Intelligent unmanned combat system

Abstract

An intelligent unmanned combat system comprises an unmanned aerial vehicle, a remote control aircraft, an unmanned vehicle, a remote control vehicle, a robot, a robot dog, a water surface unmanned ship and an underwater unmanned underwater vehicle, and computers carried by the unmanned aerial vehicle, the remote control aircraft, the unmanned vehicle, the robot dog, the water surface unmanned ship and the underwater unmanned underwater vehicle form an unmanned combat equipment edge calculation aggregate. Comprising the following steps: step 1, an unmanned combat equipment edge calculation aggregate performs intelligent unmanned reconnaissance according to a user combat instruction, and identifies an own defense target and an opposite-side attack target; step 2, the unmanned combat equipment edge calculation aggregate performs intelligent unmanned combat according to the unmanned reconnaissance data, protects an own defense target and attacks a target attacked by an opposite side; and step 3, performing intelligent unmanned combat damage evaluation on the unmanned combat equipment edge calculation aggregate in an unmanned combat process, and determining whether the state of the own defense target and the state of the opposite-side attack target meet a user combat instruction or not. According to the invention, an intelligent unmanned combat mode can be provided for various unmanned combat equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, especially to the fields of robots and artificial intelligence, and particularly to an unmanned combat equipment edge computing aggregate. BACKGROUND

[0002] With the progress of science and technology, the killing efficiency of various weapons has also made great progress. In recent local wars, a large number of soldiers on both sides have been easily killed and wounded. How to reduce the casualties of combat officers and soldiers is a problem that must be considered in modern warfare.

[0003] With the rapid development of computer hardware and intelligent technology, the intelligent combat of unmanned aircraft aggregate will become the main combat mode of future battlefield. Intelligent unmanned combat systems are systems that can perform combat tasks without or with limited human intervention, using artificial intelligence, autonomous control, sensors, and network communication technologies. They cover multiple fields such as air, ground, sea, and underwater, and have the ability of autonomous decision-making, target identification, and cooperative combat. CN118838419A discloses a task allocation method suitable for heterogeneous multi-unmanned platform, CN119416604A discloses an intelligent decision-making system for unmanned combat equipment based on reinforcement learning, and CN117650988A discloses a command and control structure modeling and evaluation method for aggregate unmanned combat scenarios.

[0004] As a multi-agent system, the traditional single-agent reinforcement learning algorithm is no longer applicable to unmanned aircraft aggregate, and multi-agent reinforcement learning algorithm has become the mainstream. Artificial intelligence technology, especially deep learning, has made significant progress in target identification and classification. Unmanned combat systems can use algorithms to achieve autonomous navigation and dynamic path planning, avoid obstacles and choose the optimal path in complex environments, achieve aggregate cooperative combat, autonomously allocate tasks and complete complex tasks. At the same time, AI technology enables unmanned combat systems to interact with operators through natural language processing and voice interaction, improving operational efficiency. Artificial intelligence technology has made significant progress in intelligent unmanned combat systems, covering target identification, autonomous navigation, threat assessment, aggregate cooperation, voice interaction, predictive maintenance, and electronic warfare.

[0005] Intelligent unmanned combat systems, as an important development direction in the future, have many significant advantages, but also have the following disadvantages: 1) Lack of intention and value judgment ability. In complex war scenarios, when moral and ethical decisions need to be made, unmanned combat systems may not be able to make the right choice; 2) Vulnerable to interference and attack. The enemy can disrupt the communication link of the unmanned combat platform through electronic interference means, causing it to lose control or fail to perform its tasks normally. In addition, the software and hardware of the unmanned combat system may also have vulnerabilities that can be exploited by the enemy for attack; 3) Risk of overfitting. The decision of the intelligent unmanned combat system usually relies on a large amount of data and algorithm training. However, in actual combat, the situation may be different from the training data, resulting in the risk of overfitting; 4) Difficulty in maintenance and support. The intelligent unmanned combat system usually has a complex technical structure, which requires professional technical personnel and equipment for maintenance and support. Moreover, as the number of unmanned combat systems may be large, the workload of maintenance and support will also increase accordingly; 5) Strategic and tactical limitations. The energy, materials, and communication of the intelligent unmanned combat system are put forward higher requirements in extreme environments. SUMMARY

[0006] The purpose of the present application is to provide an intelligent unmanned combat mode for various unmanned combat equipment, reduce the risk of personnel casualties, improve combat efficiency, handle multiple tasks in parallel, enhance the accuracy and reliability of combat, adapt to complex combat environments, and achieve distributed combat and cooperative combat.

[0007] To solve the above technical problems, the technical solution adopted by the present application is: An intelligent unmanned combat system using unmanned combat equipment for reconnaissance, combat, and damage assessment, utilizing the computers carried by each to form an unmanned combat equipment edge computing cluster; the system, when in operation, includes the following steps: Step 1: The unmanned combat equipment edge computing cluster conducts intelligent unmanned reconnaissance according to user combat instructions, identifying friendly defense targets and enemy attack targets; Step 2: The unmanned combat equipment edge computing cluster conducts intelligent unmanned combat based on unmanned reconnaissance data, protecting friendly defense targets and attacking enemy attack targets; Step 3: The unmanned combat equipment edge computing cluster conducts intelligent unmanned damage assessment during the unmanned combat process, determining whether the states of friendly defense targets and enemy attack targets meet the user combat instructions; In Step 1, the following sub-steps are included: Sub-step 1-1: The unmanned combat equipment edge computing cluster accepts user combat instructions; the unmanned combat equipment edge computing cluster calculates strategic reconnaissance tasks according to user combat instructions, including strategic reconnaissance areas, reconnaissance means, reconnaissance times, reconnaissance intensities, reconnaissance accuracies, specific reconnaissance targets, available reconnaissance tools, and reconnaissance forces; preferably, the intelligent unmanned reconnaissance uses unmanned reconnaissance tools, utilizing the computers carried by each to form an unmanned combat equipment edge computing cluster to calculate reconnaissance tasks; Sub-step 1-2, the unmanned combat equipment edge computing collective intelligence allocates reconnaissance tasks; the unmanned combat equipment edge computing collective selects available intelligent unmanned reconnaissance tools according to the user combat order, and uses the intelligent unmanned reconnaissance tool embedded edge computing platform and artificial intelligence technology to decompose the strategic reconnaissance task into a tactical reconnaissance task; the decomposition target includes the specific tactical reconnaissance area covered by the user combat order, the reconnaissance means, the reconnaissance time, the reconnaissance intensity, the reconnaissance accuracy, the specific reconnaissance target, and the available reconnaissance tools and reconnaissance forces; further, the decomposed tactical reconnaissance task is allocated to the appropriate unmanned reconnaissance tool; Sub-step 1-3, the unmanned combat equipment edge computing collective intelligence executes the reconnaissance task and identifies the own defense target and the opponent attack target; the unmanned combat equipment edge computing collective cooperates with each unmanned reconnaissance tool to execute the reconnaissance task according to the decomposed strategic and tactical reconnaissance task, and runs an artificial intelligence algorithm in the execution process to identify the own defense target and the opponent attack target to cope with the changing situation on the battlefield.

[0008] In step 1-1, the following sub-steps are adopted: Sub-step 1-1-1, initialization of an intelligent reconnaissance system; Preferably, an intelligent reconnaissance system is initialized, which includes one unmanned reconnaissance tool, a reconnaissance task number allocated to the unmanned reconnaissance tool; for each unmanned reconnaissance tool a state space , a Markov chain is used to represent: ; When the initialization is , it indicates that the unmanned reconnaissance tool is in an unallocated task state; Sub-step 1-1-2, the unmanned combat equipment edge computing collective calculates a strategic reconnaissance task according to the user combat order; The unmanned combat equipment edge computing collective obtains the user combat order, broadcasts the user combat order, requests the aggregation of unmanned reconnaissance tools, and the aggregated unmanned combat equipment edge computing collective calculates a strategic reconnaissance task that meets the user combat order according to the user combat order, including a strategic reconnaissance area, a reconnaissance means, a reconnaissance time, a reconnaissance intensity, a reconnaissance accuracy, and a specific reconnaissance target, and calculates available reconnaissance tools and reconnaissance forces; Sub-step 1-1-3, the unmanned combat equipment edge computing collective allocates available unmanned reconnaissance tools according to the strategic reconnaissance task; In the strategic reconnaissance mission of the edge computing aggregate broadcast sub-step 1-1-2 for unmanned combat equipment, the unmanned reconnaissance tools that receive the reconnaissance mission broadcast aggregate through wireless networking. Although geographically dispersed, each individual tool can exchange information with other members of the aggregate to obtain relevant information about the broadcast strategic reconnaissance mission. The probability of joining the strategic reconnaissance mission is calculated based on the unmanned reconnaissance tool's reconnaissance capabilities, reconnaissance range, and current location. Next, the edge computing aggregate of the unmanned combat equipment generates random numbers and interacts with them. Comparisons are made, and decisions are made based on the comparison results to include unmanned reconnaissance tools in strategic reconnaissance missions until a sufficient number of unmanned reconnaissance tools that can meet the requirements of strategic reconnaissance missions are allocated. In steps 1-2, the following sub-steps are used: Sub-step 1-2-1: Construct an edge computing cluster of unmanned combat equipment based on the strategic reconnaissance mission; Unmanned reconnaissance vehicles that receive strategic reconnaissance mission broadcasts gather via wireless networking to form an edge computing cluster of unmanned combat equipment. Sub-step 1-2-2: The strategic reconnaissance mission is broken down into several tactical reconnaissance missions; The edge computing cluster of unmanned combat vehicles divides the strategic reconnaissance area into several tactical reconnaissance areas, utilizing a radius of... The reconnaissance area is described by a two-dimensional circle; the unmanned reconnaissance vehicle is considered as a circle with a radius of... If the distance between the two centers is less than Therefore, it can be assumed that unmanned reconnaissance vehicles have unlimited communication and positioning capabilities; adjacent unmanned reconnaissance vehicles can quickly exchange information and determine each other's positions; Based on the strategic reconnaissance area, strategic reconnaissance missions are broken down into several sets of tactical reconnaissance missions according to different areas. Each tactical reconnaissance mission corresponds to a specific reconnaissance area, including the detailed coordinates of that area. The set of all tactical reconnaissance missions should cover the entire strategic reconnaissance area and have a moderate overlap with each other. Sub-steps 1-2-3 assign tactical reconnaissance tasks to appropriate unmanned reconnaissance vehicles; The tactical reconnaissance mission of the unmanned combat vehicle edge computing aggregate broadcast sub-step 1-2-2 is based on the reconnaissance area, reconnaissance methods, reconnaissance time, reconnaissance intensity, reconnaissance accuracy, specific reconnaissance targets, available reconnaissance tools and reconnaissance forces. According to the reconnaissance capabilities, reconnaissance range and current location of the unmanned reconnaissance vehicle, it is assigned tactical reconnaissance tasks it can perform, and the calculations are performed for each unmanned reconnaissance vehicle. state space It is represented using a Markov chain: ; at this time This indicates that the unmanned reconnaissance vehicle is in a state of assigned mission, and the corresponding number in the set is the assigned reconnaissance mission number; Sub-steps 1-2-4: Unmanned reconnaissance tools acquire the reconnaissance area for tactical reconnaissance missions; First, the unmanned reconnaissance vehicle obtains the reconnaissance area corresponding to the assigned tactical reconnaissance mission number, including the detailed geographical coordinates of that area; when the distance between the unmanned reconnaissance vehicle and the circular area exceeds... Let be the latitude and longitude coordinates of the unmanned reconnaissance vehicle reaching the area boundary; assume that the unmanned reconnaissance vehicle can obtain its current absolute latitude and longitude coordinates based on GPS signals and calculate its new direction of motion by generating a new turning angle; assume that the unmanned reconnaissance vehicle does not need to experience acceleration and deceleration phases, avoid communication conflicts, or have other hardware limitations during its movement, and that there are no issues such as turning errors. Secondly, the unmanned reconnaissance tool uses a greedy algorithm to plan tactical reconnaissance routes, that is, it prioritizes visiting the tactical reconnaissance missions closest to its own position in the reconnaissance area, and then visits the tactical reconnaissance missions farther away from itself, until all tactical reconnaissance missions have been visited. If the distance between unmanned reconnaissance vehicles is less than 1.2 times that of the unmanned reconnaissance vehicle and without receiving aggregating requests, the edge computing platform dynamically adjusts the tactical reconnaissance mission, runs a path planning algorithm on the unmanned reconnaissance vehicle, and the unmanned reconnaissance vehicle will trigger an obstacle avoidance mechanism, causing adjacent unmanned reconnaissance vehicles to generate new movement directions and paths; once the unmanned reconnaissance vehicle enters the communication range allocated by the edge computing aggregator of unmanned combat equipment, it will be able to obtain all the information; Furthermore, the edge computing cluster of unmanned combat equipment can set up high-density and high-frequency reconnaissance missions in important strategic or tactical reconnaissance areas. That is, multiple unmanned reconnaissance tools need to be densely clustered in a certain important area for repeated reconnaissance and collaborative work. When multiple unmanned reconnaissance tools enter an important reconnaissance area, they constitute a specific edge computing cluster of unmanned combat equipment in that area. The unmanned reconnaissance tools within the cluster can communicate with each other, share their respective reconnaissance information and status, and the edge computing cluster of unmanned combat equipment can continuously optimize the cluster's reconnaissance missions until the high-density and high-frequency reconnaissance missions can meet the requirements of unmanned combat.

