Distributed intelligent planning decision communication integrated terminal communication management method

Abstract

The invention discloses a distributed intelligent planning decision communication integrated terminal communication management method, which comprises the following steps that: an unmanned terminal platform converts a task demand into a plurality of dimensions of communication capability indexes based on a task demand and in combination with an expert knowledge base; constructing a hardware resource topology network, constructing and solving a multi-target optimization model on the basis of the hardware resource topology network, and outputting an optimal candidate scheme for configuring hardware and resources of the unmanned terminal platform; constructing a communication network by taking the unmanned terminal platform as a node; starting a hierarchical time synchronization mechanism based on the topological structure of the communication network, and constructing a communication sub-network; multi-level state management of the communication network is realized by constructing a network topology table, a state information table and an information maintenance rule table; and when a topological structure is changed, a task stage is switched or a node silence event is triggered, updating the network topology table and performing synchronization to realize global state consistency maintenance.

Description

Technical Field

[0001] The present invention relates to the field of communication management of unmanned terminal platforms, and particularly to a communication management method for an integrated terminal of distributed intelligent planning, decision-making, and communication. Background Art

[0002] With the rapid development of unmanned cluster technology, the market demand for intelligent, multi-functional, and high-performance cluster systems is increasing continuously. Unmanned clusters are widely used in multiple fields such as environmental monitoring, logistics distribution, and disaster emergency. With their high degree of autonomy and flexibility, they can perform diverse tasks in complex and dynamic environments. Existing cluster systems have gradually expanded from single functions to comprehensive task execution. However, most products on the market still focus on data transmission and single-function intelligent decision-making, lacking sufficient integration and collaborative processing capabilities, and unable to fully meet the requirements of complex tasks and efficient collaboration.

[0003] The decision-making and planning of cluster systems are usually closely coupled with communication. However, many current systems design the decision-making and planning module and the communication module separately, which leads to multiple problems. First of all, the independent decision-making module and communication module often cannot achieve real-time and efficient information flow and task allocation, increasing the complexity and latency of the system and affecting the timeliness and accuracy of decision-making. The communication efficiency between cluster members directly affects the decision-making quality. The separately designed modules are difficult to quickly respond to task changes in a dynamic environment, reducing the overall collaboration ability and system robustness of the cluster.

[0004] Currently, communication data links are often used by decision-making planners in a transparent transmission mode. In high-dynamic and high-real-time cluster application scenarios, high-frequency and large-volume data interactions will additionally occupy the computing power of decision-making planners. Especially for large-scale clusters, more resources will be occupied, reducing the overall efficiency of the system. Summary of the Invention

[0005] The purpose of the present invention is to provide a communication management method for an integrated terminal of distributed intelligent planning, decision-making, and communication, so as to solve problems such as low system efficiency and resource waste existing in the prior art.

[0006] To achieve the above tasks, the present invention adopts the following technical solutions: A communication management method for an integrated terminal of distributed intelligent planning, decision-making, and communication, comprising: Based on task requirements and combined with an expert knowledge base, the unmanned terminal platform quantifies the task requirements into communication capability indicators in multiple dimensions; according to the hardware and resources of the unmanned terminal platform itself, constructs a hardware resource topology network, and on this basis constructs and solves a multi-objective optimization model, and finally outputs an optimal candidate solution for configuring the hardware and resources of the unmanned terminal platform; Construct a communication network with unmanned terminal platforms as nodes; the network mode of the communication network is determined by the task type, including a centralized mode or a distributed mode; subsequently, start a hierarchical time synchronization mechanism based on the topological structure of the communication network and construct a communication subnet. Implement multi-level state management of the communication network by constructing a network topology table, a status information table, and an information maintenance rule table; among them, the network topology table records the communication status of nodes and the relationship with the communication subnet, the status information table stores the status information of each node, and the maintenance rule table defines the update frequency of each status information; when a node sends status information, it performs structured encapsulation and triggers encryption transmission and retransmission control based on the information maintenance rule table; when the topological structure changes, the task phase switches, or a node silence event is triggered, update the network topology table and synchronize it to achieve global status consistency maintenance.

