latency connection graph of the current node is obtained, thereby determining the optimal path. This allows multiple unmanned system devices to execute the current task according to the optimal path, improving the accuracy of task scheduling and resource coordination among multiple unmanned system devices. This, in turn, enhances decision support capabilities under comprehensive situational awareness, improves the collaborative efficiency and accuracy among multiple unmanned devices in the site, improves the control accuracy and efficiency of task collaboration, and ultimately improves the efficiency and accuracy of task deployment. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a flowchart illustrating a decentralized networking method for digital twin unmanned sites disclosed in an embodiment of the present invention;

[0055] Figure 2 This is a flowchart illustrating another decentralized networking method for digital twin unmanned sites disclosed in an embodiment of the present invention;

[0056] Figure 3 This is a network diagram of a forestry and grassland fire protection system for an unattended site disclosed in an embodiment of the present invention;

[0057] Figure 4This is a network diagram of a forestry and grassland fire protection system for an unattended site disclosed in an embodiment of the present invention;

[0058] Figure 5 This is a schematic diagram of an intelligent router disclosed in an embodiment of the present invention;

[0059] Figure 6 This is a schematic diagram of the structure of a decentralized networking system for a digital twin unmanned site disclosed in an embodiment of the present invention;

[0060] Figure 7 This is a schematic diagram of another decentralized networking system for digital twin unmanned sites disclosed in an embodiment of the present invention;

[0061] Figure 8 This is a schematic diagram of the structure of another decentralized networking system for digital twin unmanned sites disclosed in an embodiment of the present invention. Detailed Implementation

[0062] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0064] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0065] This invention discloses a decentralized networking method and system for unmanned digital twin sites. After obtaining the corresponding digital twin model by digital twin modeling the unmanned site, all the digital twin models of the unmanned sites serve as nodes in a self-organizing network, forming a decentralized network. The network is initialized to form a distributed structure, and a leader node is elected. When a task needs to be executed, the leader node publishes the task to the network, enabling nodes in the network to perform multi-dimensional verification of the task and vote based on a consensus mechanism. This allows the task to be claimed by the current node in the network. After claiming the task, the current node writes the task claim message into an extended field of its metadata and synchronizes it to the network, so that other nodes in the network are aware of the message and know the task claim status. This reduces the risk of multi-source heterogeneous data silos and tasks being repeatedly claimed by other nodes in the network. This approach aims to ensure timely and accurate coordination among nodes when other tasks exist or tasks change. Furthermore, based on fundamental environmental data regarding the geographic grid within the network's coverage area of ​​the current task's location and the current node's geographic grid, a multi-dimensional constraint-based optimal routing cost function is constructed. Then, the nearest neighbor node of the current node is determined based on the minimum cost of this optimal routing cost function. Through spatial analysis of the nearest neighbor node, the lowest-latency connectivity graph of the current node is obtained, thus determining the optimal path. This allows multiple unmanned system devices to execute the current task along this optimal path, improving the accuracy of task scheduling and resource coordination among multiple unmanned system devices. This, in turn, enhances decision support capabilities under integrated situational awareness, improves the collaborative efficiency and accuracy among multiple unmanned devices at the site, enhances the control accuracy and efficiency of task collaboration, and ultimately improves task deployment efficiency and accuracy. These points are explained in detail below.

[0066] Example 1

[0067] Please see Figure 1 , Figure 1 This is a flowchart illustrating a decentralized networking method for digital twin unmanned sites disclosed in an embodiment of the present invention. Figure 1 The described method can be applied to any scenario requiring task planning, such as forest and grassland fire fighting, crop planting, power line inspection, urban traffic monitoring, etc. Figure 1 As shown, the method may include the following steps:

[0068] 101. After obtaining the corresponding digital twin site model by performing digital twin modeling on the unmanned site, the digital twin site model is determined as the node of the self-organizing network, and all nodes form the network; and the network is initialized to obtain a distributed architecture network, wherein the leader node of the network is elected during the initialization process.

[0069] In this embodiment of the invention, after digital twin modeling, unmanned sites are represented as geographic entities. Each unmanned site's corresponding geographic entity possesses a globally unique identifier and / or geohash code and metadata. The network consists of multiple nodes and is constructed based on the digital twin model of each node's corresponding unmanned site and its globally unique identifier and / or geohash code. The digital twin model of each unmanned site expresses a two-way virtual-real mapping between the digital world and the objective world of the unmanned site in the current real-world scenario. The globally unique identifier and / or geohash code of each unmanned site are used by the corresponding node in the network to maintain connection attributes with its neighboring nodes. By combining the globally unique identifier and / or geohash code of the unmanned sites to establish the network, nodes can maintain connection relationships through the globally unique identifier and / or geohash code after the network is established, achieving efficient and accurate node discovery and data location, enabling subsequent accurate and rapid task allocation, network structure reconstruction, and intelligent routing. It should be noted that if the current real-world scenario changes, the corresponding digital twin model changes accordingly, and further, the network changes synchronously.

[0070] In this embodiment of the invention, optionally, all unmanned stations include one or more of the following types: portable mobile node type, relay node type, and unattended node type. Furthermore, the metadata corresponding to each unmanned station has extensible attributes, meaning that each node corresponding to an unmanned station has corresponding basic fields and extended metadata fields. The basic fields include a globally unique identifier and / or geohash code for the corresponding unmanned station and its geographic grid in the network, used to express the commonalities of different unmanned stations. The extended fields include, but are not limited to, information such as whether a task has been assigned, the type of task assigned, the task execution status, and the resources required by the task (such as computing power, storage, payload, etc.), used to express the differences between different unmanned stations, thereby ensuring a unified expression between tasks and structures and reducing heterogeneity between tasks and structures. The metadata corresponding to each unmanned station is distributed and immutable in the network. Specifically, technologies such as blockchain and IPFS (InterPlanetary File System) are used to process the metadata corresponding to the unmanned stations to ensure its distributed and immutable storage in the network.

[0071] In this embodiment of the invention, optionally, for the initial election of a leader node, the elected leader node can be randomly selected, or it can be selected based on the resource situation of nodes in the network, such as selecting the node with the most resources as the leader node. Resources include, but are not limited to, one or more of computing power, storage, payload, and carrier. For subsequent leader node elections, the elected leader node can be selected based on resources and / or historical reputation and / or the number of times it has previously served as a leader node. For example, selecting a node that has never served as a leader node before, or selecting the node with the highest historical reputation. By determining the leader node through multiple methods, the accuracy and efficiency of leader node determination are improved, while balancing the actual situation of each node in the entire network and making full use of node resources.

[0072] 102. The current task is published to the network through the leader node. After receiving the current task, the network nodes perform multi-dimensional verification of the current task and perform voting operations according to the consensus rules to obtain the current node that claims the current task. After claiming the current task, the current node writes the message that the current task has been claimed into the extended field of the current node's metadata and synchronizes it to all other nodes in the network. The current node includes the leader node or other nodes in the network.

[0073] In this embodiment of the invention, after digital twin modeling, the current task is also represented as a geographic entity. The extended fields of the current node's metadata are used to record the current node's resources, which include one or more of computing power, storage, payload, carrier, and other schedulable nodes. The extended fields of the current node's metadata also record the current node's historical service parameters, including its historical service quality score and historical service frequency. The extended fields of the current task's metadata are used to record the resources required, which include one or more of computing power, storage, payload, carrier, and other schedulable nodes.