[0009] In steps 1-3, the following sub-steps are used: Sub-step 1-3-1: Initialization of the unmanned reconnaissance tool cluster; Assuming the edge computing aggregate of unmanned combat equipment is composed of It consists of several unmanned reconnaissance vehicles, which contain An aggregate; The probability of an unmanned reconnaissance vehicle joining a cluster is given by the following formula: ; Generally speaking, when the number of group members does not exceed At the threshold, The value is greater than zero, otherwise it is equal to zero; where, Refers to unmanned reconnaissance tools In time The state space of time, Refers to unmanned reconnaissance tools In time The state space of time; Let $\frac{ ... ; Sub-step 1-3-2: Unmanned reconnaissance vehicles conduct patrol reconnaissance and collect reconnaissance data for the assigned tactical reconnaissance mission; The unmanned reconnaissance vehicle conducts patrol reconnaissance according to the tactical reconnaissance tasks assigned in sub-step 1-2-3 and the reconnaissance area in sub-step 1-2-4, ensuring that the patrol reconnaissance route covers the entire reconnaissance area, and uses cameras, infrared imaging equipment, radar, and listening equipment to collect reconnaissance data in the reconnaissance area during the patrol reconnaissance. Within a cluster, unmanned reconnaissance vehicles can obtain information about the number of unmanned reconnaissance vehicles in the cluster, their identification numbers, and probability parameters through local information exchange with neighboring unmanned reconnaissance vehicles. ; The constant represents the "willingness" of unmanned reconnaissance vehicles to leave the cluster; Furthermore, the unmanned reconnaissance tools in the cluster determine every second whether to leave the cluster and share and forward the reconnaissance data, thereby transmitting it back to the data center of their own unmanned combat system. Sub-step 1-3-3: The unmanned reconnaissance vehicle returns due to insufficient power or malfunction and the unfinished reconnaissance mission is handed over; When an unmanned reconnaissance vehicle detects low power or a malfunction, it reserves power for a return trip. If the reconnaissance mission is complete, it returns directly; if the mission is incomplete, it exchanges information with other unmanned reconnaissance vehicles in the cluster to obtain information about nearby available unmanned reconnaissance vehicles, and then calculates the probability of a nearby available unmanned reconnaissance vehicle joining the cluster. Furthermore, nearby unmanned reconnaissance vehicles need to calculate the probability of handing over unfinished reconnaissance tasks based on their remaining power, their own mission completion status, reconnaissance capabilities, reconnaissance range, and current location, and make a decision to join the cluster based on the calculation results. If any unmanned reconnaissance vehicle within the sensing range does not join the cluster, it is prohibited from joining the cluster for 20 seconds. The edge computing system of the unmanned combat equipment then dispatches a suitable, fully powered unmanned reconnaissance vehicle to take over the unfinished reconnaissance mission, ensuring... The effectiveness of unmanned reconnaissance tools joining the cluster. The probability is: ; For the current moment, the aggregate Number of members; For aggregates The threshold; It is the largest aggregate that the system can form; When there are multiple aggregates in a group system This ensures fair competition among the various clusters, meaning that unmanned reconnaissance tools tend to join clusters with higher thresholds in order to maintain the balance among the clusters. Sub-steps 1-3-4: Enhanced reconnaissance of important targets; For important areas or targets, relying on a single unmanned reconnaissance vehicle is far from sufficient. Often, two or more unmanned reconnaissance vehicles are needed to conduct continuous and intensive reconnaissance of the area from multiple angles and for extended periods, using various reconnaissance methods such as cameras, infrared imaging equipment, radar, and listening devices. In unit time Within the area, the probability of any two unmanned reconnaissance vehicles meeting is The probability of each unmanned reconnaissance vehicle encountering any other unmanned reconnaissance vehicle is... Therefore: ; The area of ​​the enclosed region; The average speed of the unmanned reconnaissance vehicle. The sensing radius of unmanned reconnaissance tools; for The area swept by unmanned reconnaissance vehicles within a given time period; The Markov process of the unit unmanned reconnaissance tool constitutes the dynamic aggregation process of the edge computing aggregate of unmanned combat equipment; by utilizing the probability parameters of the unmanned reconnaissance tool, we can obtain the average number of members in each aggregate at any time point; the change in the average number of members in the aggregate is affected by other aggregates in the system and state transition parameters. Sub-steps 1-3-5: Optimization of the distribution of unmanned reconnaissance tools within the cluster; At any given time, the distribution of unmanned reconnaissance tools within each cluster of the unmanned combat equipment edge computing aggregate can be described by difference equations: ; Represents unmanned reconnaissance tools encountering clusters within a unit of time. and the average number added; The number of unmanned reconnaissance vehicles representing the roaming status per unit time; This indicates that unmanned reconnaissance vehicles have joined the cluster. The probability of; Indicates an aggregate The initial state; This indicates the aggregate per unit time. The average number of members leaving the cluster; When the system reaches a steady state, we can obtain:

[0010] By substituting the parameters of the unmanned reconnaissance tool into the difference equation, the ratio of member quotas of any two aggregates when the system is in equilibrium can be calculated. ; Sub-steps 1-3-6 identify friendly defensive targets and enemy attack targets; The edge computing aggregate of unmanned combat equipment calculates the friendly defense zone and the enemy's strike zone based on user combat commands and reconnaissance data from sub-step 1-3-2. It runs artificial intelligence algorithms to identify targets in the friendly defense zone and the enemy's strike zone from reconnaissance data collected by cameras, infrared imaging equipment, radar, and listening devices, as well as targets approaching or intruding into the friendly defense zone. This includes infrared signals from combat personnel, equipment, and military dogs; images and radar signals from firepower configurations, field fortifications, buildings, landmines, camouflage, bridges, unmanned combat equipment, and suspicious sound signals; and calculates the identification probability of friendly defense targets and enemy strike targets. First, the identification probabilities are compared; targets with higher probabilities have higher identification rates and better reconnaissance results. Second, priorities are calculated based on the size and threat level of the identified targets; targets with higher priorities are more important. Finally, it checks whether all targets for all tactical reconnaissance missions have been identified. If so, target identification terminates; otherwise, target identification is performed sequentially, with supplementary reconnaissance conducted on targets with low identification probabilities that are difficult to identify. Sub-steps 1-3-7: Virtual reconnaissance exercise to confirm and optimize the reconnaissance mission; Furthermore, the edge computing aggregate of unmanned combat equipment can perform virtual reconnaissance exercises, verifying the rationality and feasibility of reconnaissance task allocation through online computation, and identifying and correcting errors and deviations before executing reconnaissance tasks. To test the system's stability, accuracy, and scalability, the edge computing aggregate of unmanned combat equipment can select different model parameters for calculation and comparative analysis, and can also adjust parameters accordingly. To analyze the relationship between time and the number of crew members in unmanned reconnaissance vehicles; when the number of unmanned reconnaissance vehicles is AND equals the cluster threshold Adjusting model parameters while keeping the number of simulations constant. The study investigated the trend of the average number of clusters over time; and examined the changes in the number of unmanned reconnaissance vehicles in the system. And adjust the aggregate threshold accordingly. This allows us to obtain the relationship between the average number of clusters and time under different numbers of unmanned reconnaissance vehicles; Furthermore, the edge computing aggregate of unmanned combat vehicles can optimize the execution of reconnaissance missions on the edge computing aggregate, that is, to calculate and optimize the aggregation and balance of multiple unmanned reconnaissance vehicles online, i.e., to optimize the arrangement within the work area. The unmanned reconnaissance vehicles were reassigned, and aggregation requests were initiated when the time threshold for completing the reconnaissance mission was more optimized; this was achieved by modifying the threshold and model parameters. To conduct simulation experiments, namely virtual reconnaissance exercises, can optimize the average size and quota ratio of the aggregate at different times before the reconnaissance mission is completed, better accelerate the group convergence process of multiple unmanned reconnaissance tools, and provide decision-making basis for the execution prediction of unmanned reconnaissance missions, the group convergence characteristics of unmanned reconnaissance tools, and the determination of the parameters of the group system. Sub-steps 1-3-8 complete all tactical and strategic reconnaissance missions; Each unmanned reconnaissance vehicle checks the execution status of its reconnaissance mission according to the task assignment results. After completing all reconnaissance missions, it reports the reconnaissance data and mission completion status to the unmanned combat equipment edge computing aggregate. If there are no new missions, it returns to its own base. If the unmanned combat equipment edge computing aggregate assigns a new reconnaissance mission, it skips to step 1 to execute the new reconnaissance mission. If the unmanned combat equipment edge computing aggregate assigns a new combat mission, it skips to step 2 to execute the new combat mission. If the unmanned combat equipment edge computing aggregate assigns a new battle results and damage assessment mission, it skips to step 3 to execute the new battle results and damage assessment mission.

[0011] Compared with the prior art, the present invention has the following technical effects: 1) Reduce casualties: The intelligent unmanned system described in this invention uses its own computers to form an edge computing aggregate of unmanned combat equipment; this invention can replace soldiers in performing high-risk tasks, such as reconnaissance behind enemy lines, front-line combat, battle results and damage assessment, mine clearance, and operations in nuclear, biological and chemical contaminated areas, significantly reducing the risk of casualties among combat personnel.

[0012] 2) High Efficiency and Rapid Response: Edge computing clusters of unmanned combat equipment process battlefield data in real time through AI algorithms, such as target identification and path planning, with decision-making speeds far exceeding those of humans. Furthermore, they are unaffected by physiological limitations, capable of executing missions 24 / 7 without rest, making them suitable for protracted or high-intensity warfare. Simultaneously, AI technology enables unmanned combat systems to interact with operators via voice through natural language processing, improving operational efficiency.

[0013] 3) Precision Strike and Reduced Collateral Damage: Unmanned combat vehicles (UCVs) employ an edge computing cluster system that integrates reconnaissance, combat, and damage assessment for integrated operations. Before deploying for missions, they check if the operational area has been thoroughly reconnoitered; if not, supplementary reconnaissance is conducted to ensure a true understanding of both the enemy and the objective. Furthermore, the use of laser guidance and infrared imaging technologies by unmanned reconnaissance and combat vehicles significantly reduces the probability of collateral damage to civilians and non-military targets. The UCV edge computing cluster can identify and classify targets in real time using data from multimodal sensors, including optical, infrared, and radar sensors.

[0014] 4) Collaborative Combat Capability: The edge computing aggregate of unmanned combat equipment can utilize algorithms to achieve autonomous navigation and dynamic path planning, enabling it to avoid obstacles and select the optimal path in complex environments, achieving collaborative combat, autonomously allocating tasks, and completing complex missions. The edge computing aggregate of unmanned combat equipment described in this invention represents an important research direction in the current military field, and it can solve many problems faced by traditional manned combat methods.

[0015] 5) Cost-effectiveness and scalability: The edge computing aggregate of the unmanned combat equipment utilizes an embeddable computing platform on the unmanned combat vehicle, resulting in low computing costs, good scalability, and rapid payload replacement to adapt to diverse mission requirements. This invention promotes the deep application of artificial intelligence combat systems in modern military applications, significantly improving combat efficiency and accuracy. It has wide applications in reconnaissance and surveillance, target identification and strike, decision support, cyber warfare and electronic warfare, logistics and support, training simulation, and collaborative operations, and is of great significance for future development. Attached Figure Description

[0016] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a system flowchart of the present invention. Detailed Implementation

[0017] like Figure 1 As shown, an intelligent unmanned combat system uses unmanned combat equipment for reconnaissance, combat, and assessment of combat results and damage, including the use of their respective onboard computers to form an edge computing aggregate of unmanned combat equipment; When the system is working, it includes the following steps: Step 1: The edge computing aggregate of unmanned combat equipment conducts intelligent unmanned reconnaissance based on user combat instructions, identifying friendly defensive targets and enemy attack targets; Sub-step 1-1: The unmanned combat equipment edge computing aggregate receives user combat commands; the unmanned combat equipment edge computing aggregate calculates strategic reconnaissance missions based on user combat commands, including strategic reconnaissance areas, reconnaissance methods, reconnaissance time, reconnaissance intensity, reconnaissance accuracy, specific reconnaissance targets, available reconnaissance tools and reconnaissance forces, etc.; preferably, intelligent unmanned reconnaissance uses unmanned reconnaissance tools and utilizes their respective onboard computers to form an unmanned combat equipment edge computing aggregate to calculate reconnaissance missions. Sub-step 1-1-1: Initialize the intelligent reconnaissance system; Preferably, an intelligent reconnaissance system is initialized, comprising: An unmanned reconnaissance tool, This indicates the reconnaissance mission number assigned to the unmanned reconnaissance vehicle. For each unmanned reconnaissance vehicle... state space It is represented using a Markov chain: ; During initialization At this time, it indicates that the unmanned reconnaissance vehicle is in a state of unassigned tasks.

[0018] Sub-step 1-1-2: The edge computing aggregate of unmanned combat equipment calculates strategic reconnaissance missions based on user combat instructions; The edge computing aggregate of unmanned combat equipment acquires user combat instructions and broadcasts them, requesting unmanned reconnaissance tools to gather. The gathered unmanned combat equipment edge computing aggregate calculates the user combat instructions by geographical area into a strategic reconnaissance mission that satisfies the user combat instructions, including strategic reconnaissance area, reconnaissance methods, reconnaissance time, reconnaissance intensity, reconnaissance accuracy, specific reconnaissance targets, and calculates the available reconnaissance tools and reconnaissance forces.

[0019] Sub-step 1-1-3: The edge computing aggregate of unmanned combat equipment allocates available unmanned reconnaissance tools according to the strategic reconnaissance mission; In the strategic reconnaissance mission of the edge computing aggregate broadcast sub-step 1-1-2 for unmanned combat equipment, the unmanned reconnaissance tools that receive the reconnaissance mission broadcast aggregate through wireless networking. Although geographically dispersed, each individual tool can exchange information with other members of the aggregate to obtain relevant information about the broadcast strategic reconnaissance mission. The probability of joining the strategic reconnaissance mission is calculated based on the unmanned reconnaissance tool's reconnaissance capabilities, reconnaissance range, and current location. Next, the edge computing aggregate of the unmanned combat equipment generates random numbers and interacts with them. The comparison is made, and a decision is made to include unmanned reconnaissance tools in strategic reconnaissance missions based on the comparison results, until a sufficient number of unmanned reconnaissance tools that can meet the strategic reconnaissance mission requirements are allocated.

[0020] Sub-steps 1-2: The edge computing aggregate of unmanned combat equipment intelligently allocates reconnaissance tasks; based on user combat instructions, the edge computing aggregate of unmanned combat equipment selects available intelligent unmanned reconnaissance tools, and uses the embedded edge computing platform and artificial intelligence technology of the intelligent unmanned reconnaissance tools to decompose strategic reconnaissance tasks into tactical reconnaissance tasks; the decomposed targets include the specific tactical reconnaissance area covered by the user combat instructions, reconnaissance methods, reconnaissance time, reconnaissance intensity, reconnaissance accuracy, specific reconnaissance targets, available reconnaissance tools and reconnaissance forces, etc.; further, the decomposed tactical reconnaissance tasks are allocated to appropriate unmanned reconnaissance tools; Sub-step 1-2-1: Construct an edge computing cluster of unmanned combat equipment based on the strategic reconnaissance mission; Unmanned reconnaissance vehicles that receive strategic reconnaissance mission broadcasts gather via wireless networking to form an edge computing cluster of unmanned combat equipment. Sub-step 1-2-2: The strategic reconnaissance mission is broken down into several tactical reconnaissance missions; The edge computing cluster of unmanned combat vehicles divides the strategic reconnaissance area into several tactical reconnaissance areas, utilizing a radius of... The reconnaissance area is described by a two-dimensional circle. The unmanned reconnaissance vehicle is considered as a circle with a radius of... If the distance between the two centers is less than We can then assume that unmanned reconnaissance vehicles possess unlimited communication and positioning capabilities. Adjacent unmanned reconnaissance vehicles can quickly exchange information and determine each other's positions.