[0007] Furthermore, a task requirement - communication capability index mapping relationship table is constructed in the expert knowledge base to quantify the task requirements into communication capability indexes in multiple dimensions; the dimensions of the communication capability indexes include communication capability, anti-interference level, power consumption efficiency level, concealment requirement, real-time level, and network management status.

[0008] Furthermore, the hardware resource topology network has dual node types, including a hardware node set, a resource node set, and an edge set. Among them, the hardware node set is composed of hardware nodes, and the attributes of the hardware nodes include node status and available power margin; the resource node set is composed of resource nodes, and the attributes of the resource nodes include resource availability and channel occupancy rate; the edge set is composed of the edges between the hardware nodes and the resource nodes, and the attributes of the edges include bandwidth allocation and power consumption. The constraint conditions of the hardware resource topology network include hardware compatibility constraint, power capacity constraint, and bandwidth allocation constraint.

[0009] Furthermore, construct a multi-objective optimization model according to the communication capability indexes and the hardware resource topology network, with maximizing the satisfaction of task requirements as the objective function and the hardware resource topology network as the constraint condition, use the multi-objective granular communication subnet to optimize and search for the Pareto solution set, and output multiple groups of candidate solutions; the objective function is defined as follows:

[0010]

[0011] Where is the fitness of the k th dimension in the dimensional communication capability index; The specific values of the communication capability indicators required for the candidate solutions.

[0012] Furthermore, during the solution process of the multi-objective optimization model, when parameter conflicts occur, dynamic adjustment is performed. Refer to the priority order of the dimensions in the communication capability indicators corresponding to the task requirements to adjust the conflicts, and give priority to satisfying the dimensions with higher priorities; after adjusting the conflicts, recalculate the objective function of the candidate solutions, and select the candidate solution with the maximum objective function as the final solution for output.

[0013] Furthermore, the network mode of the communication network is determined by the task type, where: When the task type requires global consistency, a centralized mode is adopted. Based on the dynamic election of node computing power and load weights, a central node is generated, a star topology is constructed, and a central node identifier is assigned; When the task type requires high fault tolerance, switch to the distributed mode, call the neighbor discovery protocol to establish a centerless Mesh network, and each node outputs the initial communication network structure including node connection relationships and mode identifiers to neighbor nodes.

[0014] Furthermore, the hierarchical synchronization mechanism is as follows: In the centralized mode, the central node of the communication network periodically broadcasts a reference clock signal, and the remaining nodes use the round-trip delay compensation algorithm to calibrate their local clocks step by step; in the distributed mode, each node exchanges timestamps with multiple neighbor nodes, constructs a clock deviation matrix, and calculates the global optimal clock reference through the Byzantine fault-tolerant consensus algorithm; when a new node joins the communication network, each node dynamically obtains the neighbor node list and uses the window sliding average filtering and Kalman prediction algorithms to complete the convergence of the clock offset.

[0015] Furthermore, the communication network structure performs communication grouping according to the task type to form communication subnets. Nodes within each communication subnet maintain data synchronization through the multicast protocol, and logical isolation strategies are adopted between communication subnets; When node addition, offline, or silent events are triggered in each communication subnet, hierarchical processing is performed: First, reconstruct the topological structure of the communication subnet; based on the current communication subnet, use the incremental Delaunay triangulation algorithm to update the connection relationships of nodes; if the triangulation fails, then enable the Kruskal greedy algorithm to construct a minimum spanning tree; Secondly, handle the silent state; move the silent nodes into the shadow topology, retain their logical state information but disable the physical layer transceiver functions; Finally, perform state synchronization; the communication subnet pushes the reconstructed topological structure through the multicast protocol. After each node receives the topological incremental data packet, it uses vector clock comparison and merging to update, and outputs the real-time topological mapping table to other neighbor nodes to complete the closed-loop consistency maintenance.