[0074] In this embodiment of the invention, for the message indicating that a current task has been claimed, in addition to the claim itself, if the current node is currently executing the current task, the message may also include the real-time execution status of the current task, such as "in progress" or "completed," and further, the task completion rate. The more task data published in this way, the more beneficial it is for other nodes in the network to accurately know the execution status of the current task, thus facilitating mutual coordination.

[0075] It should be noted that if the current node is the leader node, the publish operation at this time can be understood as the leader node publishing to itself or not executing the publish operation, and writing the message that the current task has been claimed by it into its extended field and synchronizing it to the network so that other nodes in the network can be aware of the message.

[0076] In this embodiment of the invention, the current node is specifically determined in the following way:

[0077] After receiving the current task, any node among multiple nodes in the network performs a multi-dimensional verification operation on the current task. After the current task passes, it performs a voting operation on the current task based on the consensus mechanism, obtains the voting result of the current task, and publishes the voting result of the current task to the network so that other nodes in the network can be aware of the voting result of the current task.

[0078] Any node in the network can provide feedback on the voting results of the current task it perceives, and receive feedback on the voting results.

[0079] Based on the voting results of all nodes and the feedback on the voting results, the voting importance of each node is analyzed, and the node with the highest voting importance is selected as the current node from all nodes according to the voting importance of all nodes.

[0080] In this embodiment of the invention, optional multi-dimensional verification operations include, but are not limited to, at least two of the following: task reachability verification, task legitimacy verification, and task priority verification. Task reachability includes one or more of the following: avoiding or overcoming obstacles along the route, weather conditions meeting deployment requirements, and the carrier having sufficient energy to reach the destination. Task legitimacy includes one or more of the following: having available airspace resources, having an approved mission plan, and the mission path not conflicting with prohibited / restricted areas. Task priority includes one or more of the following: priority determined by different task types, priority determined by time sequence in the queue of pending tasks, and priority determined by site task traffic optimization strategies. It should be noted that when the current task is received, multi-dimensional verification operations can be skipped, and a voting operation can be performed directly, especially when similar tasks have already been verified during the establishment of the corresponding digital twin model.

[0081] In this embodiment of the invention, when each node receives a current task, it publishes its request to execute the task and its own resources (including but not limited to computing power, storage, payload, carrier, and other schedulable nodes) to the network so that other nodes can perceive it. Upon perceiving the aforementioned message published by other nodes, each node votes on whether it agrees or disagrees with executing the current task and simultaneously publishes this result to the network. Each node provides feedback on the voting results published by other nodes, agreeing or disagreeing. Finally, for each node, its voting importance is determined based on the voting results and feedback from all other nodes. If it exceeds a preset voting importance (e.g., 2 / 3 of the other nodes agree to execute the current task), then that node becomes the current node and can claim the current task. Furthermore, if the number of nodes with a voting importance greater than or equal to the preset voting importance is greater than one, then the node with the highest voting importance is selected as the current node. This decentralized network uses a consensus mechanism for task assignment voting, ensuring fair participation from all relevant nodes and increasing the likelihood of timely and accurate task assignment. This, in turn, improves the accuracy of task scheduling and resource coordination among multiple unmanned system devices, enhances the accuracy and efficiency of task collaboration control, and ultimately improves task deployment efficiency and accuracy. Furthermore, verifying the data legality and / or integrity of tasks before voting based on the consensus mechanism ensures the executability of the voted tasks, thereby increasing the reliability of task voting and further enhancing the likelihood of timely and successful task assignment.

[0082] 103. Determine that the current task location belongs to a geographic grid within the radiation range of a node in the network. Based on the basic environmental data between the geographic grid and the geographic grid of the current node, construct the optimal routing cost function under multi-dimensional constraints. Based on the minimum cost value of the optimal routing cost function, determine the nearest neighbor node of the current node under multi-dimensional constraints. Through spatial analysis of the nearest neighbor node, dynamically construct the lowest latency connection graph of the current node. Based on the lowest latency connection graph, determine the optimal path of the current node. The optimal path of the current node is used as the basis for the current node to execute the current task.

[0083] In this embodiment of the invention, the optimal path may include, in addition to the route, the task planning speed and task planning direction of the route.

[0084] In this embodiment of the invention, the multidimensional constraints include, but are not limited to, at least two of the following: task reachability verification, task legality verification, and task priority verification. For a description of the contents of the multidimensional constraints, please refer to the above description of multidimensional verification, which will not be repeated here.

[0085] In this embodiment of the invention, the geographic grid within the radiation range of a node in the network where the current task is located is obtained in the following way: It is determined that the current task location belongs to a node in the network, and the geographic grid within the radiation range of that node is obtained from the digital twin model corresponding to that node. The geographic grid of the current node is obtained in the following way: It is determined that the digital twin model corresponding to the current node is determined, and the geographic grid of the current node is obtained from that digital twin model.

[0086] In this embodiment of the invention, optionally, the basic environmental data includes geographic environmental data, which may include, but is not limited to, natural and man-made ground environmental data such as surface data, elevation data, and terrain data within the radiation range. Geographic environmental data can be represented in the aforementioned various data formats, or it can be represented by a Digital Surface Model (DSM), Digital Elevation Model (DEM), or Digital Orthophoto Map (DOM) obtained based on the aforementioned data. Furthermore, geographic environmental data may also include electromagnetic environment data and climate data within the radiation range. The more basic environmental data available, the more beneficial it is to improving the accuracy of constructing the minimum cost function, thereby improving the accuracy of determining the minimum latency connectivity graph, and further improving the accuracy and reliability of determining the optimal path for task execution for nodes. This ensures that geographically neighboring nodes are prioritized during resource adjustments, reducing the latency of cross-regional resource scheduling, and thus improving the accuracy and efficiency of unmanned equipment task execution.

[0087] It is evident that implementation Figure 1The described method involves creating digital twin models of unmanned sites. These models then serve as nodes in a self-organizing network, forming a decentralized network. The network is initialized with a distributed structure, and a leader node is elected. When a task needs to be executed, the leader node publishes the task to the network. Nodes then perform multi-dimensional verification of the task and vote based on a consensus mechanism. The task is then claimed by the current node in the network. After claiming the task, the current node writes this claim message to an extended field in its metadata and synchronizes it to the network, allowing other nodes to be aware of the task claim status. This reduces the risk of data silos from multiple sources and heterogeneous data, and prevents tasks from being repeatedly claimed by other nodes in the network. This ensures that when other tasks exist... When tasks or missions change, other nodes can coordinate in a timely and accurate manner. Furthermore, based on the basic environmental data between the geographic grid within the radiation range of nodes in the network where the current mission is located and the geographic grid of the current node, an optimal routing cost function under multi-dimensional constraints is constructed. Then, based on the minimum cost of the optimal routing cost function, the nearest neighbor node of the current node is determined. Through spatial analysis of the nearest neighbor node, the lowest latency connection graph of the current node is obtained, thereby determining the optimal path. This allows multiple unmanned system devices to execute the current mission according to the optimal path, improving the accuracy of task scheduling and resource coordination among multiple unmanned system devices. This, in turn, enhances decision support capabilities under comprehensive situational awareness, improves the collaborative efficiency and accuracy among multiple unmanned devices in the site, improves the control accuracy and efficiency of task collaboration, and ultimately improves the efficiency and accuracy of mission deployment.