[0021] Based on the strategic reconnaissance area, strategic reconnaissance missions are broken down into several sets of tactical reconnaissance missions according to different areas. Each tactical reconnaissance mission corresponds to a specific reconnaissance area, including the detailed coordinates of that area. The set of all tactical reconnaissance missions should cover the entire strategic reconnaissance area and have a moderate overlap with each other. Sub-steps 1-2-3 assign tactical reconnaissance tasks to appropriate unmanned reconnaissance vehicles; The tactical reconnaissance mission of the unmanned combat vehicle edge computing aggregate broadcast sub-step 1-2-2 is based on the reconnaissance area, reconnaissance methods, reconnaissance time, reconnaissance intensity, reconnaissance accuracy, specific reconnaissance targets, available reconnaissance tools and reconnaissance forces. According to the reconnaissance capabilities, reconnaissance range and current location of the unmanned reconnaissance vehicle, it is assigned tactical reconnaissance tasks it can perform, and the calculations are performed for each unmanned reconnaissance vehicle. state space It is represented using a Markov chain: ; at this time This indicates that the unmanned reconnaissance tool is in a state of being assigned a mission, and the corresponding number in the set is the assigned reconnaissance mission number.

[0022] Sub-steps 1-2-4: Unmanned reconnaissance tools acquire the reconnaissance area for tactical reconnaissance missions; First, the unmanned reconnaissance vehicle obtains the reconnaissance area corresponding to the assigned tactical reconnaissance mission number, including the detailed geographical coordinates of that area. When the distance between the unmanned reconnaissance vehicle and the circular area exceeds... Let be the latitude and longitude coordinates of the area reached by the unmanned reconnaissance vehicle. Assume the unmanned reconnaissance vehicle can obtain its current absolute latitude and longitude coordinates based on GPS signals and calculate its new direction of motion by generating a new turning angle. Assume the unmanned reconnaissance vehicle does not need to undergo acceleration / deceleration phases, avoid communication conflicts, or suffer from steering errors during its movement.

[0023] Secondly, the unmanned reconnaissance tool uses a greedy algorithm to plan tactical reconnaissance routes, that is, it prioritizes visiting the tactical reconnaissance missions closest to its own position in the reconnaissance area, and then visits the tactical reconnaissance missions farther away, until all tactical reconnaissance missions have been visited.

[0024] If the distance between unmanned reconnaissance vehicles is less than If the distance is 1.2 times greater and no aggregation request is received, the edge computing platform dynamically adjusts the tactical reconnaissance mission, runs a path planning algorithm on the unmanned reconnaissance vehicle, and triggers an obstacle avoidance mechanism, causing adjacent unmanned reconnaissance vehicles to generate new directions of movement and paths. Once the unmanned reconnaissance vehicle enters the communication range allocated by the edge computing aggregation of unmanned combat equipment, it will be able to obtain all the information.

[0025] Furthermore, edge computing clusters of unmanned combat vehicles (UCVs) can set up high-density, high-frequency reconnaissance missions in important strategic or tactical reconnaissance areas. This means multiple UCVs need to be densely clustered in a specific area for repeated reconnaissance and collaborative work. When multiple UCVs enter an important reconnaissance area, they constitute a specific UCV edge computing cluster for that area. The UCVs within this cluster can communicate with each other, sharing their reconnaissance information and status. The UCV edge computing cluster can continuously optimize its reconnaissance missions until the high-density and high-frequency reconnaissance missions meet the requirements of unmanned combat operations.

[0026] Sub-steps 1-3: The edge computing aggregate of unmanned combat equipment intelligently performs reconnaissance missions, identifying friendly defensive targets and enemy strike targets; the edge computing aggregate of unmanned combat equipment, according to the decomposed strategic and tactical reconnaissance missions, is coordinated by various unmanned reconnaissance tools to perform reconnaissance missions, and runs artificial intelligence algorithms during the execution process to identify friendly defensive targets and enemy strike targets in order to cope with the ever-changing situation on the battlefield. Sub-step 1-3-1: Initialization of the unmanned reconnaissance tool cluster; Assuming the edge computing aggregate of unmanned combat equipment is composed of It consists of several unmanned reconnaissance vehicles, which contain An aggregate. The probability of an unmanned reconnaissance vehicle joining a cluster is given by the following formula: ; Generally speaking, when the number of group members does not exceed At the threshold, The value is greater than zero, otherwise it is equal to zero. Refers to unmanned reconnaissance tools In time The state space of time, Refers to unmanned reconnaissance tools In time The state space of time; Let $\frac{ ... ; Sub-step 1-3-2: Unmanned reconnaissance vehicles conduct patrol reconnaissance and collect reconnaissance data for the assigned tactical reconnaissance mission; The unmanned reconnaissance vehicle conducts patrol reconnaissance according to the tactical reconnaissance tasks assigned in sub-step 1-2-3 and the reconnaissance area in sub-step 1-2-4, ensuring that the patrol reconnaissance route covers the entire reconnaissance area, and uses cameras, infrared imaging equipment, radar, and listening equipment to collect reconnaissance data in the reconnaissance area during the patrol reconnaissance. Within a cluster, unmanned reconnaissance tools can obtain information about the number of unmanned reconnaissance tools in the cluster, their identification numbers, and probability parameters through local information exchange with neighboring unmanned reconnaissance tools.

[0027] ; is a constant representing the "willingness" of unmanned reconnaissance vehicles to leave the cluster.

[0028] Furthermore, the unmanned reconnaissance tools within the cluster determine every second whether to leave the cluster and share and forward the reconnaissance data, thereby transmitting it back to the data center of their own unmanned combat system.

[0029] Sub-step 1-3-3: The unmanned reconnaissance vehicle returns due to insufficient power or malfunction and the unfinished reconnaissance mission is handed over; When an unmanned reconnaissance vehicle detects low power or a malfunction, it reserves power for a return trip. If the reconnaissance mission is complete, it returns directly; if the mission is incomplete, it exchanges information with other unmanned reconnaissance vehicles in the cluster to obtain information about nearby available unmanned reconnaissance vehicles, and then calculates the probability of a nearby available unmanned reconnaissance vehicle joining the cluster. Furthermore, nearby unmanned reconnaissance vehicles need to calculate the probability of handing over unfinished reconnaissance tasks based on their remaining power, mission completion status, reconnaissance capabilities, reconnaissance range, and current location, and make a decision to join the cluster based on the calculation results.

[0030] If any unmanned reconnaissance vehicle within the sensing range does not join the cluster, it is prohibited from joining the cluster for 20 seconds. The edge computing system of the unmanned combat equipment then dispatches a suitable, fully powered unmanned reconnaissance vehicle to take over the unfinished reconnaissance mission, ensuring... The effectiveness of unmanned reconnaissance tools joining the cluster. The probability is: ; For the current moment, the aggregate Number of members; For aggregates The threshold; It is the largest aggregate that the system can form.

[0031] When there are multiple aggregates in a group system This ensures fair competition among the clusters, meaning that unmanned reconnaissance tools tend to join clusters with higher thresholds in order to maintain the balance among the clusters.

[0032] Sub-steps 1-3-4: Enhanced reconnaissance of important targets; For important areas or targets, relying on a single unmanned reconnaissance vehicle is far from sufficient. Often, two or more unmanned reconnaissance vehicles are needed to conduct continuous and intensive reconnaissance of the area from multiple angles and for extended periods, using various reconnaissance methods such as cameras, infrared imaging equipment, radar, and listening devices. In unit time Within the area, the probability of any two unmanned reconnaissance vehicles meeting is The probability of each unmanned reconnaissance vehicle encountering any other unmanned reconnaissance vehicle is... Therefore: ; The area of ​​the enclosed region; The average speed of the unmanned reconnaissance vehicle. The sensing radius of unmanned reconnaissance tools; for The area swept by unmanned reconnaissance vehicles within a given time period.

[0033] The Markov process of a unit-level unmanned reconnaissance vehicle constitutes the dynamic aggregation process of edge computing clusters of unmanned combat equipment. By utilizing the probability parameters of the unmanned reconnaissance vehicle, we can obtain the average number of members in each cluster at any given time. The variation in the average number of members in a cluster is influenced by other clusters in the system and state transition parameters.

[0034] Sub-steps 1-3-5: Optimization of the distribution of unmanned reconnaissance tools within the cluster; At any given time, the distribution of unmanned reconnaissance tools within each cluster of the unmanned combat equipment edge computing aggregate can be described by difference equations: ; Represents unmanned reconnaissance tools encountering clusters within a unit of time. and the average number added; The number of unmanned reconnaissance vehicles representing the roaming status per unit time; This indicates that unmanned reconnaissance vehicles have joined the cluster. The probability of; Indicates an aggregate The initial state; This indicates the aggregate per unit time. The average number of members leaving the cluster.

[0035] When the system reaches a steady state, we can obtain: ; By substituting the parameters of the unmanned reconnaissance tool into the difference equation, the ratio of member quotas of any two aggregates when the system is in equilibrium can be calculated.

[0036] ; Sub-steps 1-3-6 identify friendly defensive targets and enemy attack targets; The edge computing aggregate of unmanned combat equipment calculates the friendly defense zone and the enemy's strike zone based on user combat commands and reconnaissance data from sub-step 1-3-2. It then runs artificial intelligence algorithms to identify targets within the friendly defense zone and the enemy's strike zone from reconnaissance data collected by cameras, infrared imaging equipment, radar, and listening devices. This includes targets approaching or intruding into the friendly defense zone, such as infrared signals from personnel, equipment, and military dogs; images and radar signals from firepower deployments, field fortifications, buildings, landmines, camouflage, bridges, unmanned combat equipment, and suspicious sound signals. The probability of identifying friendly defense targets and enemy strike targets is calculated. First, the identification probabilities are compared; targets with higher probabilities have higher identification rates and better reconnaissance results. Second, priorities are calculated based on the size and threat level of the identified targets; targets with higher priorities are more important. Finally, it checks whether all targets for all tactical reconnaissance missions have been identified. If so, target identification terminates; otherwise, target identification is performed sequentially, with supplementary reconnaissance conducted on targets with low identification probabilities that are difficult to identify.

[0037] Sub-steps 1-3-7: Virtual reconnaissance exercise to confirm and optimize the reconnaissance mission; Furthermore, the edge computing aggregate of unmanned combat equipment can perform virtual reconnaissance exercises, verifying the rationality and feasibility of reconnaissance task allocation through online computation, and identifying and correcting errors and deviations before executing reconnaissance tasks. To test the system's stability, accuracy, and scalability, the edge computing aggregate of unmanned combat equipment can select different model parameters for calculation and comparative analysis, and can also adjust parameters... Let's analyze the relationship between time and the number of crew members in unmanned reconnaissance vehicles. When the number of unmanned reconnaissance vehicles is... AND equals the cluster threshold Adjusting model parameters while keeping the number of simulations constant. The study investigated the trend of the average number of clusters over time by changing the number of unmanned reconnaissance vehicles in the system. And adjust the aggregate threshold accordingly. This allows us to obtain the relationship between the average number of clusters and time under different numbers of unmanned reconnaissance vehicles.

[0038] Furthermore, the edge computing aggregate of unmanned combat vehicles can optimize the execution of reconnaissance missions on the edge computing aggregate, that is, to calculate and optimize the aggregation and balance of multiple unmanned reconnaissance vehicles online, i.e., to optimize the arrangement within the work area. The unmanned reconnaissance vehicles were reassigned, and aggregation requests were initiated when the time threshold for completing the reconnaissance mission was more favorable. This was achieved by modifying the threshold and model parameters. To conduct simulation experiments, namely virtual reconnaissance exercises, can optimize the average size and quota ratio of the aggregate at different times before the reconnaissance mission is completed, better accelerate the group convergence process of multiple unmanned reconnaissance tools, and provide decision-making basis for the execution prediction of unmanned reconnaissance missions, the group convergence characteristics of unmanned reconnaissance tools, and the determination of the parameters of the group system.

[0039] Sub-steps 1-3-8 complete all tactical and strategic reconnaissance missions; Each unmanned reconnaissance vehicle checks the execution status of its reconnaissance mission according to the task assignment results. After completing all reconnaissance missions, it reports the reconnaissance data and mission completion status to the unmanned combat equipment edge computing aggregate. If there are no new missions, it returns to its own base. If the unmanned combat equipment edge computing aggregate assigns a new reconnaissance mission, it skips to step 1 to execute the new reconnaissance mission. If the unmanned combat equipment edge computing aggregate assigns a new combat mission, it skips to step 2 to execute the new combat mission. If the unmanned combat equipment edge computing aggregate assigns a new battle results and damage assessment mission, it skips to step 3 to execute the new battle results and damage assessment mission. Step 2: The edge computing aggregate of unmanned combat equipment conducts intelligent unmanned combat based on unmanned reconnaissance data to protect its own defensive targets and attack the enemy's strike targets; Sub-step 2-1: The unmanned combat equipment edge computing aggregate formulates combat missions. Based on user combat instructions and the friendly defense targets and enemy attack targets identified in sub-steps 1-3-6, the unmanned combat equipment edge computing aggregate formulates combat missions, including combat area, combat methods, combat time, combat intensity, combat accuracy, specific combat targets, available combat tools and combat forces. Preferably, intelligent unmanned combat uses unmanned combat tools and their respective onboard computers to form an unmanned combat equipment edge computing aggregate to calculate combat missions. Sub-step 2-1-1: Initialization of the intelligent combat system; Preferably, an intelligent combat system is initialized, including... An unmanned combat vehicle, This indicates the combat mission number assigned to the unmanned combat vehicle. For each unmanned combat vehicle... state space It is represented using a Markov chain: ; During initialization This indicates that the unmanned combat vehicle is in a state of unassigned mission.

[0040] Sub-step 2-1-2: The edge computing aggregate of unmanned combat equipment allocates available unmanned combat tools according to user combat instructions and combat missions; Unmanned combat vehicles (UCVs) use edge computing to broadcast user combat commands and missions. UCVs receiving these broadcasts aggregate via wireless networking, meaning they are geographically dispersed, but each can exchange information with other UCV members within the aggregate to obtain relevant information about the broadcast missions. The probability of joining the mission is calculated based on the UCV's combat capabilities, operational range, and current location.