[0016] Further, the network topology table is used to record the unique identifier of the node, the communication subnet number it belongs to, the working status, the communication link quality index, the central node identifier, and the timestamp of the most recent data update; The status information table stores the local status information with each node as an independent unit, including location information, motion parameters, and user-defined status variables; The information maintenance rule table defines the update frequencies of the network topology status and the status information of all nodes; where the status information includes location information, motion parameters, and user-defined status variables; based on this table, status query instructions are periodically generated or a broadcast push mechanism is triggered; where the status query instructions are used to query the status information of a certain node; the broadcast push mechanism is used for a node to push its own status information to other nodes.

[0017] Further, based on the update frequencies defined in the information maintenance rule table, the current node performs protocol encapsulation on the status information to be sent to construct a data packet; where, according to the type of the status information, the unique identifier of the target node or the hardware address of the node is bound, and the reachability of the target node is verified through the network topology table; transmission policy parameters are configured, including the switch status of the acknowledgment mechanism, the data encryption flag, the sending period, and the number of retransmissions; the update frequency of the status information is synchronously updated to the information maintenance rule table to be incorporated into the global scheduling.

[0018] Further, data transmission is performed based on the update frequencies and encapsulation policies in the information maintenance rule table: For data packets with the encryption flag enabled, encryption is performed using an encryption algorithm; unicast, multicast, or broadcast communication is triggered according to the update frequency. If the switch status of the acknowledgment mechanism is enabled, the target node needs to return an acknowledgment signal containing a hash check value within the timeout window. The current node detects the acknowledgment status and retransmits according to the exponential backoff algorithm after the timeout. After the transmission is completed, the corresponding content in the status information table is updated according to the transmitted status information.

[0019] Further, when the topology structure of the communication subnet changes, the task phase switches, or a node's silent event is triggered, the network topology table is updated; after the update, each node writes its local status information into the status information table according to the update frequency in the information maintenance rule table cycle. At the same time, a status information query request is sent to neighboring nodes through the status query instructions. After receiving the status information difference data returned by the neighboring nodes, a timestamp comparison or version number conflict detection mechanism is used to update the local status information table.

[0020] An unmanned terminal platform, on which the above-mentioned distributed intelligent planning decision-making communication integrated terminal communication management method is deployed.

[0021] A computer-readable storage medium stores a computer program; when the computer program is executed by a processor, it implements the communication management method of the distributed intelligent planning decision-making communication integrated terminal.

[0022] Compared with the prior art, the present invention has the following technical features: The present invention is applicable to the communication management of unmanned cluster systems, especially focusing on aspects such as communication peripheral management, node collaborative supervision, information update and transmission, etc.; this method can dynamically configure communication hardware and network topology according to task requirements and ensure efficient and stable communication; through priority scheduling and rule-driven, it takes into account the requirements of information transmission efficiency, security and real-time performance, and solves the problems of low efficiency, response delay, resource waste, etc. caused by the separation of communication and decision-making in the prior art. Description of the Drawings

[0023] Figure 1 is the communication management framework of the decision-making planning communication integrated terminal; Figure 2 is the information flow of the decision-making planning communication integrated terminal; Figure 3 is the information table that needs to be automatically maintained. Detailed Embodiments

[0024] The present invention provides a communication management method for a distributed intelligent planning decision-making communication integrated terminal, which is deployed on unmanned terminal platforms such as unmanned aerial vehicles and unmanned vehicles; during the execution of tasks by the unmanned terminal platform, this method is used to dynamically configure communication hardware, collaborate on node supervision, and intelligently maintain the information transmission process.

[0025] A communication management method for a distributed intelligent planning decision-making communication integrated terminal of the present invention includes dynamic configuration of hardware peripherals, collaborative supervision of cluster nodes and dynamic maintenance of network topology, and dynamic information maintenance, where: 1. Dynamic configuration of hardware peripherals.

[0026] Based on task requirements and combined with an expert knowledge base, the unmanned terminal platform quantifies the task requirements into communication capability indicators in multiple dimensions; according to the hardware and resources of the unmanned terminal platform itself, it constructs a hardware resource topology network, and on this basis constructs and solves a multi-objective optimization model, and finally outputs an optimal candidate solution to configure the hardware and resources of the unmanned terminal platform.