[0088] In an optional embodiment, before determining the optimal path for the current node based on the lowest latency connectivity graph, the method may further include the following steps:

[0089] Determine the number of nodes in the lowest latency connectivity graph. When the number is equal to 1, perform the above operation of determining the optimal path for the current node based on the lowest latency connectivity graph.

[0090] When the number is greater than or equal to 2, obtain the historical service quality score of the intermediate routing node of each lowest latency connection graph, and determine the reputation value of the intermediate routing node based on the historical service quality score of each intermediate routing node; based on the reputation values ​​of all intermediate routing nodes, select the lowest latency connection graph corresponding to the intermediate routing node with the highest reputation value from all intermediate routing nodes, and perform the above operation of determining the optimal path of the current node based on the lowest latency connection graph.

[0091] In this optional embodiment, the historical service quality score of each routing node can be determined by dimensions such as execution efficiency and execution accuracy when it performs historical tasks. For example, the higher the execution efficiency, the higher its historical service quality score, and the higher its reputation value.

[0092] In this optional embodiment, for each lowest latency connection graph, the number of intermediate routing nodes is greater than or equal to 1. When it is equal to 1, its corresponding reputation value refers to the reputation value of the unique intermediate routing node. When it is greater than 1, its corresponding reputation value refers to the average reputation value of all intermediate routing nodes or the reputation value of the intermediate routing node with the highest reputation value.

[0093] As can be seen, this optional embodiment, when multiple lowest-latency connection graphs exist, further selects the lowest-latency connection graph containing the intermediate routing node with the higher reputation value as the optimal path for the current node to execute the current task, thereby improving the accuracy of determining the optimal path. This helps to improve the accuracy of task allocation priority, thereby reducing resource monopoly, waste, or unevenness, and improving the execution efficiency and accuracy of the current task.

[0094] In another alternative embodiment, the method may further include the following steps:

[0095] When all intermediate routing nodes have the same reputation value, calculate the historical service frequency of each intermediate routing node, and based on the historical service frequency of all intermediate routing nodes, filter the lowest latency connection graph corresponding to the intermediate routing node with the lowest historical service frequency.

[0096] Among these, determining the optimal path for the current node based on the lowest-latency connectivity graph includes:

[0097] The optimal path for the current node is determined based on the lowest latency connection graph corresponding to the intermediate routing node with the lowest historical service frequency.

[0098] As can be seen, this optional embodiment, when there are multiple lowest latency connection graphs, if the reputation values ​​of the intermediate routing nodes in each lowest latency connection graph are equal, then the lowest latency connection graph corresponding to the intermediate routing node with the lowest historical service frequency is selected as the optimal path for the current task. While accurately selecting the optimal path, it also considers the node usage of the entire network, balances all network nodes, improves node resource utilization, and reduces node aging, which is beneficial to improving the lifespan of nodes and the entire network.

[0099] In another alternative embodiment, the method may further include the following steps:

[0100] After the current node claims the current task, it determines whether it needs to divide the task into subtasks based on the resources that the current node can provide and the resources required by the current task.

[0101] When it is determined that subtasks need to be divided, the current task is recursively decomposed into subtasks through the current node to obtain one or more indivisible atomic tasks.

[0102] Based on the resources required for each atomic task and the resources that other nodes in the network can provide, nodes in the network that have the ability to complete the atomic task are selected as candidate nodes. According to the consensus rules of the atomic task, the current node that will claim the atomic task is determined from all candidate nodes. After claiming the atomic task, the current node writes the message that the atomic task has been claimed into the extended field of the current node's metadata and synchronizes it to all other nodes in the network.

[0103] For any atomic task, update the atomic task to the current task and trigger the execution of the above operation to determine that the location of the current task belongs to the geographic grid within the radiation range of the nodes in the network.

[0104] In this optional embodiment, specifically, after the current node receives the current task, if the resources that the current node can provide are insufficient to meet the resources required by the current task (e.g., the computing power that the current node can provide is less than the resources required by the current task), it indicates that subtask partitioning is required. After partitioning into multiple atomic tasks, nodes capable of completing the atomic tasks are selected as candidate nodes based on the resources that other nodes in the network can provide, according to the resources required by each atomic task. Based on consensus rules, these candidate nodes then claim the atomic tasks, resulting in the current node for each atomic task. It should be noted that when multiple atomic tasks exist, the current node for an atomic task can be either the node that partitioned the task or a candidate node. That is, the node that partitioned the task can execute some of the atomic tasks itself, publish the remaining atomic tasks to the candidate nodes in the network for them to claim, or it can directly publish all atomic tasks to the candidate nodes in the network for them to claim. For any atomic task, the consensus rule is determined based on the specific type of the atomic task, including but not limited to PoW (Proof of Work), PoS (Proof of Stake), DPoS (Delegated Proof of Stake), Paxos (Paxos Consensus Algorithm), Raft (Raft Consensus Algorithm), and PBFT (Practical Byzantine Fault Tolerance). For details on other assignment processes, please refer to the aforementioned descriptions; they will not be repeated here. Furthermore, when it is determined that subtask partitioning is unnecessary, the operation of determining that the current task's location belongs to a geographical grid within the network's node coverage area is directly triggered. Each atomic task corresponds to a unique current node, where each current node includes the leader node or other nodes in the network. For a description of resources, please refer to the aforementioned descriptions; they will not be repeated here.

[0105] As can be seen, in this optional embodiment, after the current node receives the current task, it first analyzes and compares its own resource status with the resource consumption required by the current task. If the comparison shows that the current node's resources cannot meet the resource requirements of the current task, it divides the task into subtasks, then selects candidate nodes based on the resource status of network nodes and the resource consumption required by the atomic tasks, and claims the atomic tasks based on consensus rules. This allows multiple nodes to work together to complete the current task, which helps improve the coordination accuracy between current tasks and improves the accuracy and efficiency of the current task execution.

[0106] In yet another optional embodiment, the method may further include the following steps:

[0107] When the current conditions of the network are detected to meet the predetermined reconstruction conditions, the corresponding reconstruction type of the network is determined;

[0108] Based on the reconstruction type of the network, a reconstruction operation matching that reconstruction type is performed on the network. The reconstructed network is used as the basis for the reconstruction of the task to be executed.

[0109] In this optional embodiment, when a first condition and / or a second condition are detected in the network, it indicates that the current network condition is reconfigured. The first condition includes, but is not limited to, a node going offline, a node coming online, a node malfunctioning, or a change in the task being executed; in this case, the reconfiguration type includes node reconfiguration. The second condition includes, but is not limited to, changes in node resources or a partial network disconnection from the cloud; in this case, the reconfiguration type includes resource reconfiguration.

[0110] In this optional embodiment, the reconstructed network is used as the basis for task execution, which includes the current task or other tasks. Specifically, when the task is the current task, it means that the network reconstruction occurs during the execution of the current task.

[0111] In this optional embodiment, optionally, based on the reconstruction type corresponding to the network, a reconstruction operation matching the reconstruction type is performed on the network, including:

[0112] When the network reconstruction type includes node reconstruction, the network reconstruction operation is performed based on the globally unique identifiers and / or geohash codes of other schedulable nodes recorded in the extended fields of the metadata of the online nodes in the network. Specifically, when a node in the network changes, other online nodes in the network perceive the change based on the globally unique identifiers and / or geohash codes of the node whose status has changed, obtaining a change perception result. When the node whose status has changed is an offline or faulty node in the network, the change perception result is an offline perception result or a fault perception result; when the node whose status has changed is an online node in the network, the change perception result is an online perception result.