[0041] Unmanned combat vehicles (UCVs) that receive combat mission broadcasts aggregate via wireless networking. Although geographically dispersed, each UCV can exchange information with other members within the aggregate to obtain relevant mission information. The probability of joining the aggregate is calculated based on the UCV's combat capabilities, operational range, and current location. Next, the unmanned combat vehicle generates a random number and... Comparisons are made, and decisions are made based on the comparison results, until a sufficient number of unmanned combat vehicles are allocated for the combat mission. Sub-step 2-1-3: Edge computing aggregate computing of unmanned combat equipment for combat missions; The edge computing aggregate of unmanned combat equipment, based on user combat instructions and the friendly and enemy targets identified in sub-steps 1-3-6, formulates friendly defensive combat missions and enemy strike combat missions, including the combat areas, combat methods, combat time, combat intensity, combat accuracy, specific combat targets, available combat tools and combat forces, etc.; preferably, intelligent unmanned combat uses unmanned combat tools; preferably, priority is given to defending the friendly defensive targets with high priority in sub-steps 1-3-6, or priority is given to striking the enemy strike targets with high priority in sub-steps 1-3-6; preferably, the friendly defensive combat mission includes constructing 1-3 lines of defense, digging field fortifications for each line of defense, optimizing the combination of unmanned combat tools and firepower configuration to ensure that firepower can cover the entire friendly defensive area, and actively striking targets approaching or invading the friendly defensive area; preferably, the enemy strike combat mission includes 1-3 rounds of strikes to ensure that firepower can cover all enemy strike targets in the entire enemy strike area, and to carry out saturation strikes on important targets, fortified targets, as well as hidden targets and newly added targets that were not discovered during reconnaissance; Sub-step 2-2: The edge computing aggregate of the unmanned combat equipment checks whether the combat mission has been reconnoitered. Based on the input combat mission, the edge computing aggregate of the unmanned combat equipment obtains the relevant reconnaissance data from step 1 and checks whether the combat area and combat target corresponding to the combat mission have been reconnoitered. If so, proceed to step 2-3. If there are combat areas and combat targets that have not been reconnoitered, or if the reconnaissance targets are moving at high speed, then set the combat area and combat target as the reconnaissance area and reconnaissance target, and return to step 1 to deploy the reconnaissance mission to supplement the reconnaissance. Sub-steps 2-3: The edge computing aggregate of unmanned combat equipment intelligently allocates combat tasks; the edge computing aggregate of unmanned combat equipment selects available intelligent unmanned combat tools according to the user's combat instructions, and decomposes the combat tasks using the embedded edge computing platform and artificial intelligence technology of the intelligent unmanned combat tools; the decomposed targets include more specific combat areas, combat methods, combat time, combat intensity, combat precision, specific combat objectives, available combat tools and combat forces, etc. Sub-step 2-3-1: Construct an edge computing aggregate of unmanned combat equipment based on the combat mission; Unmanned combat vehicles that receive combat mission broadcasts gather through wireless networking to form an edge computing cluster of unmanned combat equipment. Sub-step 2-3-2: Division of operational areas; Edge computing and aggregate computing for unmanned combat equipment to divide the combat zone, using a radius of... The combat zone is described by a two-dimensional circle. The unmanned combat vehicle is considered as a circle with a radius of... If the distance between the two centers is less than Therefore, it can be assumed that unmanned combat vehicles possess unlimited communication and positioning capabilities. Adjacent unmanned combat vehicles can quickly exchange information and determine each other's positions.

[0042] Based on the operational area, the operational mission is broken down into several sets of operational sub-missions. Each combat sub-task corresponds to a specific combat area, including the detailed coordinates of that area, friendly defensive targets, and enemy attack targets; Sub-step 2-3-3: Assign combat sub-tasks to appropriate unmanned combat vehicles; The operational tasks of the unmanned combat vehicle (UCV) edge computing aggregate broadcast sub-step 2-1-3 are determined from the operational sub-tasks of friendly defense and enemy attack, operational methods, operational time, operational intensity, operational accuracy, specific operational targets, available operational tools and forces. Based on the UCV's operational capabilities, operational range, and current location, the UCVs are assigned operational sub-tasks they are capable of performing, and the calculations are performed for each UCV. state space It is represented using a Markov chain: ; at this time This indicates that the unmanned combat vehicle is in a state of assigned mission, and the corresponding number in the set is the assigned combat sub-mission number.

[0043] Sub-steps 2-3-4: Unmanned combat vehicles acquire the combat sub-task combat area; First, the unmanned combat vehicle obtains the operational area of ​​the corresponding sub-task based on the assigned operational sub-task number, including the detailed geographical coordinates of that area. When the distance between the unmanned combat vehicle and the circular area exceeds... This is considered as the unmanned combat vehicle reaching the area boundary. Assume the unmanned combat vehicle can obtain its current absolute coordinates and calculate its new direction of motion by generating a new turning angle. Assume the unmanned combat vehicle does not need to undergo acceleration / deceleration phases, avoid communication conflicts (due to hardware limitations), and has no steering errors during its movement.

[0044] Secondly, unmanned combat vehicles use a greedy algorithm to plan their combat routes, that is, they prioritize visiting the combat sub-tasks closest to their own location, and then visit the combat sub-tasks farther away, until all combat sub-tasks have been visited.

[0045] If the distance between unmanned combat vehicles is less than If the target is 1.2 times larger and no aggregation request is received, the edge computing platform dynamically adjusts the combat mission and runs a path planning algorithm on the unmanned combat vehicle. The unmanned combat vehicle will trigger an obstacle avoidance mechanism, causing adjacent unmanned combat vehicles to generate new directions of movement and paths. Once the unmanned combat vehicle enters the communication range allocated by the edge computing aggregation of unmanned combat equipment, it will be able to obtain all information and participate in combat sub-missions.

[0046] Furthermore, edge computing aggregates of unmanned combat vehicles (UCVs) can assign high-density, high-frequency combat sub-tasks to important combat areas or targets. This means multiple UCVs need to be densely clustered at a key location for repeated combat and collaborative operations. When multiple UCVs enter a key combat position, they form a specific edge computing aggregate for that location. The UCVs within the aggregate can communicate with each other, sharing their combat information and status. The edge computing aggregate can continuously optimize its combat sub-tasks until the high-density, high-frequency combat sub-tasks meet the operational requirements.

[0047] Sub-steps 2-4: The edge computing aggregate of unmanned combat equipment intelligently executes combat missions; the edge computing aggregate of unmanned combat equipment coordinates the execution of combat missions by various unmanned combat tools according to the decomposed combat missions, and runs artificial intelligence algorithms in the execution process to cope with the ever-changing situation on the battlefield; Sub-step 2-4-1: Initialization of the unmanned combat vehicle cluster; Assuming the edge computing aggregate of unmanned combat equipment is composed of It consists of several unmanned combat vehicles, which contain An aggregate. The probability of an unmanned combat vehicle joining a cluster while roaming is: ; Generally speaking, when the number of group members does not exceed At the threshold, The value is greater than zero, otherwise it is equal to zero. Refers to unmanned combat vehicles In time The state space of time, Refers to unmanned combat vehicles In time The state space of time; The probability of an unmanned combat vehicle leaving the cluster is: ; Sub-step 2-4-2: Unmanned combat vehicles perform defense or attack according to the assigned combat sub-tasks; According to the combat sub-tasks assigned in sub-step 2-3-3 and the combat area in sub-step 2-3-4, unmanned combat vehicles conduct patrol defense or direct strikes, ensuring that the patrol defense route covers the entire combat area defended by the friendly forces or attacked by the enemy, and that direct strikes can effectively destroy the targets corresponding to the combat sub-tasks. During patrol defense, cameras, infrared imaging equipment, radar, and listening equipment are used to identify friend or foe in the combat area, ensuring that the enemy is struck quickly without harming friendly forces. Within a cluster, unmanned combat vehicles can obtain information such as the number of unmanned combat vehicles in the group, their serial numbers, and probability parameters through local information transmission between adjacent unmanned combat vehicles.

[0048] ; is a constant representing the "willingness" of unmanned combat vehicles to leave the cluster.

[0049] Furthermore, the unmanned combat vehicles in the cluster determine every second whether to leave the cluster and share and forward reconnaissance data so that multiple unmanned combat vehicles can carry out coordinated defense or coordinated attack.

[0050] Sub-steps 2-4-3: The unmanned combat vehicle returns due to insufficient power or malfunction and the handover of unfinished combat sub-tasks; When an unmanned combat vehicle detects low power or a malfunction, it will reserve power for a return trip. If the combat sub-mission has been completed, it will return directly; if the combat sub-mission has not been completed, it will exchange information with other unmanned combat vehicle members in the cluster to obtain relevant information about nearby available unmanned combat vehicles, and then calculate the probability of the incomplete combat mission handover, i.e., the probability of a nearby available unmanned combat vehicle joining the cluster. Furthermore, nearby unmanned combat vehicles need to calculate the probability of handing over unfinished combat missions based on their remaining power, mission completion status, combat capabilities, combat range, and current location, and make a decision to join the cluster based on the calculation results.

[0051] If any unmanned combat vehicles within the sensing range do not join the cluster, they are prohibited from joining the cluster for 20 seconds. The edge computing system of the unmanned combat vehicles then dispatches a suitable, fully powered unmanned combat vehicle to take over the unfinished combat sub-tasks, ensuring... The effectiveness of unmanned combat vehicles joining the cluster The probability is: ; For the current moment, the aggregate Number of members; For aggregates The threshold; It is the largest aggregate that the system can form.

[0052] When there are multiple aggregates in a group system This ensures fair competition among the various clusters, meaning that unmanned combat vehicles tend to join clusters with higher thresholds in order to maintain the balance among the clusters.

[0053] Sub-step 2-4-4: The unmanned combat vehicle calls for fire support; If the unmanned combat vehicle finds that the combat sub-tasks assigned in sub-step 2-3-3 cannot be completed, including when its own casualties are too great to continue effective combat, when the enemy has captured the three lines of defense in its own defensive area and cannot drive them away, when its own defensive targets have been attacked or destroyed, when the enemy's targets have not been effectively destroyed after three rounds of attacks, or when a large number of enemy reinforcements have appeared, it needs to call for fire support from the edge computing cluster of the unmanned combat vehicle and dispatch unmanned combat vehicles with the same or greater lethality or targeted lethality to enhance its ability to complete the combat sub-tasks. In unit time Within the space, the probability of any two unmanned combat vehicles encountering each other is... The probability of each unmanned combat vehicle encountering any other unmanned combat vehicle is: Therefore: ; The area of ​​the enclosed region; The average speed of the unmanned combat vehicle; To determine the sensing radius of unmanned combat vehicles; for The area swept by unmanned combat vehicles within a given time period.

[0054] The Markov process of a unit-level unmanned combat vehicle constitutes the dynamic aggregation process of edge computing clusters of unmanned combat vehicles. By utilizing the probabilistic parameters of the unmanned combat vehicle, we can obtain the average number of members in each cluster at any given time. The variation in the average number of members in a cluster is influenced by other clusters in the system and state transition parameters.

[0055] Sub-steps 2-4-5: Optimization of the distribution of unmanned combat vehicles within the cluster; At any given time, the distribution of unmanned combat vehicles within each aggregate of edge computing clusters in the system can be described by difference equations: ; Represents unmanned combat vehicles encountering clusters within a unit of time. and the average number added; This indicates the number of unmanned combat vehicles in a roaming state per unit of time. This indicates that unmanned combat vehicles have joined the cluster. The probability of; Indicates an aggregate The initial state; This indicates the aggregate per unit time. The average number of members leaving the cluster.

[0056] When the system reaches a steady state, we can obtain: ; By substituting the parameters of unmanned combat vehicles into the difference equation, the ratio of member quotas of any two aggregates when the system is in equilibrium can be calculated.

[0057] ; Sub-steps 2-4-6 involve multi-layered defense and multiple rounds of attacks to complete the combat mission; During the execution of combat missions, edge computing aggregates of unmanned combat equipment adhere to the principles of multi-layered defense and multi-round attacks. They are not limited to three lines of defense or three rounds of attacks. They make full use of the remaining unmanned combat tools to block the enemy from attacking their own defensive targets and do their best to attack the enemy's targets to ensure the completion of combat missions.

[0058] Sub-steps 2-4-7 involve virtual combat exercises to confirm and optimize combat missions; Furthermore, the edge computing aggregate of unmanned combat equipment can compute virtual combat exercises on its own, verifying the rationality and feasibility of combat mission allocation through online computation, and identifying and correcting errors and deviations before the execution of combat missions. To test the system's stability, accuracy, and scalability, the edge computing aggregate of unmanned combat equipment can select different model parameters for calculation and comparative analysis, and can also adjust parameters... Let's analyze the relationship between time and the number of crew members in unmanned combat vehicles. When the number of unmanned combat vehicles is... AND equals the cluster threshold Adjusting model parameters while keeping the number of simulations constant. The study investigates the changing trend of the average number of clusters over time. This is achieved by varying the number of unmanned combat vehicles in the system. And adjust the aggregate threshold accordingly. This allows us to obtain the relationship between the average number of clusters and time under different numbers of unmanned combat vehicles.

[0059] Furthermore, the edge computing aggregate of unmanned combat vehicles can optimize the execution of combat missions on the edge computing aggregate of unmanned combat vehicles, that is, to calculate and optimize the aggregation and balance of multiple unmanned combat vehicles online, that is, to optimize the arrangement within the working area. The unmanned combat vehicles are reassigned, and aggregation requests are initiated when the time threshold for completing the combat mission is more favorable. This is achieved by modifying the threshold and model parameters. To conduct simulation experiments, namely virtual combat exercises, can optimize the average size and quota ratio of the aggregate at different times before the completion of the combat mission, better accelerate the group convergence process of multiple unmanned combat vehicles, and provide decision-making basis for the execution prediction of unmanned combat missions, the group convergence characteristics of unmanned combat vehicles, and the determination of the parameters of the group system.