[0027] Step 1.1, quantitative analysis and parsing of task requirements.

[0028] The unmanned terminal platform parses the tasks sent by the upper-layer task software to obtain task requirements, and based on the task requirement - communication capability index mapping relation table in the preset expert knowledge base, quantifies the task requirements into communication capability indexes in multiple dimensions; the dimensions of the communication capability indexes include communication capability, anti-jamming level, power consumption and energy efficiency level, concealment requirement, real-time level, and network management status; among them, the first five dimensions adopt a three-level quantization scale of low, medium, and high, and the last dimension of network management status is divided into silent state, online state, and offline state; the dimension combination of specific communication capability indexes is determined through the task type; in an exemplary implementation, the index combination corresponding to the cluster collaborative reconnaissance task is: communication capability (high), anti-jamming level (medium), power consumption and energy efficiency level (low), concealment requirement (medium), real-time level (low), network management status (online).

[0029] After determining the task type through task parsing, different task types have different task requirements for communication. Some task requirements require a large communication bandwidth, and some task requirements require communication silence and stealth; therefore, it is necessary to consult the expert knowledge base according to the task type to determine the task requirements.

[0030] Step 1.2, hardware resource topology network modeling. The hardware connection status, power margin, and channel resource availability on the unmanned terminal platform are monitored in real time through the data acquisition interface, and a hardware resource topology network with dual node types is established. This topology network includes: Hardware node set , where represents one of the hardware nodes; the attributes of the hardware node include node status ; available power margin ; where represents the m th hardware node, represents the maximum power margin of the hardware node; the hardware nodes are different hardware on the unmanned terminal platform, including network ports, serial ports, antennas, etc.

[0031] Resource node set , where represents one of the resource nodes; the attributes of the resource node include resource availability , channel occupancy rate ; where represents the n th resource node; the resource nodes are different resources on the unmanned terminal platform, including bandwidth, power, frequency band, etc.

[0032] Edge set , representing the association relationship between the hardware node and the resource node, characterized by the attributes of the edges in the edge set, specifically including bandwidth allocation , power consumption ; where represents the maximum bandwidth, represents the total power.

[0033] The topological network satisfies the following constraint conditions: 1) Hardware compatibility constraint:

[0034] Among them, the node state of the hardware node directly affects resource availability: , where represents the hardware node h whether it supports the resource node r .

[0035] 2) Power capacity constraint: The hardware node h and each resource node r The sum of the power consumption between them should be less than the available power margin of the hardware node : .

[0036] 3) Bandwidth allocation constraint: The sum of the bandwidth allocations h between each hardware node r and the resource node should be less than the total bandwidth : .

[0037] 4) Frequency band isolation constraint: If the resource node r is a frequency band, it is required to satisfy frequency band isolation: If the frequency bands of the i th j and r i and r j resource nodes conflict, then the corresponding resource availability: .

[0038] Step 1.3, multi-objective optimization solution.

[0039] Construct a multi-objective optimization model based on the communication ability index in Step 1.1 and the hardware resource topological network in Step 1.2. Taking maximizing the satisfaction of task requirements as the objective function and the hardware resource topological network as the constraint condition, use the multi-objective granular communication subnet to optimize and search for the Pareto solution set, and the output is N groups of candidate solutions (solutions that satisfy the constraint conditions and have the maximum objective function); the objective function is defined as follows:

[0040]

[0041] where is the fitness of the k th dimension in the communication capability index; is a candidate solution, which contains the attributes of the edges, i.e., the specific values of bandwidth allocation and power consumption; is the specific value of the communication capability index required by the candidate solution.

[0042] Step 1.4, Conflict arbitration and priority decision.