[0113] When the network's corresponding reconstruction type includes resource reconstruction, obtain the task execution status of the current node in the network, and perform resource reconstruction operations on the network based on the task execution status of the current node.

[0114] In this embodiment of the invention, nodes in the network publish data to other nodes through spatiotemporal data management and publishing services, and autonomously exchange metadata, such as node status and resource availability, so that nodes can be aware of each other. Specifically, when a node in the network changes, online nodes in the network perceive the change based on the node's globally unique identifier and / or geohash code, obtain the change perception result, and reconstruct the network based on the change perception result.

[0115] In this embodiment of the invention, optionally, the task execution status includes one or more of the following: task completion rate, task execution stage, and the geographic grid where the node executing the task is located. In this case, the node executing the task can publish its task execution status to the network during the task execution process. Nodes can perceive each other through globally unique identifiers and / or geo-hash codes, obtain the task execution status, and make resource adjustments, such as re-routing intelligently.

[0116] As can be seen, in this optional embodiment, whether during task execution or in an idle state, nodes in the network can perceive any changes in the entire network through the globally unique identifier and / or geohash code of the nodes, i.e., the connection attributes between nodes. Based on the perceived node changes, the network can reconstruct nodes, or perceive the task execution status of nodes currently executing tasks, so as to readjust the routing. That is, whether a node goes offline or online, or a node fails and causes node reconstruction, or a local network disconnects from the cloud and causes routing adjustments, the remaining nodes and their network can maintain basic network functions. This supports decentralized self-organizing networks and topology reconstruction for multiple unmanned equipment corresponding to sites, thereby ensuring the adaptive reconstruction of the network to carry tasks. It breaks through the limitations of decision support capabilities under comprehensive situational awareness, enabling the entire network to form a network quickly and accurately, improving the control accuracy and efficiency of task collaboration, or providing a precise, stable and reliable network for task execution.

[0117] In yet another optional embodiment, the method may further include the following steps:

[0118] When the node whose state has changed is the current node, it is determined whether the current node is the leader node. If it is determined that the current node is not the leader node, based on the leader node and the current node's execution status for the current task, the current task is re-published to the network so that the current node in the network can claim the current task.

[0119] When a leader node is identified, other leader nodes that are online and match the current task are determined from the network based on the current task execution status and the geographical grid of the leader node in the digital twin model corresponding to the network.

[0120] Update other leader nodes to become leader nodes and re-execute the operation of publishing the current task to the network so that the current node in the network can claim the current task.

[0121] It should be noted that if the current task is divided into multiple atomic tasks and executed by multiple current nodes, the other leader nodes can be one of these current nodes or other nodes in the network. For details on how to determine other leader nodes, please refer to the method for determining leader nodes in Example 1, which will not be repeated here.

[0122] As can be seen, this optional embodiment, when it is determined that the node that has changed is a node that is executing a task, further determines whether the leader node has changed. If not, the task claiming operation is re-performed based on the current task execution status; if so, the leader node is re-selected according to the current actual situation, so that the task can be quickly and accurately claimed by the reconstructed node in a timely manner, thereby further ensuring the accuracy and efficiency of task coordination control.

[0123] Example 2

[0124] Please see Figure 2 , Figure 2 This is a flowchart illustrating another decentralized networking method for digital twin unmanned sites disclosed in an embodiment of the present invention. Figure 2 The described method can be applied to any scenario requiring unmanned task planning, such as forest and grassland fire fighting, crop planting, power line inspection, urban traffic monitoring, etc. Figure 2 As shown, the method may include the following steps:

[0125] 201. When the current conditions of the network are detected to meet the predetermined reconstruction conditions, determine the reconstruction type of the network.

[0126] In this embodiment of the invention, optionally, when a first situation and / or a second situation are detected in the network, it indicates that the current network conditions are reconfigurable. The first situation includes, but is not limited to, a node going offline, a node coming online, a node malfunctioning, or a change in the task being executed; in this case, the reconfiguration type includes node reconfiguration. The second situation includes, but is not limited to, changes in node resources or a partial network disconnection from the cloud; in this case, the reconfiguration type includes resource reconfiguration.

[0127] 202. Based on the reconstruction type of the network, perform a reconstruction operation that matches the reconstruction type on the network. The reconstructed network is used as the basis for the reconstruction of the task to be executed.

[0128] In this embodiment of the invention, the reconstructed network is used as the basis for task execution, which includes the current task or other tasks. Specifically, when the task is the current task, it indicates that the network reconstruction occurs during the execution of the current task.

[0129] It should be noted that Embodiment 2 can exist independently or it can occur in the scenario of performing a task as described in Embodiment 1.

[0130] In this embodiment of the invention, optionally, according to the reconstruction type corresponding to the network, a reconstruction operation matching the reconstruction type is performed on the network, including:

[0131] When the network reconstruction type includes node reconstruction, the network reconstruction operation is performed based on the globally unique identifiers and / or geohash codes of other schedulable nodes recorded in the extended fields of the metadata of the online nodes in the network. Specifically, when a node in the network changes, other online nodes in the network perceive the change based on the globally unique identifiers and / or geohash codes of the node whose status has changed, obtaining a change perception result. When the node whose status has changed is an offline or faulty node in the network, the change perception result is an offline perception result or a fault perception result; when the node whose status has changed is an online node in the network, the change perception result is an online perception result.

[0132] When the network's corresponding reconstruction type includes resource reconstruction, obtain the task execution status of the current node in the network, and perform resource reconstruction operations on the network based on the task execution status of the current node.

[0133] In this embodiment of the invention, nodes in the network publish data to other nodes through spatiotemporal data management and publishing services, and autonomously exchange metadata, such as node status and resource availability, so that nodes can be aware of each other. Specifically, when a node in the network changes, online nodes in the network perceive the change based on the node's globally unique identifier and / or geohash code, obtain the change perception result, and reconstruct the network based on the change perception result.

[0134] In this embodiment of the invention, optionally, the task execution status includes one or more of the following: task completion rate, task execution stage, and the geographic grid where the node executing the task is located. In this case, the node executing the task can publish its task execution status to the network during the task execution process. Nodes can perceive each other through globally unique identifiers and / or geo-hash codes, obtain the task execution status, and make resource adjustments, such as re-routing intelligently.

[0135] It is evident that implementation Figure 2The described method allows network nodes to perceive any changes across the entire network, whether during task execution or in idle state, through their globally unique identifiers and / or geohash codes (i.e., the connection attributes between nodes). Based on these perceived changes, the method reconstructs the network nodes or senses the task execution status of nodes currently performing tasks to readjust routing. This means that regardless of whether a node goes offline or online, fails leading to node reconstruction, or experiences a local network disconnection from the cloud causing routing adjustments, basic network functionality can be maintained through the remaining nodes and their network. This supports decentralized self-organizing networks and topology reconstruction for multiple unmanned devices corresponding to various sites, ensuring adaptive reconstruction of the network to support tasks. It overcomes the limitations of decision support capabilities under comprehensive situational awareness, enabling rapid and accurate network formation across the entire network, improving the accuracy and efficiency of task coordination control, or providing a precise, stable, and reliable network for task execution.

[0136] In an optional embodiment, the method may further include the following steps:

[0137] When the node whose state has changed is the current node, it is determined whether the current node is the leader node. If it is determined that the current node is not the leader node, based on the leader node and the current node's execution status for the current task, the current task is re-published to the network so that the current node in the network can claim the current task.