[0060] Sub-steps 2-4-8 complete the operation; After each unmanned combat vehicle completes all combat missions, it reports combat data and mission completion status to the unmanned combat equipment edge computing cluster. If there are no new missions, it returns to its own base. If the unmanned combat equipment edge computing cluster assigns a new reconnaissance mission, it skips to step 1 to execute the new reconnaissance mission. If the unmanned combat equipment edge computing cluster assigns a new combat mission, it skips to step 2 to execute the new combat mission. If the unmanned combat equipment edge computing cluster assigns a new battle results and damage assessment mission, it skips to step 3 to execute the new battle results and damage assessment mission. Step 3: The edge computing aggregate of unmanned combat equipment conducts intelligent unmanned combat results and damage assessment during unmanned combat to determine whether the status of friendly defensive targets and enemy attack targets meets the user's combat instructions. Sub-step 3-1: The unmanned combat equipment edge computing aggregate calculates the battle results and damage assessment task; the unmanned combat equipment edge computing aggregate calculates the battle results and damage assessment task based on the user's combat instructions and the friendly defense targets and enemy attack targets identified in sub-step 1-3-6, including the battle results and damage assessment area, battle results and damage assessment methods, battle results and damage assessment time, battle results and damage assessment intensity, battle results and damage assessment accuracy, specific battle results and damage assessment targets, available battle results and damage assessment tools, and battle results and damage assessment forces; preferably, intelligent unmanned battle results and damage assessment uses unmanned reconnaissance tools, and each of them uses its onboard computers to form an unmanned combat equipment edge computing aggregate to calculate the battle results and damage assessment task; the unmanned combat equipment edge computing aggregate independently acquires the relevant reconnaissance data from step 1 and the relevant combat data from step 2 based on the battle results and damage assessment task; Sub-step 3-1-1: Initialize the intelligent combat system; Preferably, an intelligent combat system is initialized, including... A tool for assessing the combat results and damage of unmanned aerial vehicles. This indicates the task number assigned to the unmanned combat results and damage assessment tool. For each unmanned combat results and damage assessment tool... state space It is represented using a Markov chain: ; During initialization This indicates that the battle results and damage assessment tool is in an unassigned task state.

[0061] Sub-step 3-1-2: The edge computing aggregate of unmanned combat equipment allocates available unmanned combat results and damage assessment tools according to user combat instructions and combat results and damage assessment tasks. The edge computing aggregate of unmanned combat equipment broadcasts user combat commands and unmanned combat results and damage assessment tasks. Unmanned combat results and damage assessment tools that receive the broadcast of unmanned combat results and damage assessment tasks are aggregated through wireless networking. That is, they are geographically dispersed, but each individual can exchange information with other unmanned combat results and damage assessment tool members in the aggregate to obtain relevant information about the broadcast unmanned combat results and damage assessment tasks. The probability of joining the unmanned combat results and damage assessment task is calculated based on the reconnaissance capabilities, reconnaissance range, and current location of the unmanned combat results and damage assessment tools until a sufficient number of unmanned combat results and damage assessment tools are allocated for the unmanned combat results and damage assessment task; Sub-step 3-1-3, the edge computing aggregate of unmanned combat equipment calculates the combat results and damage assessment task; The edge computing aggregate of unmanned combat equipment formulates tasks for assessing friendly combat losses and enemy strike results based on user combat instructions and the friendly defense targets and enemy strike targets identified in sub-steps 1-3-6. This includes the combat areas of friendly defense and enemy strike, available combat tools and forces, specific combat targets, high-priority friendly defense targets, and high-priority enemy strike targets. Furthermore, it calculates the number and location of targets for assessing friendly combat losses and enemy strike results. Sub-step 3-2: The edge computing aggregate of unmanned combat equipment intelligently allocates battle result and damage assessment tasks; the edge computing aggregate of unmanned combat equipment selects available intelligent unmanned reconnaissance tools according to the user's combat instructions, and decomposes the battle result and damage assessment tasks using the embedded edge computing platform and artificial intelligence technology of the intelligent unmanned reconnaissance tools; the decomposed targets include more specific battle result and damage assessment areas, battle result and damage assessment methods, battle result and damage assessment time, battle result and damage assessment intensity, battle result and damage assessment accuracy, specific battle result and damage assessment targets, available battle result and damage assessment tools, and battle result and damage assessment forces, etc. Sub-step 3-2-1: Construction of edge computing aggregates for unmanned combat equipment; Unmanned combat results and damage assessment tools that receive the combat results and damage assessment mission broadcast gather through wireless networking to form an edge computing aggregation of unmanned combat equipment. Sub-step 3-2-2: Delineation of battle results and damage assessment areas; Edge computing aggregates of unmanned combat equipment are used for combat result and damage assessment and area division, utilizing a radius of... A two-dimensional circle is used to describe the area for assessing battle results and damage. The battle results and damage assessment tool is considered as a circle with a radius of... If the distance between the two centers is less than Therefore, it can be assumed that the combat results and damage assessment tools possess unlimited communication and positioning capabilities. Adjacent combat results and damage assessment tools can quickly exchange information and determine each other's locations.

[0062] Based on the area for assessing battle results and damage, the task of assessing battle results and damage is broken down into several sets of tasks. Each battle result and damage assessment task corresponds to a specific battle result and damage assessment area, including the detailed coordinates of that area; Sub-step 3-2-3 assigns the task of assessing battle results and damage to a suitable unmanned battle results and damage assessment tool; The edge computing aggregate broadcast sub-step 3-2-2 of the unmanned combat equipment's battle results and damage assessment task involves assessing the battle results and damage of both friendly and enemy attacks from the perspectives of friendly defense and enemy attack results and damage assessment tasks. This includes the assessment methods, assessment time, assessment intensity, assessment accuracy, specific assessment targets, available assessment tools, and assessment forces. Based on the unmanned battle results and damage assessment tools' capabilities, assessment range, and current location, the task assigns them the battle results and damage assessment tasks they are capable of performing. Finally, it calculates the battle results and damage assessment tasks for each unmanned battle results and damage assessment tool. state space It is represented using a Markov chain: ; at this time This indicates that the unmanned combat results and damage assessment tool is in a task-assigned state, and the corresponding number in the set is the assigned combat results and damage assessment task number.

[0063] Sub-step 3-2-4: Obtain the battle results and damage assessment area; First, the unmanned combat damage assessment tool obtains the corresponding combat damage assessment area, including its detailed geographic coordinates, based on the assigned combat damage assessment task number. When the distance between the combat damage assessment tool and the circular area exceeds... The battle damage assessment tool is considered to have reached the area boundary. Assume the tool can obtain its current absolute coordinates and calculate its new direction of movement by generating a new turning angle. Assume the tool's movement does not require acceleration / deceleration phases, avoids communication conflicts, or suffers from steering errors.

[0064] Secondly, the unmanned reconnaissance tools use a greedy algorithm to plan their battle damage assessment routes. This means they prioritize visiting the battle damage assessment tasks closest to their own location, then move on to those further away, until all battle damage assessment tasks have been visited. If the distance between the battle damage assessment tools is less than... If the target is 1.2 times larger and no aggregation request is received, the edge computing platform dynamically adjusts the combat mission and runs a path planning algorithm on the battle results and damage assessment tool. The battle results and damage assessment tool will trigger an obstacle avoidance mechanism, causing adjacent battle results and damage assessment tools to generate new movement directions and paths. Once the battle results and damage assessment tool enters the communication range allocated by the edge computing aggregation of unmanned combat equipment, it will be able to obtain all information and participate in the battle results and damage assessment mission.

[0065] Furthermore, the edge computing aggregate of unmanned combat equipment can set high-density, high-frequency battle result and damage assessment tasks for important battle result and damage assessment areas or targets. This means that multiple unmanned battle result and damage assessment tools need to be densely clustered at a specific important location for repeated battle result and damage assessments and collaborative work. When multiple unmanned battle result and damage assessment tools enter an important battle result and damage assessment location, they constitute a specific unmanned combat equipment edge computing aggregate at that location. The unmanned battle result and damage assessment tools within the aggregate can communicate with each other, sharing their respective battle result and damage assessment information and status. The unmanned combat equipment edge computing aggregate can continuously optimize the aggregate's battle result and damage assessment tasks until the high-density and high-frequency battle result and damage assessment tasks meet operational requirements.

[0066] Sub-step 3-3: The edge computing aggregate of unmanned combat equipment intelligently executes the task of assessing the results and damage of combat. The edge computing aggregate of unmanned combat equipment, based on the decomposed task of assessing the results and damage of combat, is jointly executed by various unmanned reconnaissance tools to calculate the target loss in the friendly defense area and the enemy's strike area, that is, the damage to the target before and after the start of the combat mission. Artificial intelligence algorithms are run during the execution process to cope with the ever-changing situation on the battlefield. Sub-step 3-3-1: Initialization of the battle results and damage assessment tool aggregate; Assuming the edge computing aggregate of unmanned combat equipment is composed of It consists of a set of tools for assessing battle results and damage, which include An aggregate. The probability of the battle results and damage assessment tool being added to a cluster, under roaming conditions, is: ; Generally speaking, when the number of group members does not exceed At the threshold, The value is greater than zero, otherwise it is equal to zero. This refers to unmanned combat results and damage assessment tools. In time The state space of time, This refers to unmanned combat results and damage assessment tools. In time The state space of time; The probability of the battle results and damage assessment tool leaving the cluster is: [The probability of the battle results and damage assessment tool leaving the cluster under clustered conditions is:] ; Sub-step 3-3-2: Conduct a patrol assessment of the assigned battle results and damages and collect battle results and damages assessment data. The unmanned combat results and damage assessment tool conducts a patrol-based combat results and damage assessment based on the combat results and damage assessment tasks assigned in sub-step 3-2-3 and the combat results and damage assessment areas in sub-step 3-2-4. It ensures that the patrol-based combat results and damage assessment route covers the entire combat results and damage assessment area, and uses cameras, infrared imaging equipment, radar, and listening equipment to collect combat results and damage assessment data in the combat results and damage assessment area during the patrol-based combat results and damage assessment. Within a cluster, unmanned combat results and damage assessment tools can obtain information such as the number of unmanned combat results and damage assessment tools, the number of unmanned combat tools, and probability parameters in the cluster through local information transmission between adjacent unmanned combat results and damage assessment tools.

[0067] ; The constant represents the "willingness" of the battle results and damage assessment tool to leave the cluster. Every second, the battle results and damage assessment tool in the cluster determines whether to leave the cluster and shares and forwards the battle results and damage assessment data, thereby transmitting it back to the data center of its own unmanned combat system.

[0068] Sub-step 3-3-3: The unmanned combat results and damage assessment tool returns due to insufficient power or malfunction and the unfinished combat results and damage assessment task is handed over. When a battle damage assessment tool detects insufficient power or a malfunction, it will reserve power for a return trip. If the battle damage assessment mission has been completed, it will return directly. If the mission has not been completed, it will exchange information with other unmanned battle damage assessment tools in the cluster to obtain relevant information about nearby available unmanned battle damage assessment tools. It will then calculate the probability of a failed mission handover, i.e., the probability of a nearby available unmanned battle damage assessment tool joining the cluster. Furthermore, nearby unmanned combat damage assessment tools need to calculate the probability of handing over unfinished combat damage assessment tasks based on their remaining power, their own mission completion status, combat damage assessment capabilities, combat damage assessment range, and current location, and make a decision to join the cluster based on the calculation results.

[0069] If a combat damage assessment tool within the perception range does not join the cluster, it is prohibited from joining the cluster for 20 seconds. The edge computing cluster of unmanned combat vehicles then dispatches a suitable, fully powered unmanned combat damage assessment tool to take over the unfinished combat damage assessment task, ensuring... The effectiveness of the battle results and damage assessment tool was incorporated into the aggregate. The probability is: ; For the current moment, the aggregate Number of members; For aggregates The threshold; It is the largest aggregate that the system can form.

[0070] When there are multiple aggregates in a group system This ensures fair competition among the clusters, meaning that the battle results and losses assessment tool tends to include clusters with higher thresholds in order to maintain the balance among the clusters.

[0071] Sub-step 3-3-4: Enhance the assessment of combat results and damage for important targets; For important areas or targets, relying on a single unmanned combat results and damage assessment tool is far from sufficient. Often, two or more unmanned combat results and damage assessment tools are needed to conduct continuous and intensive combat results and damage assessments of the assessment area from multiple angles and over a long period of time using various methods such as cameras, infrared imaging equipment, radar, and listening equipment. In unit time Within this context, the probability of any two battle result / damage assessment tools meeting is... The probability that each battle result / damage assessment tool encounters any other battle result / damage assessment tool is: Therefore: ; The area of ​​the enclosed region; The average speed of the movement of the tool for assessing battle results and damage; The perception radius of the battle results and damage assessment tool; for The area covered by the battle results and damage assessment tool within a given time period.

[0072] The Markov process of the unit battle results and damage assessment tool constitutes the dynamic aggregation process of edge computing clusters of unmanned combat equipment. By utilizing the probabilistic parameters of the battle results and damage assessment tool, we can obtain the average number of members in each cluster at any time point. The variation in the average number of members in a cluster is affected by other clusters in the system and state transition parameters.

[0073] Sub-step 3-3-5: Optimization of the distribution of battle results and damage assessment tools within the cluster; At any given time, the distribution of unmanned combat vehicles within each aggregate of edge computing clusters in the system can be described by difference equations: ; The tool for assessing battle results and damage per unit time encounters an aggregate. and the average number added; The number of tools for assessing combat results and damages representing the roaming status per unit of time. This indicates that the battle results and damage assessment tool has been incorporated into the cluster. The probability of; Indicates an aggregate The initial state; This indicates the aggregate per unit time. The average number of members leaving the cluster.

[0074] When the system reaches a steady state, we can obtain:

[0075] By substituting the parameters of the battle results and losses assessment tool into the difference equation, the ratio of member quotas of any two aggregates at system equilibrium can be calculated.

[0076] ; Sub-step 3-3-6: Calculate the battle damage to our own defensive targets and the results of the enemy's attack on the targets; The edge computing aggregate of unmanned combat equipment calculates the target loss situation in the friendly defense area and the enemy's strike area based on the battle result and damage assessment task assigned in sub-step 3-2-3 and the battle result and damage assessment area in sub-step 3-2-4, that is, the damage situation of the targets before and after the start of the combat mission; it runs artificial intelligence algorithms to identify targets in the friendly defense area and the enemy's strike area from the battle result and damage assessment data collected by cameras, infrared imaging equipment, radar, and listening equipment, as well as targets approaching and intruding into the friendly defense area, including infrared signals of combat personnel, equipment, and military dogs, firepower configuration, field fortifications, buildings, landmines, camouflage, bridges, images and radar signals of unmanned combat equipment, and suspicious sound signals, and calculates the loss probability of friendly defense targets and enemy strike targets. First, a comparison of loss probabilities is performed. Targets with higher probabilities have a higher loss rate and are more severely affected by attacks, while those with lower probabilities have a better defensive effect. Second, priorities are calculated based on the size and threat level of the identified targets. Targets with higher priorities are more important. For high-priority defensive targets, the lower the probability of loss, the better (i.e., lower battle damage). For high-priority attack targets, the higher the probability of loss, the better (i.e., greater battle results). Finally, it is checked whether all targets in all battle results and battle damage assessment tasks have been calculated. If so, the battle results and battle damage assessment calculation is terminated. Otherwise, the battle results and battle damage assessment calculation is performed sequentially, and supplementary battle results and battle damage assessments are conducted for targets with low identification probabilities that are difficult to identify.