[0043] When parameter conflicts occur during the solution process of Step 1.3, such as high bandwidth requirements for high-power operation of the hardware while there are also low-power consumption requirements; when there are parameter conflicts, perform dynamic adjustment, refer to the priority order of the dimensions in the communication capability index corresponding to the task requirements to adjust the conflicts, and give priority to satisfying the dimensions with higher priorities; after adjusting the conflicts, recalculate the objective function of the candidate solution, and take the candidate solution with the maximum objective function as the final solution for output.

[0044] Through Step 1, the hardware management of the unmanned terminal platform is realized, and the hardware nodes and resource nodes are configured according to the final solution output in Step 1 (bandwidth allocation and power consumption between the hardware nodes and resource nodes), so that the state of the unmanned terminal platform better conforms to the task requirements, thereby enabling the unmanned terminal platform to better execute the tasks at the software level in Steps 2 and 3.

[0045] 2. Cluster node collaborative supervision and network topology dynamic maintenance.

[0046] Construct a communication network with the unmanned terminal platform as a node; the network mode of the communication network is determined by the task type, including a centralized mode or a distributed mode; among them, the centralized mode establishes a star topology through election, and the distributed mode constructs a Mesh network through the neighbor discovery protocol; then, based on the topological structure of the communication network, start a hierarchical time synchronization mechanism and construct a communication subnet.

[0047] For the centralized mode, the central node broadcasts and compensates and calibrates the RTT. In the distributed mode, an improved PTP protocol is used to construct a clock deviation matrix for consensus synchronization. Furthermore, a dynamic communication group is constructed according to the task type to form a data synchronization subnet. Finally, the topology is dynamically maintained through the enqueue and dequeue management mechanism, and the connection relationship is updated by using incremental Delaunay triangulation or greedy algorithm. When dealing with silent nodes, a shadow topology is constructed and group multicast synchronization is implemented to achieve real-time update of the communication network topology and maintenance of state consistency. In the communication network, the unmanned terminal platforms act as nodes, and the communication attributes between the unmanned terminal platforms act as the edges between the nodes.

[0048] Step 2.1, Network mode decision and initialization. When the unmanned terminal platform is initialized or the task requirements change, the unmanned terminal platform is used as a node to construct an initial communication network structure. The task type is obtained by parsing the task sent by the upper-layer task software, and the network mode is selected based on the task type: For the global consistency requirement, the centralized mode is adopted. Based on the node computing power and load (the usage of the current node computing power), the weight (the weight is obtained by normalizing the available computing power of all nodes), a central node is dynamically elected to construct a star topology and assign a central node identifier. For the high fault tolerance requirement, switch to the distributed mode, call the neighbor discovery protocol to establish a centerless Mesh network, and each node outputs the initial communication network structure including the node connection relationship and mode identifier to the neighbor nodes. The mode identifier is used to identify that the current network mode is the centralized mode or the distributed mode.

[0049] Global consistency requirement: For some task types, the information obtained by all unmanned terminal platforms needs to be consistent. High fault tolerance requirement: The information obtained by all unmanned terminal platforms does not need to be consistent. It is allowed that the information obtained by individual unmanned terminal platforms has deviations, and the system can still operate normally with some deviations.

[0050] Step 2.2, Hierarchical time synchronization mechanism.

[0051] Based on the communication network structure constructed in Step 2.1, the time synchronization of the nodes is started according to the network mode: In the centralized mode, the central node periodically broadcasts a reference clock signal, and the remaining nodes use the round-trip delay (RTT) compensation algorithm to calibrate the local clock step by step. In the distributed mode, each node exchanges IEEE 1588 improved PTP protocol timestamps with at least 3 neighbor nodes, constructs a clock deviation matrix, and calculates the global optimal clock reference through the Byzantine fault-tolerant consensus algorithm.

[0052] When a new node is added to the communication network structure, each node dynamically obtains the neighbor node list, and uses the window sliding average filtering and Kalman prediction algorithms to complete the convergence of the clock offset within a 10ms time window.

[0053] Step 2.3, Construction of dynamic communication subnetworks.