[0138] When a leader node is identified, other leader nodes that are online and match the current task are determined from the network based on the current task execution status and the geographical grid of the leader node in the digital twin model corresponding to the network.

[0139] Update other leader nodes to become leader nodes and re-execute the operation of publishing the current task to the network so that the current node in the network can claim the current task.

[0140] It should be noted that if the current task is divided into multiple atomic tasks and executed by multiple current nodes, the other leader nodes can be one of these current nodes or other nodes in the network. For details on how to determine other leader nodes, please refer to the method for determining leader nodes in Example 1, which will not be repeated here.

[0141] As can be seen, this optional embodiment, when it is determined that the node that has changed is a node that is executing a task, further determines whether the leader node has changed. If not, the task claiming operation is re-performed based on the current task execution status; if so, the leader node is re-selected according to the current actual situation, so that the task can be quickly and accurately claimed by the reconstructed node in a timely manner, thereby further ensuring the accuracy and efficiency of task coordination control.

[0142] To enable those skilled in the art to better understand this technical solution, a certain type of unmanned hangar site will be used as an example. The following will combine... Figure 3-5 This technical solution will be described. Specifically, Figure 3 This is a network diagram of a perimeter control system for an unattended site disclosed in an embodiment of the present invention. Figure 4 This is a network diagram of another unattended site perimeter control system disclosed in an embodiment of the present invention. Figure 5 This is a schematic diagram of an intelligent router disclosed in an embodiment of the present invention. Figure 3 , 4 As shown, this perimeter control system includes two types of unmanned sites: unmanned hangars (or airports, aircraft nests) and unmanned underground buried monitoring equipment. Specifically:

[0143] After constructing digital twin models of the two types of unattended stations in the perimeter control system, all unattended stations are represented as geographical entities. They automatically obtain globally unique identifiers and neighbor node connectivity attributes for both types of stations. The extended fields of the geographical entities express the neighbor node connectivity attributes. During this network initialization process, a leader node is autonomously elected, thus forming a hybrid self-organizing network of both types of stations. Figure 3 , 4 As shown.

[0144] Specifically, Class II sites are generally deployed along the perimeter at the front end, such as... Figure 3 , 4 The stations a, b, c, d, e, f, g, h, i, j, and k shown are distributed relatively evenly along a winding boundary line. One type of station is deployed at the rear in a certain defensive formation, such as... Figure 3 , 4The A, B, and C sites shown partially overlap in coverage, forming a network that interconnects to create a perimeter defense system. One type of site can sense targets using the payload of its unmanned reconnaissance equipment (e.g., equipped with electro-optical pods, using visible light / infrared cameras supplemented by AI functions to track, lock onto, or verify targets intruding into the perimeter). The second type of site can sense targets using sensors in underground monitoring equipment (e.g., underground monitoring equipment using vibration sensing supplemented by AI functions to determine the type and number of targets intruding into the perimeter under different geological conditions). When a second-type site senses an intrusion, it transmits the information to the leader node, which then issues a close-in reconnaissance mission to the network. Nodes in the network then vote on the mission according to a consensus mechanism / rules to determine which node will accept it. The entire process of this reconnaissance mission's status changes is published to the network through a decentralized metadata service.

[0145] Specifically, such as Figure 3 As shown, the entire perimeter control system's task is assumed to be task 0. The main process of task 0 is to automatically dispatch unmanned systems that can be dispatched by the Class I station to the vicinity of the suspicious target (i.e., on the same side as the Class I station) based on the location of the suspicious target detected by the Class II station that has intruded into the inner boundary (i.e., on the same side as the Class I station) to conduct reconnaissance with maximum efficiency. The system compares and verifies the location with the targets detected by the Class II station. If the target is confirmed as a threat, it is tracked and locked, and real-time images of the threat are continuously captured. This real-time image data is published as an accessible resource within the network, and the unique identifier of the real-time image resource is shared throughout the network. Class I station B is elected as the leader node, which decomposes task 0 of the entire perimeter control system into sub-tasks 1, 2, and 3 (e.g., ...). Figure 3As shown by the middle arrow, tasks are assigned to three stations, A, B, and C, respectively. At this point, station A is paired with task 1, station B with task 2, and station C with task 3. The paired station and task pairs are traversed to achieve intelligent routing for tasks 1, 2, and 3 in sequence. Taking task 1 as an example of intelligent routing: First, GeoHash technology is used to quickly determine the geographic grid where the suspicious target that has intruded into the inner boundary is located, and it is determined that the geographic grid is within the coverage / radiation range of type II nodes a and b and type I node A. Based on the timeliness requirements decomposed from task 0, for task 1, based on the basic environmental data within the coverage / radiation range of type II nodes a and type I nodes A, the optimal routing cost function of multiple task paths is dynamically constructed through spatial analysis using the shortest time strategy. The task planning scheme with the minimum cost value of the optimal routing cost function (i.e., the highest timeliness) is selected for execution. That is, based on the minimum cost value of the optimal routing cost function, the nearest neighbor node of type I node A under multi-dimensional constraints is determined, and the lowest latency connection graph of type I node A is dynamically constructed through spatial analysis of the nearest neighbor node. Based on the lowest latency connection graph, the optimal path of type I node A is determined, thereby realizing the geographic awareness of the node. Furthermore, assuming that the minimum cost of using Class II nodes a and b as route endpoints is equal, and that the historical service quality score recorded in the metadata of Class II site a is 100 while the historical service quality score of Class II node b is 80, then according to the reputation scheduling mechanism, the Class II site a, which can better guarantee the quality of task completion, is selected as the route endpoint. Finally, in the entire network, the four-dimensional coordinates of stations A, B, C, and a, b, c, d, e, f, g, h, i, j, and k are all in the same spatiotemporal reference system. Each station is clock-aligned and can check the current node service status according to a preset period (e.g., 8 hours), automatically expanding, shrinking, and reclaiming idle resources (including but not limited to node computing power after completing the previous task, node storage, data published and shared in the network by nodes, and unmanned equipment controlled by nodes). Alternatively, if a type of station B goes offline due to force majeure, a type of station A is re-elected as the leader node. A then takes the lead in allocating subtasks, with type A assigned to subtasks 1 and 2, and type C assigned to subtask 3. This achieves the reconstruction of the entire perimeter control system network and its carried tasks, resulting in a new network, such as... Figure 4 As shown. More specifically, for smart routing, it can be achieved through methods such as... Figure 5 Please provide an explanation, such as Figure 5As shown, a network consisting of six stations A, B, C, D, E, and F is required to perform close-range reconnaissance tasks. The preceding steps are similar and will not be repeated. In implementing intelligent routing, the geographic grids corresponding to the six nodes A, B, C, D, E, and F are obtained. The location of the task objective belongs to a grid within the geographic range radiated by node E. An optimal routing cost function is constructed under multi-dimensional constraints, including but not limited to task reachability and legitimacy. Based on the minimum cost of the optimal routing cost function, the nearest neighbors of A under multi-dimensional constraints are determined to be B and C. A minimum latency connectivity graph of the network is dynamically constructed through spatial analysis (composed of the blue paths in the figure, including paths ACE and ABE). Based on the minimum latency connectivity graph, the optimal route candidate nodes are determined to be A, B, C, and E. Assuming that nodes B and C have the same historical service quality score, but node B has a lower historical service frequency than node C, then based on a comprehensive ranking of nodes with higher historical service quality and lower historical service frequency, the set of current nodes through which the optimal route passes is finally determined to be (A, B, E). By sequentially concatenating the current node sets, the optimal path is determined to be ABE, thus realizing the intelligent routing mechanism of the network.