[0077] Sub-step 3-3-7, virtual battle results and damage assessment exercise to confirm and optimize the battle results and damage assessment task; The edge computing aggregate of unmanned combat equipment can initially assess the results and losses on its own. That is, based on the relevant reconnaissance data in step 1 and the relevant combat data in step 2, it compares the results and losses assessment task results in sub-step 3-3-6 to initially calculate the results and losses assessment data. Furthermore, the edge computing aggregate of unmanned combat equipment can perform virtual battle result and damage assessment exercises on the unmanned combat equipment edge computing aggregate platform. This means that it verifies the rationality and feasibility of the battle result and damage assessment task allocation through online calculations, identifying and correcting errors and deviations before executing the battle result and damage assessment task. To test the system's stability, accuracy, and scalability, the unmanned combat equipment edge computing aggregate can select different model parameters for calculation and comparative analysis, and can also adjust the parameters... This analysis examines the relationship between time and the number of personnel in unmanned combat results and damage assessment tools. The number of personnel in the combat results and damage assessment tool is... AND equals the cluster threshold Adjusting model parameters while keeping the number of simulations constant. The study investigated the changing trend of the average number of aggregates over time. This was achieved by varying the number of battle results and damage assessment tools within the system. And adjust the aggregate threshold accordingly. This allows us to obtain the relationship between the average number of clusters and time under different numbers of battle results and damage assessment tools.

[0078] Furthermore, the edge computing aggregate of unmanned combat equipment can perform calculations and optimize the execution of combat result and damage assessment tasks on the edge computing aggregate of unmanned combat equipment. That is, it can calculate and optimize the aggregation and balance of multiple unmanned combat result and damage assessment tools online, that is, optimize the arrangement within the working area. The unmanned aerial vehicle (UAV) combat results and damage assessment tool was reassigned, and an aggregation request was initiated if the time threshold for completing the combat results and damage assessment task was more favorable. This was achieved by modifying the threshold and model parameters. To conduct simulation experiments, namely virtual battle results and damage assessment exercises, can optimize the average size and quota ratio of the aggregate at different times before the battle results and damage assessment task is completed. This can better accelerate the group convergence process of multiple unmanned battle results and damage assessment tools, and provide decision-making basis for the execution prediction of unmanned reconnaissance battle results and damage assessment tasks, the group convergence characteristics of unmanned battle results and damage assessment tools, and the determination of the parameters of the group system.

[0079] Sub-step 3-3-8, the assessment of battle results and losses is completed; After each unmanned combat damage assessment tool completes all combat damage assessment tasks, it reports the combat damage assessment data and task completion status to the unmanned combat equipment edge computing aggregate. The unmanned combat equipment edge computing aggregate determines whether the status of friendly defensive targets and enemy attack targets meets the user's combat instructions. If the user's combat instructions are met, and there are no other new tasks, it returns to its own base, and the operation ends. If the user's combat instructions are not met, and the unmanned combat equipment edge computing aggregate assigns a new reconnaissance task, it jumps to step 1 to execute the new reconnaissance task; if the unmanned combat equipment edge computing aggregate assigns a new combat task, it jumps to step 2 to execute the new combat task; if the unmanned combat equipment edge computing aggregate assigns a new combat damage assessment task, it jumps to step 3 to execute the new combat damage assessment task.

Claims

1. An intelligent unmanned combat system, characterized in that, The system utilizes unmanned combat equipment for reconnaissance, combat, and assessment of battle results and damage, forming an edge computing aggregate of unmanned combat equipment using their onboard computers; the system, when in operation, includes the following steps: Step 1: The edge computing aggregate of unmanned combat equipment conducts intelligent unmanned reconnaissance based on user combat instructions, identifying friendly defensive targets and enemy attack targets; Step 2: The edge computing aggregate of unmanned combat equipment conducts intelligent unmanned combat based on unmanned reconnaissance data to protect its own defensive targets and attack the enemy's strike targets; Step 3: The edge computing aggregate of unmanned combat equipment conducts intelligent unmanned combat results and damage assessment during unmanned combat to determine whether the status of friendly defensive targets and enemy attack targets meets the user's combat instructions. Step 1 includes the following sub-steps: Sub-step 1-1: The unmanned combat equipment edge computing aggregate receives user combat commands; the unmanned combat equipment edge computing aggregate calculates strategic reconnaissance missions based on user combat commands, including strategic reconnaissance areas, reconnaissance methods, reconnaissance time, reconnaissance intensity, reconnaissance accuracy, specific reconnaissance targets, available reconnaissance tools and reconnaissance forces; intelligent unmanned reconnaissance uses unmanned reconnaissance tools and utilizes their respective onboard computers to form an unmanned combat equipment edge computing aggregate to calculate reconnaissance missions. Sub-steps 1-2: The edge computing aggregate of unmanned combat equipment intelligently allocates reconnaissance tasks; based on user operational instructions, the edge computing aggregate selects available intelligent unmanned reconnaissance tools, and uses the embedded edge computing platform and artificial intelligence technology of the intelligent unmanned reconnaissance tools to decompose strategic reconnaissance tasks into tactical reconnaissance tasks; the decomposed targets include the specific tactical reconnaissance areas covered by user operational instructions, reconnaissance methods, reconnaissance time, reconnaissance intensity, reconnaissance accuracy, specific reconnaissance targets, available reconnaissance tools, and reconnaissance forces; the decomposed tactical reconnaissance tasks are then allocated to appropriate unmanned reconnaissance tools; Sub-steps 1-3: The edge computing aggregate of unmanned combat equipment intelligently performs reconnaissance missions, identifying friendly defensive targets and enemy strike targets; the edge computing aggregate of unmanned combat equipment, according to the decomposed strategic and tactical reconnaissance missions, is coordinated by various unmanned reconnaissance tools to perform reconnaissance missions, and runs artificial intelligence algorithms during the execution process to identify friendly defensive targets and enemy strike targets in order to cope with the ever-changing situation on the battlefield.

2. The system according to claim 1, characterized in that, In step 1-1, the following sub-steps are used: Sub-step 1-1-1: Initialize the intelligent reconnaissance system; Initialize an intelligent reconnaissance system, including An unmanned reconnaissance tool, This indicates the reconnaissance mission number assigned to the unmanned reconnaissance vehicle; for each unmanned reconnaissance vehicle state space It is represented using a Markov chain: ； During initialization At this time, it indicates that the unmanned reconnaissance vehicle is in a state of unassigned tasks; Sub-step 1-1-2: The edge computing aggregate of unmanned combat equipment calculates strategic reconnaissance missions based on user combat instructions; The edge computing aggregate of unmanned combat equipment acquires user combat instructions and broadcasts them, requesting unmanned reconnaissance tools to gather. The gathered unmanned combat equipment edge computing aggregate calculates the user combat instructions into a strategic reconnaissance mission that satisfies the user combat instructions by geographical area, including strategic reconnaissance area, reconnaissance methods, reconnaissance time, reconnaissance intensity, reconnaissance accuracy, specific reconnaissance targets, and calculates the available reconnaissance tools and reconnaissance forces. Sub-step 1-1-3: The edge computing aggregate of unmanned combat equipment allocates available unmanned reconnaissance tools according to the strategic reconnaissance mission; In the strategic reconnaissance mission of the edge computing aggregate broadcast sub-step 1-1-2 for unmanned combat equipment, the unmanned reconnaissance tools that receive the reconnaissance mission broadcast aggregate through wireless networking. Although geographically dispersed, each individual tool can exchange information with other members of the aggregate to obtain relevant information about the broadcast strategic reconnaissance mission. The probability of joining the strategic reconnaissance mission is calculated based on the unmanned reconnaissance tool's reconnaissance capabilities, reconnaissance range, and current location. Next, the edge computing aggregate of the unmanned combat equipment generates random numbers and interacts with them. Comparisons are made, and decisions are made based on the comparison results to include unmanned reconnaissance tools in strategic reconnaissance missions until a sufficient number of unmanned reconnaissance tools that can meet the requirements of strategic reconnaissance missions are allocated. In steps 1-2, the following sub-steps are used: Sub-step 1-2-1: Construct an edge computing cluster of unmanned combat equipment based on the strategic reconnaissance mission; Unmanned reconnaissance vehicles that receive strategic reconnaissance mission broadcasts gather via wireless networking to form an edge computing cluster of unmanned combat equipment. Sub-step 1-2-2: The strategic reconnaissance mission is broken down into several tactical reconnaissance missions; The edge computing cluster of unmanned combat vehicles divides the strategic reconnaissance area into several tactical reconnaissance areas, utilizing a radius of... The reconnaissance area is described by a two-dimensional circle; the unmanned reconnaissance vehicle is considered as a circle with a radius of... If the distance between the two centers is less than Therefore, it can be assumed that unmanned reconnaissance vehicles have unlimited communication and positioning capabilities; adjacent unmanned reconnaissance vehicles can quickly exchange information and determine each other's positions; Based on the strategic reconnaissance area, strategic reconnaissance missions are broken down into several sets of tactical reconnaissance missions according to different areas. Each tactical reconnaissance mission corresponds to a specific reconnaissance area, including the detailed coordinates of that area. The set of all tactical reconnaissance missions should cover the entire strategic reconnaissance area and have a moderate overlap with each other. Sub-steps 1-2-3 assign tactical reconnaissance tasks to appropriate unmanned reconnaissance vehicles; The tactical reconnaissance mission of the unmanned combat vehicle edge computing aggregate broadcast sub-step 1-2-2 is based on the reconnaissance area, reconnaissance methods, reconnaissance time, reconnaissance intensity, reconnaissance accuracy, specific reconnaissance targets, available reconnaissance tools and reconnaissance forces. According to the reconnaissance capabilities, reconnaissance range and current location of the unmanned reconnaissance vehicle, it is assigned tactical reconnaissance tasks it can perform, and the calculations are performed for each unmanned reconnaissance vehicle. state space It is represented using a Markov chain: ； at this time This indicates that the unmanned reconnaissance vehicle is in a state of assigned mission, and the corresponding number in the set is the assigned reconnaissance mission number; Sub-steps 1-2-4: Unmanned reconnaissance tools acquire the reconnaissance area for tactical reconnaissance missions; First, the unmanned reconnaissance vehicle obtains the reconnaissance area corresponding to the assigned tactical reconnaissance mission number, including the detailed geographical coordinates of that area; when the distance between the unmanned reconnaissance vehicle and the circular area exceeds... Let be the latitude and longitude coordinates of the unmanned reconnaissance vehicle reaching the area boundary; assume that the unmanned reconnaissance vehicle can obtain its current absolute latitude and longitude coordinates based on GPS signals and calculate its new direction of motion by generating a new turning angle; assume that the unmanned reconnaissance vehicle does not need to experience acceleration and deceleration phases, avoid communication conflicts, or have other hardware limitations during its movement, and that there are no issues such as turning errors. Secondly, the unmanned reconnaissance tool uses a greedy algorithm to plan tactical reconnaissance routes, that is, it prioritizes visiting the tactical reconnaissance missions closest to its own position in the reconnaissance area, and then visits the tactical reconnaissance missions farther away from itself, until all tactical reconnaissance missions have been visited. If the distance between unmanned reconnaissance vehicles is less than 1.2 times that of the unmanned reconnaissance vehicle and without receiving aggregating requests, the edge computing platform dynamically adjusts the tactical reconnaissance mission, runs a path planning algorithm on the unmanned reconnaissance vehicle, and the unmanned reconnaissance vehicle will trigger an obstacle avoidance mechanism, causing adjacent unmanned reconnaissance vehicles to generate new movement directions and paths; once the unmanned reconnaissance vehicle enters the communication range allocated by the edge computing aggregator of unmanned combat equipment, it will be able to obtain all the information; Furthermore, the edge computing cluster of unmanned combat equipment can set up high-density and high-frequency reconnaissance missions in important strategic or tactical reconnaissance areas. That is, multiple unmanned reconnaissance tools need to be densely clustered in a certain important area for repeated reconnaissance and collaborative work. When multiple unmanned reconnaissance tools enter an important reconnaissance area, they constitute a specific edge computing cluster of unmanned combat equipment in that area. The unmanned reconnaissance tools within the cluster can communicate with each other, share their respective reconnaissance information and status, and the edge computing cluster of unmanned combat equipment can continuously optimize the cluster's reconnaissance missions until the high-density and high-frequency reconnaissance missions can meet the requirements of unmanned combat.