[0054] Based on the time synchronization results of the nodes in Step 2.2, the communication network structure performs communication grouping according to the task type to form communication subnetworks. Nodes within each communication subnetwork maintain data synchronization through the multicast protocol, and logical isolation strategies are adopted between communication subnetworks to ensure that data is only transmitted within authorized communication subnetworks.

[0055] Step 2.4, Ingress / egress queue management and topology reconstruction.

[0056] When node addition, offline, or silent events occur in each communication subnetwork, hierarchical processing is performed: First, reconstruct the topological structure of the communication subnetwork; based on the current communication subnetwork, use the incremental Delaunay triangulation algorithm to update the connection relationships of the nodes; if the triangulation fails, then enable the Kruskal greedy algorithm to construct a minimum spanning tree; Second, handle the silent state; move the silent nodes (nodes that do not participate in communication) into the shadow topology, retain their logical state information but disable the physical layer transceiver functions (retain the topological positions of the silent nodes in the communication subnetwork but do not send or receive information, so as to facilitate the communication of the silent nodes after they go online later) to avoid communication conflicts; Finally, perform state synchronization; the communication subnetwork pushes the reconstructed topological structure through the multicast protocol. After each node receives the topological incremental data packet, it uses vector clock comparison and merging to update, and outputs the real-time topological mapping table to other neighbor nodes to complete the closed-loop consistency maintenance; among them, the real-time topological mapping table contains the latest topological structure of the communication subnetwork in the current state.

[0057] 3. Dynamic information maintenance.

[0058] Implement multi-level state management of the communication network by constructing a network topology table, a status information table, and an information maintenance rule table; among them, the network topology table records the communication status of the nodes and the relationship with the communication subnetworks, the status information table stores the status information of each node, and the maintenance rule table defines the update frequency of each status information; when a node sends status information, it performs structured encapsulation and triggers encryption transmission and retransmission control based on the information maintenance rule table; when the topological structure changes, the task phase switches, or a node silent event is triggered, update the network topology table and perform synchronization to achieve global state consistency maintenance.

[0059] The dynamic information maintenance link includes the following steps: Step 3.1, Construction of multi-level status tables.

[0060] Establish three types of status management tables: a network topology table, a status information table, and an information maintenance rule table: The network topology table is used to record the unique identifier of a node (identifying the node number), the communication subnet number it belongs to, the working status, the communication link quality index, the central node identifier (identifying whether the node is a central node), and the timestamp of the most recent update of the status information.

[0061] The status information table stores the local status information with each node as an independent unit, including location information, motion parameters, and user-defined status variables; the number of status information tables corresponds one-to-one with the number of nodes; among them, the user-defined status variables support the user to customize and expand other status variables.

[0062] The information maintenance rule table defines the network topology status (the connectivity relationship between the current node and other nodes) and the update frequency of the status information of all nodes; among them, the status information includes location information, motion parameters, and user-defined status variables; based on this table, status query instructions are periodically generated or a broadcast push mechanism is triggered; among them, the status query instructions are used to query the status information of a certain node; the broadcast push mechanism is used for a node to push its own status information to other nodes.

[0063] Step 3.2, Structured encapsulation of the information to be sent.

[0064] Based on the update frequency defined in the information maintenance rule table, the current node performs protocol encapsulation on the status information to be sent to construct a data packet; among them, according to the type of status information, the unique identifier of the target node or the hardware address of the node is bound, and the reachability of the target node is verified through the network topology table; transmission policy parameters are configured, including the switch status of the acknowledgment mechanism, the data encryption flag, the sending period, and the number of retransmissions; the update frequency of the status information is synchronously updated to the information maintenance rule table to be incorporated into the global scheduling.

[0065] Step 3.3, Encrypted transmission and reliability guarantee.

[0066] Execute data transmission based on the update frequency and encapsulation policy in the information maintenance rule table: For the data packet with the encryption flag enabled, the content is encrypted using the AES-256 algorithm; unicast, multicast, or broadcast communication is triggered according to the update frequency. If the switch status of the acknowledgment mechanism is enabled, the target node needs to return an acknowledgment signal containing the hash check value within the timeout window. The current node detects the acknowledgment status and retransmits according to the exponential backoff algorithm after the timeout. The upper limit of the number of retransmissions is set to 3 times; after the transmission is completed, the corresponding content in the status information table is updated according to the transmitted status information.