[0146] Example 3

[0147] Please see Figure 6 , Figure 6 This is a schematic diagram of the structure of a decentralized networking system for digital twin unmanned sites disclosed in an embodiment of the present invention. This system can be applied to any scenario requiring task planning, such as forest fire fighting, crop planting, power line inspection, urban traffic monitoring, etc. Figure 6 As shown, the system includes:

[0148] The networking module 301 is used to determine the digital twin site model as a node of the self-organizing network after obtaining the corresponding digital twin site model by digital twin modeling of the unmanned site, and all nodes form the network; and to perform initialization operation on the network to obtain a distributed architecture network, wherein the network autonomously elects a leader node during the initialization process.

[0149] The task claiming module 302 is used to publish the current task to the network through the leader node. After receiving the current task, the network nodes perform multi-dimensional verification of the current task and execute voting operations according to the consensus rules to obtain the current node that claims the current task. After claiming the current task, the current node writes the message that the current task has been claimed into the extended field of the current node's metadata and synchronizes it to all other nodes in the network. The current node includes the leader node or other nodes in the network.

[0150] The routing module 303 is used to determine that the location of the current task belongs to the geographic grid within the radiation range of the nodes in the network, and to construct the optimal routing cost function under multi-dimensional constraints based on the basic environmental data between the geographic grid and the geographic grid of the current node. Based on the minimum cost of the optimal routing cost function, it determines the nearest neighbor node of the current node under multi-dimensional constraints, and dynamically constructs the lowest latency connection graph of the current node through spatial analysis of the nearest neighbor node. Based on the lowest latency connection graph, it determines the optimal path of the current node, whereby the optimal path of the current node is used as the basis for the current node to execute the current task.

[0151] In this embodiment of the invention, after digital twin modeling, both the unmanned site and the current task are represented as geographic entities. Each geographic entity corresponding to an unmanned site has a globally unique identifier, a geohash code, and metadata. The metadata corresponding to each unmanned site has extensible attributes.

[0152] The metadata corresponding to each unmanned site is stored in a distributed manner across the network and is tamper-proof;

[0153] The extended fields of the current node's metadata are used to record the current node's resources. The current node's resources include one or more of computing power, storage, load, carrier, and other schedulable nodes. The extended fields of the current node's metadata are also used to record the current node's historical service parameters, which include the current node's historical service quality score and historical service count.

[0154] The extended fields of the current task's metadata are used to record the resources required, including computing power, storage, payload, carrier, and one or more other nodes that need to be scheduled.

[0155] The geographic grid within the coverage area of ​​a node in the network where the current task is located is obtained using the following method:

[0156] Determine that the current task location belongs to a node in the network, and obtain the geographic grid within the range of the node's radiation area from the digital twin model corresponding to that node;

[0157] The geographic grid of the current node is obtained in the following way:

[0158] Determine the digital twin model corresponding to the current node, and obtain the geographic grid of the current node from the digital twin model;

[0159] Multi-dimensional verification includes at least two of the following: mission reachability verification, mission legality verification, and mission priority verification. Mission reachability includes one or more of the following: avoiding or overcoming obstacles along the route, weather conditions meeting the deployment requirements, and the carrier's energy support to reach the destination.

[0160] Mission legitimacy includes one or more of the following: having available airspace resources, having an approved mission plan, and ensuring that the mission path does not conflict with prohibited / restricted areas.

[0161] Task priority includes one or more of the following: priority determined by different task types, priority determined by time sequence in the queue of tasks to be executed, and priority determined by site task traffic optimization strategies.

[0162] It is evident that implementation Figure 6 The described system, after obtaining corresponding digital twin models of unmanned sites through digital twin modeling, uses all the digital twin models of unmanned sites as nodes in a self-organizing network, forming a decentralized network. The network is initialized to form a distributed structure, and a leader node is elected. When a task needs to be executed, the leader node publishes the task to the network, allowing other nodes to perform multi-dimensional verification of the task and vote based on a consensus mechanism. This ensures that the task is claimed by the current node in the network. After claiming the task, the current node writes the task claim message into an extended field of its metadata and synchronizes it to the network, enabling other nodes to be aware of the task claim status. This reduces the possibility of data silos from multiple sources and heterogeneous structures, and prevents tasks from being repeatedly claimed by other nodes in the network. This ensures that when other tasks exist... When tasks or missions change, other nodes can coordinate in a timely and accurate manner. Furthermore, based on the basic environmental data between the geographic grid within the radiation range of nodes in the network where the current mission is located and the geographic grid of the current node, an optimal routing cost function under multi-dimensional constraints is constructed. Then, based on the minimum cost of the optimal routing cost function, the nearest neighbor node of the current node is determined. Through spatial analysis of the nearest neighbor node, the lowest latency connection graph of the current node is obtained, thereby determining the optimal path. This allows multiple unmanned system devices to execute the current mission according to the optimal path, improving the accuracy of task scheduling and resource coordination among multiple unmanned system devices. This, in turn, enhances decision support capabilities under comprehensive situational awareness, improves the collaborative efficiency and accuracy among multiple unmanned devices in the site, improves the control accuracy and efficiency of task collaboration, and ultimately improves the efficiency and accuracy of mission deployment.

[0163] In an optional embodiment, Figure 7 This is a schematic diagram of another decentralized networking system for digital twin unmanned sites disclosed in an embodiment of the present invention. Figure 7 As shown, the system may also include:

[0164] The judgment module 304 is used to determine whether subtask division is needed after the current node has claimed the current task, based on the resources that the current node can provide and the resources required by the current task.

[0165] The task decomposition module 305 is used to recursively decompose the current task into subtasks through the current node when it is determined that subtasks need to be divided, so as to obtain one or more indivisible atomic tasks.

[0166] The filtering module 306 is used to filter nodes in the network that have the ability to complete the atomic task as candidate nodes based on the resources required for each atomic task and the resources that other nodes in the network can provide. According to the consensus rules of the atomic task, the current node is determined from all candidate nodes to claim the atomic task. After claiming the atomic task, the current node writes the message that the atomic task has been claimed into the extended field of the current node's metadata and synchronizes it to all other nodes in the network. For any atomic task, the atomic task is updated to the current task, and the routing module 303 is triggered to perform the above-mentioned operation of determining that the location of the current task belongs to the geographical grid within the radiation range of the nodes in the network.

[0167] It is evident that implementation Figure 7 After receiving the current task, the current node in the described system first analyzes and compares its own resources with the resources required by the current task. If the current node's resources cannot meet the resource requirements of the current task, it divides the task into subtasks. Then, it selects candidate nodes based on the resource status of network nodes and the resources required by the atomic tasks, and assigns atomic tasks based on consensus rules. This allows multiple nodes to work together to complete the current task, which helps improve the coordination accuracy between current tasks and improves the accuracy and efficiency of the current task execution.