3. The system according to claim 1, characterized in that, In steps 1-3, the following sub-steps are used: Sub-step 1-3-1: Initialization of the unmanned reconnaissance tool cluster; Assuming the edge computing aggregate of unmanned combat equipment is composed of It consists of several unmanned reconnaissance vehicles, which contain An aggregate; The probability of an unmanned reconnaissance vehicle joining a cluster is given by the following formula: ； Generally speaking, when the number of group members does not exceed At the threshold, The value is greater than zero, otherwise it is equal to zero; where, Refers to unmanned reconnaissance tools In time The state space of time, Refers to unmanned reconnaissance tools In time The state space of time; Let $\frac{ ... ； Sub-step 1-3-2: Unmanned reconnaissance vehicles conduct patrol reconnaissance and collect reconnaissance data for the assigned tactical reconnaissance mission; The unmanned reconnaissance vehicle conducts patrol reconnaissance according to the tactical reconnaissance tasks assigned in sub-step 1-2-3 and the reconnaissance area in sub-step 1-2-4, ensuring that the patrol reconnaissance route covers the entire reconnaissance area, and uses cameras, infrared imaging equipment, radar, and listening equipment to collect reconnaissance data in the reconnaissance area during the patrol reconnaissance. Within a cluster, unmanned reconnaissance vehicles can obtain information about the number of unmanned reconnaissance vehicles in the cluster, their identification numbers, and probability parameters through local information exchange with neighboring unmanned reconnaissance vehicles. ； The constant represents the "willingness" of unmanned reconnaissance vehicles to leave the cluster; Furthermore, the unmanned reconnaissance tools in the cluster determine every second whether to leave the cluster and share and forward the reconnaissance data, thereby transmitting it back to the data center of their own unmanned combat system. Sub-step 1-3-3: The unmanned reconnaissance vehicle returns due to insufficient power or malfunction and the unfinished reconnaissance mission is handed over; When an unmanned reconnaissance vehicle detects low power or a malfunction, it reserves power for a return trip. If the reconnaissance mission is complete, it returns directly; if the mission is incomplete, it exchanges information with other unmanned reconnaissance vehicles in the cluster to obtain information about nearby available unmanned reconnaissance vehicles, and then calculates the probability of a nearby available unmanned reconnaissance vehicle joining the cluster. Furthermore, nearby unmanned reconnaissance vehicles need to calculate the probability of handing over unfinished reconnaissance tasks based on their remaining power, their own mission completion status, reconnaissance capabilities, reconnaissance range, and current location, and make a decision to join the cluster based on the calculation results. If unmanned reconnaissance vehicles within the sensing range do not join the cluster, they are prohibited from joining the cluster for a certain period of time. The edge computing system of the unmanned combat equipment then dispatches a suitable, fully powered unmanned reconnaissance vehicle to take over the unfinished reconnaissance mission, ensuring... The effectiveness of unmanned reconnaissance tools joining the cluster. The probability is: ； For the current moment, the aggregate Number of members; For aggregates The threshold; It is the largest aggregate that the system can form; When there are multiple aggregates in a group system This ensures fair competition among the various clusters, meaning that unmanned reconnaissance tools tend to join clusters with higher thresholds in order to maintain the balance among the clusters. Sub-steps 1-3-4: Enhanced reconnaissance of important targets; For important areas or targets, relying on a single unmanned reconnaissance vehicle is far from sufficient. Often, two or more unmanned reconnaissance vehicles are needed to conduct continuous and intensive reconnaissance of the area from multiple angles and for extended periods, using various reconnaissance methods such as cameras, infrared imaging equipment, radar, and listening devices. In unit time Within the area, the probability of any two unmanned reconnaissance vehicles meeting is The probability of each unmanned reconnaissance vehicle encountering any other unmanned reconnaissance vehicle is... Therefore: ； The area of ​​the enclosed region; The average speed of the unmanned reconnaissance vehicle. The sensing radius of unmanned reconnaissance tools; for The area swept by unmanned reconnaissance vehicles within a given time period; The Markov process of the unit unmanned reconnaissance tool constitutes the dynamic aggregation process of the edge computing aggregate of unmanned combat equipment; by utilizing the probability parameters of the unmanned reconnaissance tool, we can obtain the average number of members in each aggregate at any time point; the change in the average number of members in the aggregate is affected by other aggregates in the system and state transition parameters. Sub-steps 1-3-5: Optimization of the distribution of unmanned reconnaissance tools within the cluster; At any given time, the distribution of unmanned reconnaissance tools within each cluster of the unmanned combat equipment edge computing aggregate can be described by difference equations: ； Represents unmanned reconnaissance tools encountering clusters within a unit of time. and the average number added; The number of unmanned reconnaissance vehicles representing the roaming status per unit time; This indicates that unmanned reconnaissance vehicles have joined the cluster. The probability of; Indicates an aggregate The initial state; This indicates the aggregate per unit time. The average number of members leaving the cluster; When the system reaches a steady state, we can obtain: By substituting the parameters of the unmanned reconnaissance tool into the difference equation, the ratio of member quotas of any two aggregates when the system is in equilibrium can be calculated. ； Sub-steps 1-3-6 identify friendly defensive targets and enemy attack targets; The edge computing aggregate of unmanned combat equipment calculates the friendly defense zone and the enemy's strike zone based on user combat instructions and reconnaissance data from sub-step 1-3-2. It then runs artificial intelligence algorithms to identify targets in the friendly defense zone and the enemy's strike zone from reconnaissance data collected by cameras, infrared imaging equipment, radar, and listening devices, as well as targets approaching or intruding into the friendly defense zone. This includes infrared signals from combat personnel, equipment, and military dogs; images and radar signals from firepower configurations, field fortifications, buildings, landmines, camouflage, bridges, unmanned combat equipment, and suspicious sound signals; and calculates the identification probability of friendly defense targets and enemy strike targets. First, it compares the identification probabilities; targets with higher probabilities have higher identification rates and better reconnaissance results. Second, it calculates the priority based on the size and threat level of the identified targets; targets with higher priority are more important. Finally, it checks whether all targets in all tactical reconnaissance missions have been identified. If so, target identification terminates; otherwise, it sequentially performs target identification, supplementing reconnaissance of targets with low identification probabilities. Sub-steps 1-3-7: Virtual reconnaissance exercise to confirm and optimize the reconnaissance mission; Furthermore, the edge computing aggregate of unmanned combat equipment can perform virtual reconnaissance exercises, verifying the rationality and feasibility of reconnaissance task allocation through online computation, and identifying and correcting errors and deviations before executing reconnaissance tasks. To test the system's stability, accuracy, and scalability, the edge computing aggregate of unmanned combat equipment can select different model parameters for calculation and comparative analysis, and can also adjust parameters accordingly. To analyze the relationship between time and the number of crew members in unmanned reconnaissance vehicles; when the number of unmanned reconnaissance vehicles is AND equals the cluster threshold Adjusting model parameters while keeping the number of simulations constant. The study investigated the trend of the average number of clusters over time; and examined the changes in the number of unmanned reconnaissance vehicles in the system. And adjust the aggregate threshold accordingly. This allows us to obtain the relationship between the average number of clusters and time under different numbers of unmanned reconnaissance vehicles; Furthermore, the edge computing aggregate of unmanned combat vehicles can optimize the execution of reconnaissance missions on the edge computing aggregate, that is, to calculate and optimize the aggregation and balance of multiple unmanned reconnaissance vehicles online, i.e., to optimize the arrangement within the work area. The unmanned reconnaissance vehicles were reassigned, and aggregation requests were initiated when the time threshold for completing the reconnaissance mission was more optimized; this was achieved by modifying the threshold and model parameters. To conduct simulation experiments, namely virtual reconnaissance exercises, can optimize the average size and quota ratio of the aggregate at different times before the reconnaissance mission is completed, better accelerate the group convergence process of multiple unmanned reconnaissance tools, and provide decision-making basis for the execution prediction of unmanned reconnaissance missions, the group convergence characteristics of unmanned reconnaissance tools, and the determination of the parameters of the group system. Sub-steps 1-3-8 complete all tactical and strategic reconnaissance missions; Each unmanned reconnaissance vehicle checks the execution status of its reconnaissance mission according to the task assignment results. After completing all reconnaissance missions, it reports the reconnaissance data and mission completion status to the unmanned combat equipment edge computing aggregate. If there are no new missions, it returns to its own base. If the unmanned combat equipment edge computing aggregate assigns a new reconnaissance mission, it skips to step 1 to execute the new reconnaissance mission. If the unmanned combat equipment edge computing aggregate assigns a new combat mission, it skips to step 2 to execute the new combat mission. If the unmanned combat equipment edge computing aggregate assigns a new battle results and damage assessment mission, it skips to step 3 to execute the new battle results and damage assessment mission.

4. The system according to any one of claims 1 to 3, characterized in that, Step 2 includes the following sub-steps: Sub-step 2-1: The edge computing aggregate of unmanned combat equipment formulates combat missions. Based on user combat instructions and the friendly defense targets and enemy attack targets identified in sub-steps 1-3-6, the edge computing aggregate of unmanned combat equipment formulates combat missions, including combat area, combat methods, combat time, combat intensity, combat accuracy, specific combat targets, available combat tools and combat forces. Preferably, intelligent unmanned combat uses unmanned combat tools, including drones, remote-controlled aircraft, unmanned vehicles, remote-controlled cars, robots, robot dogs, unmanned surface vessels, and unmanned underwater vehicles, and uses their onboard computers to form an edge computing aggregate of unmanned combat equipment to calculate combat missions. Sub-step 2-2: The edge computing aggregate of unmanned combat equipment checks whether the combat mission has been reconnoitered. The edge computing aggregate of unmanned combat equipment automatically acquires the relevant reconnaissance data from step 1 based on the input combat mission, checks whether the combat area and combat target corresponding to the combat mission have been reconnoitered. If so, it proceeds to steps 2-3. If there are combat areas and combat targets that have not been reconnoitered, or if the reconnaissance targets are moving at high speed, it sets the combat area and combat target as the reconnaissance area and reconnaissance target, and returns to step 1 to deploy reconnaissance missions to supplement reconnaissance. Sub-steps 2-3: The edge computing aggregate of unmanned combat equipment intelligently allocates combat tasks; the edge computing aggregate of unmanned combat equipment selects available intelligent unmanned combat tools according to the user's combat instructions, and decomposes the combat tasks using the embedded edge computing platform and artificial intelligence technology of the intelligent unmanned combat tools; the decomposed targets include more specific combat areas, combat methods, combat time, combat intensity, combat precision, specific combat objectives, available combat tools and combat forces; Sub-steps 2-4: The edge computing aggregate of unmanned combat equipment intelligently executes combat missions; the edge computing aggregate of unmanned combat equipment coordinates the execution of combat missions by various unmanned combat tools according to the decomposed combat missions, and runs artificial intelligence algorithms in the execution process to cope with the ever-changing situation on the battlefield.

5. The system according to claim 4, characterized in that, In step 2-1, the following steps are adopted: Sub-step 2-1-1: Initialization of the intelligent combat system; Preferably, an intelligent combat system is initialized, including... An unmanned combat vehicle, This indicates the combat mission number assigned to each unmanned combat vehicle; for each unmanned combat vehicle state space It is represented using a Markov chain: ； During initialization This indicates that the unmanned combat vehicle is in a state of unassigned mission. Sub-step 2-1-2: The edge computing aggregate of unmanned combat equipment allocates available unmanned combat tools according to user combat instructions and combat missions; Unmanned combat vehicles (UCVs) use edge computing to broadcast user combat commands and missions. UCVs receiving these broadcasts aggregate via wireless networking, meaning they are geographically dispersed, but each can exchange information with other UCVs within the aggregate to obtain relevant information about the broadcast missions. The probability of joining the mission is calculated based on the UCV's combat capabilities, operational range, and current location. Next, the unmanned combat vehicle generates a random number and... Comparisons are made, and decisions are made based on the comparison results, until a sufficient number of unmanned combat vehicles are allocated for the combat mission. Sub-step 2-1-3: Edge computing aggregate computing of unmanned combat equipment for combat missions; The edge computing aggregate of unmanned combat equipment, based on user combat commands and the friendly and enemy attack targets identified in sub-step 1-3-6, formulates friendly defensive combat missions and enemy attack missions, including the combat area, combat methods, combat time, combat intensity, combat accuracy, specific combat targets, available combat tools and combat forces, etc.; preferably, intelligent unmanned combat uses unmanned combat tools, including drones, remote-controlled aircraft, unmanned vehicles, remote-controlled cars, robots, robot dogs, unmanned surface vessels, and unmanned underwater vehicles; preferably, priority is given to defense in sub-step 1-3-6. The objectives are: to prioritize friendly defensive targets or to strike enemy targets with high priority in sub-steps 1-3-6; preferably, the friendly defensive mission includes constructing 1-3 lines of defense, digging field fortifications for each line of defense, optimizing the combination of unmanned combat vehicles and firepower configurations to ensure that firepower can cover the entire friendly defensive area, and actively striking targets that approach or intrude into the friendly defensive area; preferably, the enemy strike mission includes 1-3 rounds of strikes to ensure that firepower can cover all enemy targets in the entire enemy strike area, and to carry out saturation strikes on important targets, fortified targets, as well as hidden targets and newly added targets that were not discovered during reconnaissance. In steps 2-3, the following sub-steps are used: Sub-step 2-3-1: Construct an edge computing aggregate of unmanned combat equipment based on the combat mission; Unmanned combat vehicles that receive combat mission broadcasts gather through wireless networking to form an edge computing cluster of unmanned combat equipment. Sub-step 2-3-2: Division of operational areas; Edge computing and aggregate computing for unmanned combat equipment to divide the combat zone, using a radius of... The combat zone is described by a two-dimensional circle; unmanned combat vehicles are considered as a circle with a radius of... If the distance between the two centers is less than Therefore, it can be assumed that unmanned combat vehicles have unlimited communication and positioning capabilities; adjacent unmanned combat vehicles can quickly exchange information and determine each other's positions; Based on the operational area, the operational mission is broken down into several sets of operational sub-missions. Each combat sub-task corresponds to a specific combat area, including the detailed coordinates of that area, friendly defensive targets, and enemy attack targets; Sub-step 2-3-3: Assign combat sub-tasks to appropriate unmanned combat vehicles; The operational tasks of the unmanned combat vehicle (UCV) edge computing aggregate broadcast sub-step 2-1-3 are determined from the operational sub-tasks of friendly defense and enemy attack, operational methods, operational time, operational intensity, operational accuracy, specific operational targets, available operational tools and forces. Based on the UCV's operational capabilities, operational range, and current location, the UCVs are assigned operational sub-tasks they are capable of performing, and the calculations are performed for each UCV. state space It is represented using a Markov chain: ； at this time This indicates that the unmanned combat vehicle is in a state of assigned mission, and the corresponding number in the set is the assigned combat sub-mission number; Sub-steps 2-3-4: Unmanned combat vehicles acquire the combat sub-task combat area; First, the unmanned combat vehicle obtains the operational area of ​​the corresponding sub-task based on the assigned operational sub-task number, including the detailed geographical coordinates of that area; when the distance between the unmanned combat vehicle and the circular area exceeds... We assume that the unmanned combat vehicle has reached the boundary of the area; we assume that the unmanned combat vehicle can obtain its current absolute coordinates and calculate its new direction of motion by generating a new turning angle; we assume that the unmanned combat vehicle does not need to go through acceleration and deceleration phases, avoid communication conflicts and other hardware limitations during its movement, and has no issues such as turning errors. Secondly, unmanned combat vehicles use a greedy algorithm to plan their combat routes, that is, they prioritize visiting the combat sub-tasks closest to their own location, and then visit the combat sub-tasks farther away from them, until all combat sub-tasks have been visited. If the distance between unmanned combat vehicles is less than If the number of unmanned combat vehicles reaches a certain multiple and no aggregation request is received, the edge computing platform dynamically adjusts the combat mission and runs a path planning algorithm on the unmanned combat vehicle. The unmanned combat vehicle will trigger an obstacle avoidance mechanism, causing adjacent unmanned combat vehicles to generate new movement directions and paths. Once the unmanned combat vehicle enters the communication range allocated by the edge computing aggregation of unmanned combat equipment, it will be able to obtain all the information and participate in the combat sub-mission. Furthermore, the edge computing aggregate of unmanned combat vehicles can set high-density and high-frequency combat sub-tasks for important combat areas or targets. That is, multiple unmanned combat vehicles need to be densely gathered in a certain important position to repeatedly fight and work together. When multiple unmanned combat vehicles enter an important combat position, they constitute a specific edge computing aggregate of unmanned combat vehicles at that position. The unmanned combat vehicles within the aggregate can communicate with each other, share their respective combat information and status, and the edge computing aggregate of unmanned combat vehicles can continuously optimize the aggregate's combat sub-tasks until the high-density and high-frequency combat sub-tasks can meet the combat requirements.