[0067] Step 3.4, Dynamic synchronization and closed-loop maintenance of the status table.

[0068] When the topology structure of the communication subnet changes, the task phase switches, or a node's silent event is triggered, the network topology table is updated. After the update, each node writes its local status information into the status information table according to the update frequency in the information maintenance rule table. At the same time, a status information query request is sent to neighboring nodes through a status query instruction. After receiving the status information difference data returned by the neighboring nodes, a timestamp comparison or version number conflict detection mechanism is used to update the local status information table to ensure strong consistency of the network-wide status information.

[0069] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features. These modifications or replacements 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 application, and should all be included in the protection scope of the present application.

Claims

1. A communication management method for an integrated terminal of distributed intelligent planning, decision-making and communication, characterized in that, Including: The unmanned terminal platform quantifies the task requirements into communication capability indicators in multiple dimensions based on the task requirements and in combination with the expert knowledge base; According to the hardware and resources of the unmanned terminal platform itself, a hardware resource topology network is constructed, and on this basis, a multi-objective optimization model is constructed and solved, and finally an optimal candidate solution is output to configure the hardware and resources of the unmanned terminal platform; A communication network is constructed with the unmanned terminal platform as a node; The network mode of the communication network is determined by the task type, including a centralized mode or a distributed mode; subsequently, a hierarchical time synchronization mechanism is started based on the topological structure of the communication network, and a communication subnet is constructed; The multi-level state management of the communication network is realized by constructing a network topology table, a status information table and an information maintenance rule table; among them, the network topology table records the communication status of the nodes and the relationship with the communication subnet, the status information table stores the status information of each node, and the maintenance rule table defines the update frequency of each status information; when the node sends the status information, it is structured and encapsulated, and the encrypted transmission and retransmission control are triggered based on the information maintenance rule table; When the topological structure changes, the task phase switches, or a node silence event is triggered, the network topology table is updated and synchronized to achieve global state consistency maintenance.

2. The communication management method of the distributed intelligent planning, decision-making and communication integrated terminal according to claim 1, wherein The hardware resource topology network has two types of nodes, which include a hardware node set, a resource node set, and an edge set; Among them, the hardware node set is composed of hardware nodes, and the attributes of the hardware nodes include node status and available power margin; the resource node set is composed of resource nodes, and the attributes of the resource nodes include resource availability and channel occupancy rate; the edge set is composed of the edges between the hardware nodes and the resource nodes, and the attributes of the edges include bandwidth allocation and power consumption; The constraint conditions of the hardware resource topology network include hardware compatibility constraints, power capacity constraints, and bandwidth allocation constraints.

3. The communication management method of the distributed intelligent planning decision-making communication integrated terminal according to claim 1, characterized in that A multi-objective optimization model is constructed according to the communication capability indicators and the hardware resource topology network, with maximizing the satisfaction of task requirements as the objective function and the hardware resource topology network as the constraint condition, and the multi-objective particle communication subnet optimization search Pareto solution set is used to output multiple groups of candidate solutions; the objective function is defined as follows: Among them is the fitness of the k th dimension in the communication capability index; is a candidate solution, which includes the attributes of the edges, namely the specific values of bandwidth allocation and power consumption; is the specific value of the communication capability index required by the candidate solution.

4. The communication management method of the distributed intelligent planning, decision-making and communication integrated terminal according to claim 1, characterized in that, The network mode of the communication network is determined by the task type, where: When the task type requires global consistency, the centralized mode is adopted, and the central node is generated by dynamic election based on the node computing power and load weight, and a star topology is constructed and the central node identifier is assigned; When the task type requires high fault tolerance, it switches to the distributed mode, and the neighbor discovery protocol is called to establish a centerless Mesh network, and each node outputs the initial communication network structure including the node connection relationship and the mode identifier to the neighbor nodes.