[0168] In another alternative embodiment, such as Figure 7 As shown, the system also includes:

[0169] The determination module 307 is used to determine the reconstruction type of the network when the current conditions of the network are detected to meet the predetermined reconstruction conditions;

[0170] The networking module 301 is also used to perform a reconstruction operation on the network that matches the reconstruction type according to the reconstruction type of the network, wherein the reconstructed network is used as the basis for the reconstruction of the task to be executed;

[0171] The networking module 301, based on the network's corresponding reconstruction type, specifies the specific methods for performing reconstruction operations that match that reconstruction type, including:

[0172] When the network reconstruction type includes node reconstruction, the network reconstruction operation is performed based on the globally unique identifiers and / or geohash codes of other schedulable nodes recorded in the extended fields of the metadata of the online nodes in the network. Specifically, when a node in the network changes, other online nodes in the network perceive the change based on the globally unique identifiers and / or geohash codes of the node whose status has changed, obtaining a change perception result. When the node whose status has changed is an offline or faulty node in the network, the change perception result is an offline perception result or a fault perception result; when the node whose status has changed is an online node in the network, the change perception result is an online perception result.

[0173] When the network's corresponding reconstruction type includes resource reconstruction, obtain the task execution status of the current node in the network, and perform resource reconstruction operations on the network based on the task execution status of the current node.

[0174] It is evident that implementation Figure 7 The described system allows network nodes to perceive any changes across the entire network, whether during task execution or in idle state, through their globally unique identifiers and / or geo-hash codes (i.e., the connectivity attributes between nodes). Based on these perceived changes, the system can reconstruct the network or assess the task execution status of nodes currently performing tasks to readjust routing. This means that even if a node goes offline or online, malfunctions leading to node reconstruction, or experiences a local network disconnection from the cloud causing routing adjustments, the remaining nodes and their network can maintain basic network functionality. The system supports decentralized self-organizing networks and topology reconstruction for multiple unmanned devices at corresponding sites, ensuring adaptive reconstruction of the network to support tasks. This overcomes the limitations of decision support capabilities under comprehensive situational awareness, enabling rapid and accurate network formation across the entire network, improving the accuracy and efficiency of task coordination control, or providing a precise, stable, and reliable network for task execution.

[0175] In yet another alternative embodiment, such as Figure 7As shown, the networking module 301 is also used to determine whether the current node is a leader node when the node whose state has changed is the current node. If it is determined that the current node is not a leader node, it re-publishes the current task to the network based on the current node's execution status for the current task, so that the current node in the network can claim the current task. If it is determined that the current node is a leader node, it determines other leader nodes that are online and match the current task from the network based on the current task's execution status and the geographical grid of the leader node in the digital twin model corresponding to the network. It updates the other leader nodes as leader nodes and re-executes the above-mentioned operation of publishing the current task to the network so that the current node in the network can claim the current task.

[0176] It is evident that implementation Figure 7 The described system, upon determining that a node that has changed is one that is currently executing a task, further determines whether the leader node has changed. If not, it re-accepts the task based on the current task execution status; if so, it re-selects the leader node based on the current situation. This ensures that tasks can be quickly and accurately claimed by the reconstructed node, further guaranteeing the accuracy and efficiency of task coordination control.

[0177] In yet another alternative embodiment, such as Figure 7 As shown, the routing module 303 is further configured to determine the number of minimum latency connection graphs before determining the optimal path of the current node based on the minimum latency connection graph. When the number is equal to 1, the operation of determining the optimal path of the current node based on the minimum latency connection graph is performed. When the number is greater than or equal to 2, the historical service quality score of the intermediate routing nodes in each minimum latency connection graph is obtained, and the reputation value of the intermediate routing node is determined based on the historical service quality score of each intermediate routing node. Based on the reputation values ​​of all intermediate routing nodes, the minimum latency connection graph corresponding to the intermediate routing node with the highest reputation value is selected from all intermediate routing nodes, and the operation of determining the optimal path of the current node based on the minimum latency connection graph is performed.

[0178] The routing module 303 is also used to calculate the historical service frequency of each intermediate routing node when all intermediate routing nodes have the same reputation value, and to filter the lowest latency connection graph corresponding to the intermediate routing node with the lowest historical service frequency based on the historical service frequency of all intermediate routing nodes.

[0179] The specific method by which the routing module 303 determines the optimal path for the current node based on the lowest-latency connectivity graph includes:

[0180] The optimal path for the current node is determined based on the lowest latency connection graph corresponding to the intermediate routing node with the lowest historical service frequency.

[0181] It is evident that implementation Figure 7 The described system, when multiple lowest-latency connection graphs exist, further selects the lowest-latency connection graph containing the intermediate routing node with the higher reputation value as the optimal path for the current node to execute the current task. This improves the accuracy of optimal path determination, thereby improving the accuracy of task allocation priority, reducing resource monopolies, waste, or uneven distribution, and improving the execution efficiency and accuracy of the current task. Furthermore, when multiple lowest-latency connection graphs exist, if the reputation values ​​of the intermediate routing nodes in each graph are equal, the system selects the lowest-latency connection graph corresponding to the intermediate routing node with the lowest historical service frequency as the optimal path for the current task. This accurately selects the optimal path while considering the overall network node usage, balancing all network nodes, improving node resource utilization, reducing node aging, and ultimately extending the lifespan of nodes and the entire network.

[0182] Example 4

[0183] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of another decentralized networking system for digital twin unmanned sites disclosed in this invention. This system can be applied to any scenario requiring task planning, such as forest fire fighting, crop planting, power line inspection, urban traffic monitoring, etc. Figure 8 As shown, the system may include:

[0184] Memory 401 storing executable program code;

[0185] Processor 402 coupled to memory 401;

[0186] Furthermore, it may also include an input interface 403 and an output interface 404 coupled to the processor 402;

[0187] The processor 402 calls the executable program code stored in the memory 401 to execute some or all of the steps in the decentralized networking method for digital twin unmanned sites disclosed in Embodiment 1 or Embodiment 2 of the present invention.

[0188] Example 5

[0189] This invention discloses a computer storage medium storing computer instructions. When these computer instructions are invoked, they are used to execute some or all of the steps in the decentralized networking method for digital twin unmanned sites disclosed in Embodiment 1 or Embodiment 2 of this invention.

[0190] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0191] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0192] Finally, it should be noted that the decentralized networking method and system for digital twin unmanned sites disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A decentralized networking method for digital twin unmanned sites, characterized in that, The method includes: After obtaining the corresponding digital twin site model by performing digital twin modeling on the unmanned site, the digital twin site model is determined as the node of the self-organizing network, and all the nodes form the network topology; and the network is initialized to obtain a distributed architecture network, wherein the leader node of the network is elected during the initialization process; The leader node publishes the current task to the network, wherein, upon receiving the current task, the network nodes perform multi-dimensional verification of the current task and execute a voting operation according to consensus rules to obtain a current node that claims the current task; wherein, after claiming the current task, the current node writes the message that the current task has been claimed into the extended field of the current node's metadata and synchronizes it to all other nodes in the network; wherein, the current node includes the leader node or other nodes in the network; The system determines that the location of the current task belongs to a geographic grid within the radiation range of a node in the network. Based on the basic environmental data between the geographic grid and the geographic grid of the current node, it constructs an optimal routing cost function under multi-dimensional constraints. Based on the minimum cost value of the optimal routing cost function, it determines the nearest neighbor node of the current node under the multi-dimensional constraints. Through spatial analysis of the nearest neighbor node, it dynamically constructs a minimum latency connectivity graph of the current node. Based on the minimum latency connectivity graph, it determines the optimal path of the current node, wherein the optimal path of the current node is used as the basis for the current node to execute the current task.