5. The communication management method of the distributed intelligent planning, decision-making and communication integrated terminal according to claim 1, characterized in that The hierarchical synchronization mechanism is as follows: In the centralized mode, the central node of the communication network periodically broadcasts a reference clock signal, and the remaining nodes use the round-trip delay compensation algorithm to calibrate their local clocks step by step; in the distributed mode, each node exchanges timestamps with multiple neighbor nodes, constructs a clock deviation matrix, and calculates the global optimal clock reference through the Byzantine fault-tolerant consensus algorithm; when a new node joins the communication network, each node dynamically obtains the neighbor node list and uses the window sliding average filtering and Kalman prediction algorithms to complete the convergence of the clock offset.

6. The communication management method for the distributed intelligent planning, decision-making, and communication integrated terminal according to claim 1, characterized in that, The communication network structure performs communication grouping according to the task type to form communication subnets. Nodes within each communication subnet maintain data synchronization through the multicast protocol, and logical isolation strategies are adopted between communication subnets. When there are node addition, offline, or silent event triggers in each communication subnet, hierarchical processing is performed: First, the topological structure of the communication subnet is reconstructed; based on the current communication subnet, the incremental Delaunay triangulation algorithm is used to update the connection relationship of the nodes; if the triangulation fails, the Kruskal greedy algorithm is enabled to construct a minimum spanning tree. Secondly, the silent state is processed. The silent nodes are moved into the shadow topology, and their logical state information is retained while the physical layer transceiver functions are disabled. Finally, state synchronization is performed. The communication subnet pushes the reconstructed topological structure through the multicast protocol. After each node receives the topological incremental data packet, it uses vector clock comparison to merge and update, and outputs the real-time topological mapping table to other neighbor nodes to complete the closed-loop consistency maintenance.

7. The communication management method of the distributed intelligent planning, decision-making and communication integrated terminal according to claim 1, characterized in that The network topology table is used to record the unique identifier of the node, the communication subnet number it belongs to, the working state, the communication link quality index, the central node identifier, and the most recent update timestamp of the data. The status information table stores the local status information with the node as an independent unit, including location information, motion parameters, and user-defined status variables. The information maintenance rule table defines the update frequencies of the network topology status and the status information of all nodes. Among them, the status information includes location information, motion parameters, and user-defined status variables; based on this table, status query instructions are periodically generated or the broadcast push mechanism is triggered; among them, the status query instructions are used to query the status information of a certain node; the broadcast push mechanism is used for a node to push its own status information to other nodes.

8. The communication management method of the distributed intelligent planning, decision-making and communication integrated terminal according to claim 1, wherein Based on the update frequencies defined in the information maintenance rule table, the current node performs protocol encapsulation on the status information to be sent to construct a data packet; among them, the unique identifier of the target node or the hardware address of the node is bound according to the type of the status information, and the reachability of the target node is verified through the network topology table; the transmission policy parameters are configured, including the switch state of the acknowledgment mechanism, the data encryption flag, the sending period, and the number of repeated transmissions; the update frequency of the status information is synchronously updated to the information maintenance rule table to be incorporated into the global scheduling.

9. The communication management method of the distributed intelligent planning, decision-making and communication integrated terminal according to claim 1, characterized in that Data transmission is performed based on the update frequencies and encapsulation strategies in the information maintenance rule table. For data packets with the encryption flag enabled, encryption is performed using an encryption algorithm; unicast, multicast, or broadcast communication is triggered according to the update frequency. If the acknowledgment mechanism switch is enabled, the target node needs to return an acknowledgment signal containing a hash check value within the timeout window. The current node detects the acknowledgment status and retransmits according to the exponential backoff algorithm after the timeout. After the transmission is completed, the corresponding content in the status information table is updated according to the transmitted status information.

10. An unmanned terminal platform, characterized in that, The distributed intelligent planning decision-making communication integrated terminal communication management method according to any one of claims 1-9 is deployed on this platform.

Images