2. The decentralized networking method for digital twin unmanned sites according to claim 1, characterized in that, After digital twin modeling, both the unmanned station and the current task are expressed as geographic entities. Each geographic entity corresponding to the unmanned station has a globally unique identifier and / or geohash code and metadata. The metadata corresponding to each unmanned station has extensible attributes. The metadata corresponding to each unmanned station is stored in a distributed manner in the network and is tamper-proof; The extended fields of the metadata of the current node are used to record the resources of the current node, wherein the resources of the current node include one or more of computing power, storage, load, carrier, and other schedulable nodes. The extended fields of the metadata of the current node are also used to record the historical service parameters of the current node, wherein the historical service parameters of the current node include the historical service quality score and the number of historical services. The extended fields of the metadata of the current task are used to record the resources required, including computing power, storage, payload, carrier, and other nodes that need to be scheduled.

3. The decentralized networking method for digital twin unmanned sites according to claim 1 or 2, characterized in that, The method further includes: After the current node claims the current task, it determines whether subtask division is necessary based on the resources it can provide and the resources required by the current task. When it is determined that subtasks need to be divided, the current task is recursively decomposed into subtasks through the current node to obtain one or more indivisible atomic tasks. Based on the resources required for each atomic task and the resources that other nodes in the network can provide, nodes in the network that have the ability to complete the atomic task are selected as candidate nodes. According to the consensus rules of the atomic task, the current node that claims the atomic task is determined from all the candidate nodes. After claiming the atomic task, the current node writes the message that the atomic task has been claimed into the extended field of the current node's metadata and synchronizes it to all other nodes in the network. For any of the atomic tasks, update the atomic task to the current task and trigger the operation of determining that the location of the current task belongs to the geographic grid within the radiation range of the nodes in the network.

4. The decentralized networking method for digital twin unmanned sites according to claim 1 or 2, characterized in that, The geographic grid within the radiation range of the nodes in the network where the current task is located is obtained through the following methods: Determine that the current task's location belongs to a node in the network, and obtain the geographic grid within the node's radiation range from the digital twin model corresponding to that node; The geographic grid of the current node is obtained in the following way: Determine the digital twin model corresponding to the current node, and obtain the geographic grid of the current node from the digital twin model; The multi-dimensional verification includes at least two of the following: task reachability verification, task legality verification, and task priority verification. The task reachability includes one or more of the following: avoiding or overcoming obstacles along the route, weather conditions meeting the deployment requirements, and the carrier's energy support to reach the destination. The legitimacy of the mission includes one or more of the following: having airspace resources, having an approved mission plan, and the mission path not conflicting with prohibited / restricted areas. The task priority includes one or more of the following: priority determined by different task types, priority determined by time sequence in the queue of tasks to be executed, and priority determined by site task traffic optimization strategies.

5. The decentralized networking method for digital twin unmanned sites according to claim 1 or 2, characterized in that, The method further includes: When it is detected that the current conditions of the network meet the predetermined reconstruction conditions, the reconstruction type corresponding to the network is determined; Based on the reconstruction type corresponding to the network, a reconstruction operation matching the reconstruction type is performed on the network, wherein the reconstructed network is used as the basis for the reconstruction of the task to be executed; The step of performing a reconstruction operation on the network that matches the reconstruction type according to the network's corresponding reconstruction type includes: When the network's corresponding reconstruction type includes a node reconstruction type, a reconstruction operation is performed on the network based on the globally unique identifiers and / or geohash codes of other schedulable nodes recorded in the extended fields of their metadata for nodes that are currently online. When a node in the network changes, other nodes currently online perceive the change based on the globally unique identifiers and / or geohash codes of the node whose status has changed, obtaining a change perception result. Specifically, when the node whose status has changed is an offline node or a faulty node in the network, the change perception result is an offline perception result or a fault perception result; when the node whose status has changed is an online node in the network, the change perception result is an online perception result. When the reconstruction type corresponding to the network includes the resource reconstruction type, the task execution status of the current node in the network is obtained, and the resource reconstruction operation of the network is performed according to the task execution status of the current node.

6. The decentralized networking method for digital twin unmanned sites according to claim 5, characterized in that, The method further includes: When the node whose state has changed is the current node, it is determined whether the current node is the leader node. If it is determined that the current node is not the leader node, based on the leader node and according to the current node's execution status of the current task, the current task is re-published to the network so that the current node in the network can claim the current task. When the leader node is identified, other leader nodes that are online and match the current task are determined from the network based on the task execution status of the current task and the geographical grid of the leader node in the digital twin model corresponding to the network. The other leader nodes are updated to become the leader node, and the operation of publishing the current task to the network is re-executed so that the current node in the network can claim the current task.

7. The decentralized networking method for digital twin unmanned sites according to any one of claims 1, 2, and 6, characterized in that, Before determining the optimal path for the current node based on the lowest latency connectivity graph, the method further includes: Determine the number of the lowest latency connection graphs. When the number is equal to 1, perform the operation of determining the optimal path of the current node based on the lowest latency connection graphs. When the number is greater than or equal to 2, obtain the historical quality of service score of each intermediate routing node in the lowest latency connection graph, and determine the reputation value of the intermediate routing node based on the historical quality of service score of each intermediate routing node; based on the reputation values ​​of all intermediate routing nodes, select the lowest latency connection graph corresponding to the intermediate routing node with the highest reputation value from all intermediate routing nodes, and perform the operation of determining the optimal path of the current node based on the lowest latency connection graph; The method further includes: When the reputation values ​​of all intermediate routing nodes are equal, the historical service frequency of each intermediate routing node is calculated, and based on the historical service frequency of all intermediate routing nodes, the lowest latency connection graph corresponding to the intermediate routing node with the lowest historical service frequency is selected. The step of determining the optimal path for the current node based on the lowest latency connectivity graph includes: The optimal path for the current node is determined based on the lowest latency connection graph corresponding to the intermediate routing node with the lowest historical service frequency.

8. A decentralized networking system for digital twin unmanned sites, characterized in that, The system includes: The networking module is used to determine the digital twin site model as a node of the self-organizing network after obtaining the corresponding digital twin site model by performing digital twin modeling on the unmanned site, and all the nodes form a network; and to perform an initialization operation on the network to obtain a distributed architecture network, wherein the leader node of the network is elected during the initialization process; The task claiming module is used to publish the current task to the network through the leader node. Upon receiving the current task, the network nodes perform multi-dimensional verification and execute a voting operation according to consensus rules to determine the current node that claims the task. After claiming the task, the current node writes the message that it has claimed the task into an extended field of its metadata and synchronizes it to all other nodes in the network. The current node includes the leader node or other nodes in the network. The routing module is used to determine that the location of the current task belongs to a geographic grid within the radiation range of a node in the network, and to construct an optimal routing cost function under multi-dimensional constraints based on the basic environmental data between the geographic grid and the geographic grid of the current node. Based on the minimum cost value of the optimal routing cost function, it determines the nearest neighbor node of the current node under the multi-dimensional constraints, and dynamically constructs a minimum latency connectivity graph of the current node through spatial analysis of the nearest neighbor node. Finally, it determines the optimal path of the current node based on the minimum latency connectivity graph, wherein the optimal path of the current node serves as the basis for the current node to execute the current task.

9. A decentralized networking system for digital twin unmanned sites, characterized in that, The system includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the decentralized networking method for digital twin unmanned sites as described in any one of claims 1-7.

10. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the decentralized networking method for digital twin unmanned sites as described in any one of claims 1-7.

Images