6. The system according to claim 4, characterized in that, In steps 2-4, the following sub-steps are used: Sub-step 2-4-1: Initialization of the unmanned combat vehicle cluster; Assuming the edge computing aggregate of unmanned combat equipment is composed of It consists of several unmanned combat vehicles, which contain An aggregate; The probability of an unmanned combat vehicle joining a cluster while roaming is: ； When the number of group members does not exceed At the threshold, The value is greater than zero, otherwise it is equal to zero; where, Refers to unmanned combat vehicles In time The state space of time, Refers to unmanned combat vehicles In time The state space of time; The probability of an unmanned combat vehicle leaving the cluster is: ； Sub-step 2-4-2: Unmanned combat vehicles perform defense or attack according to the assigned combat sub-tasks; According to the combat sub-tasks assigned in sub-step 2-3-3 and the combat area in sub-step 2-3-4, unmanned combat vehicles conduct patrol defense or direct strikes, ensuring that the patrol defense route covers the entire combat area defended by the friendly forces or attacked by the enemy, and that direct strikes can effectively destroy the targets corresponding to the combat sub-tasks. During patrol defense, cameras, infrared imaging equipment, radar, and listening equipment are used to identify friend or foe in the combat area, ensuring that the enemy is struck quickly without harming friendly forces. Within a cluster, unmanned combat vehicles can obtain information such as the number of unmanned combat vehicles in the group, their serial numbers, and probability parameters through local information transmission between adjacent unmanned combat vehicles. ； The constant represents the "willingness" of unmanned combat vehicles to leave the cluster; Furthermore, the unmanned combat vehicles in the cluster determine every second whether to leave the cluster and share and forward reconnaissance data so that multiple unmanned combat vehicles can carry out coordinated defense or coordinated attack. Sub-steps 2-4-3: The unmanned combat vehicle returns due to insufficient power or malfunction and the handover of unfinished combat sub-tasks; When an unmanned combat vehicle detects low power or a malfunction, it will reserve power for a return trip. If the combat sub-mission has been completed, it will return directly; if the combat sub-mission has not been completed, it will exchange information with other unmanned combat vehicle members in the cluster to obtain relevant information about nearby available unmanned combat vehicles, and then calculate the probability of the incomplete combat mission handover, i.e., the probability of a nearby available unmanned combat vehicle joining the cluster. The available unmanned combat vehicles in the vicinity need to calculate the possibility of handing over unfinished combat missions based on their remaining power, mission completion status, combat capabilities, combat range, and current location, and make a decision to join the cluster based on the calculation results. If unmanned combat vehicles within the sensing range do not join the cluster, they are prohibited from joining the cluster for a certain period of time. The edge computing of the unmanned combat equipment will then dispatch a suitable unmanned combat vehicle with sufficient power to take over the unfinished combat sub-tasks, ensuring... The effectiveness of unmanned combat vehicles joining the cluster The probability is: ； For the current moment, the aggregate Number of members; For aggregates The threshold; It is the largest aggregate that the system can form; When there are multiple aggregates in a group system This ensures fair competition among the various clusters, meaning that unmanned combat vehicles tend to join clusters with higher thresholds in order to maintain the balance among the clusters; Sub-step 2-4-4: The unmanned combat vehicle calls for fire support; If the unmanned combat vehicle finds that the combat sub-tasks assigned in sub-step 2-3-3 cannot be completed, including when its own casualties are too great to continue effective combat, when multiple lines of defense in its own defensive area are captured by the enemy and cannot be driven out, when its own defensive targets are attacked or destroyed, when the enemy's targets are not effectively destroyed after several rounds of attacks, or when a large number of enemy reinforcements appear, it needs to call for fire support from the edge computing cluster of the unmanned combat vehicle and dispatch unmanned combat vehicles with the same or greater lethality or targeted lethality to enhance its ability to complete the combat sub-tasks; In unit time Within the space, the probability of any two unmanned combat vehicles encountering each other is... The probability of each unmanned combat vehicle encountering any other unmanned combat vehicle is: Therefore: ； The area of ​​the enclosed region; The average speed of the unmanned combat vehicle; To determine the sensing radius of unmanned combat vehicles; for The area swept by unmanned combat vehicles within a given time period; The Markov process of a unit unmanned combat vehicle constitutes the dynamic aggregation process of the edge computing aggregate of unmanned combat equipment; by utilizing the probability parameters of the unmanned combat vehicle, we can obtain the average number of members in each aggregate at any time point; the change in the average number of members in the aggregate is affected by other aggregates in the system and state transition parameters. Sub-steps 2-4-5: Optimization of the distribution of unmanned combat vehicles within the cluster; At any given time, the distribution of unmanned combat vehicles within each aggregate of edge computing clusters in the system can be described by difference equations: ； Represents unmanned combat vehicles encountering clusters within a unit of time. and the average number added; This indicates the number of unmanned combat vehicles in a roaming state per unit of time. This indicates that unmanned combat vehicles have joined the cluster. The probability of; Indicates an aggregate The initial state; This indicates the aggregate per unit time. The average number of members leaving the cluster; When the system reaches a steady state, we can obtain: ； By substituting the parameters of unmanned combat vehicles into the difference equation, the ratio of member quotas of any two aggregates when the system is in equilibrium can be calculated. ； Sub-steps 2-4-6 involve multi-layered defense and multiple rounds of attacks to complete the combat mission; During the execution of combat missions, the edge computing aggregate of unmanned combat equipment adheres to the principle of multi-layered defense and multi-round attack, making full use of the remaining unmanned combat tools to block the enemy from attacking friendly defensive targets, and making every effort to attack the enemy's strike targets to ensure the completion of combat missions. Sub-steps 2-4-7 involve virtual combat exercises to confirm and optimize combat missions; The edge computing aggregate of unmanned combat equipment can compute virtual combat exercises on its own, verifying the rationality and feasibility of combat mission allocation through online computation, and identifying and correcting errors and deviations before the execution of combat missions. To test the system's stability, accuracy, and scalability, the edge computing aggregate can select different model parameters for calculation and comparative analysis, and can adjust these parameters as needed. To analyze the relationship between time and the number of crew members in unmanned combat vehicles; when the number of unmanned combat vehicles is AND equals the cluster threshold Adjusting model parameters while keeping the number of simulations constant. The study also investigated the trend of the average number of clusters over time; and examined the changes in the number of unmanned combat vehicles in the system. And adjust the aggregate threshold accordingly. This allows us to obtain the relationship between the average number of clusters and time under different numbers of unmanned combat vehicles; The edge computing aggregate of unmanned combat vehicles can perform calculations and optimize the execution of combat missions on the edge computing aggregate of unmanned combat vehicles, that is, to calculate and optimize the aggregation and balance of multiple unmanned combat vehicles online, that is, to optimize the arrangement within the working area. The unmanned combat vehicles are reassigned, and aggregation requests are initiated when the time threshold for completing the combat mission is more favorable; by modifying the threshold and model parameters... To conduct simulation experiments, namely virtual combat exercises, can optimize the average size and quota ratio of the aggregate at different times before the completion of combat missions, better accelerate the group convergence process of multiple unmanned combat vehicles, and provide decision-making basis for the execution prediction of unmanned combat missions, the group convergence characteristics of unmanned combat vehicles, and the determination of group system parameters. Sub-steps 2-4-8 complete the operation; After each unmanned combat vehicle completes all combat missions, it reports combat data and mission completion status to the unmanned combat equipment edge computing cluster. If there are no new missions, it returns to its own base. If the unmanned combat equipment edge computing cluster assigns a new reconnaissance mission, it skips to step 1 to execute the new reconnaissance mission. If the unmanned combat equipment edge computing cluster assigns a new combat mission, it skips to step 2 to execute the new combat mission. If the unmanned combat equipment edge computing cluster assigns a new battle results and damage assessment mission, it skips to step 3 to execute the new battle results and damage assessment mission.

7. The system according to claim 1, 2, 3, 5, or 6, characterized in that, Step 3 includes the following sub-steps: Sub-step 3-1: The edge computing aggregate of unmanned combat equipment calculates the battle results and damage assessment task; the edge computing aggregate of unmanned combat equipment calculates the battle results and damage assessment task according to the user's combat instructions and the friendly defense targets and enemy attack targets identified in sub-step 1-3-6, including the battle results and damage assessment area, battle results and damage assessment methods, battle results and damage assessment time, battle results and damage assessment intensity, battle results and damage assessment accuracy, specific battle results and damage assessment targets, available battle results and damage assessment tools, and battle results and damage assessment forces; preferably, intelligent unmanned battle results and damage assessment uses unmanned reconnaissance tools, including drones, remote-controlled aircraft, unmanned vehicles, remote-controlled cars, robots, robot dogs, surface unmanned boats, and underwater unmanned submersibles, and uses their onboard computers to form an edge computing aggregate of unmanned combat equipment to calculate the battle results and damage assessment task; Sub-step 3-2: The edge computing aggregate of unmanned combat equipment intelligently allocates battle results and damage assessment tasks; the edge computing aggregate of unmanned combat equipment selects available intelligent unmanned reconnaissance tools according to the user's combat instructions, and decomposes the battle results and damage assessment tasks using the embedded edge computing platform and artificial intelligence technology of the intelligent unmanned reconnaissance tools; the decomposed targets include more specific battle results and damage assessment areas, battle results and damage assessment methods, battle results and damage assessment time, battle results and damage assessment intensity, battle results and damage assessment accuracy, specific battle results and damage assessment targets, available battle results and damage assessment tools, and battle results and damage assessment forces; Sub-step 3-3: The edge computing aggregate of unmanned combat equipment intelligently executes the task of assessing the results and damage of combat. According to the decomposed task of assessing the results and damage of combat, the edge computing aggregate of unmanned combat equipment is jointly executed by various unmanned reconnaissance tools to calculate the target loss in the friendly defense area and the enemy's strike area, that is, the damage to the target before and after the start of the combat mission. During the execution process, artificial intelligence algorithms are run to cope with the ever-changing situation on the battlefield.

8. The system according to claim 7, characterized in that, In sub-step 3-1, the following sub-steps are used: Sub-step 3-1-1: Initialize the intelligent combat system; Initialize an intelligent combat system, including A tool for assessing the combat results and damage of unmanned aerial vehicles. This indicates the task number assigned to the unmanned combat results and damage assessment tool; for each unmanned combat results and damage assessment tool state space It is represented using a Markov chain: ； During initialization This indicates that the battle results and damage assessment tool is in an unassigned task state; Sub-step 3-1-2: The edge computing aggregate of unmanned combat equipment allocates available unmanned combat results and damage assessment tools according to user combat instructions and combat results and damage assessment tasks. The edge computing aggregate of unmanned combat equipment broadcasts user combat commands and unmanned combat results and damage assessment tasks. Unmanned combat results and damage assessment tools that receive the broadcast of unmanned combat results and damage assessment tasks are aggregated through wireless networking. That is, they are geographically dispersed, but each individual can exchange information with other unmanned combat results and damage assessment tool members in the aggregate to obtain relevant information about the broadcast unmanned combat results and damage assessment tasks. The probability of joining the unmanned combat results and damage assessment task is calculated based on the reconnaissance capabilities, reconnaissance range, and current location of the unmanned combat results and damage assessment tools until a sufficient number of unmanned combat results and damage assessment tools are allocated for the unmanned combat results and damage assessment task; Sub-step 3-1-3, the edge computing aggregate of unmanned combat equipment calculates the combat results and damage assessment task; The edge computing aggregate of unmanned combat equipment formulates tasks for assessing friendly combat losses and enemy strike results based on user combat instructions and the friendly defense targets and enemy strike targets identified in sub-steps 1-3-6. This includes the combat areas of friendly defense and enemy strike, available combat tools and forces, specific combat targets, high-priority friendly defense targets, and high-priority enemy strike targets. Furthermore, it calculates the number and location of targets for assessing friendly combat losses and enemy strike results.

9. The system according to claim 7, characterized in that, In sub-step 3-2, the following sub-steps are used: Sub-step 3-2-1: Construction of edge computing aggregates for unmanned combat equipment; Unmanned combat results and damage assessment tools that receive the combat results and damage assessment mission broadcast gather through wireless networking to form an edge computing aggregation of unmanned combat equipment. Sub-step 3-2-2: Delineation of battle results and damage assessment areas; Edge computing aggregates of unmanned combat equipment are used for combat result and damage assessment and area division, utilizing a radius of... A two-dimensional circle is used to describe the area for assessing battle results and damage; the battle results and damage assessment tool is considered as a circle with a radius of... If the distance between the two centers is less than Therefore, it can be assumed that the combat results and damage assessment tools have unlimited communication and positioning capabilities; adjacent combat results and damage assessment tools can quickly exchange information and determine each other's positions. Based on the area for assessing battle results and damage, the task of assessing battle results and damage is broken down into several sets of tasks. Each battle result and damage assessment task corresponds to a specific battle result and damage assessment area, including the detailed coordinates of that area; Sub-step 3-2-3 assigns the task of assessing battle results and damage to a suitable unmanned battle results and damage assessment tool; The edge computing aggregate broadcast sub-step 3-2-2 of the unmanned combat equipment's battle results and damage assessment task involves assessing the battle results and damage of both friendly and enemy attacks from the perspectives of friendly defense and enemy attack results and damage assessment tasks. This includes the assessment methods, assessment time, assessment intensity, assessment accuracy, specific assessment targets, available assessment tools, and assessment forces. Based on the unmanned battle results and damage assessment tools' capabilities, assessment range, and current location, the task assigns them the battle results and damage assessment tasks they are capable of performing. Finally, it calculates the battle results and damage assessment tasks for each unmanned battle results and damage assessment tool. state space It is represented using a Markov chain: ； at this time This indicates that the unmanned combat results and damage assessment tool is in the state of being assigned a task, and the corresponding number in the set is the assigned combat results and damage assessment task number; Sub-step 3-2-4: Obtain the battle results and damage assessment area; First, the unmanned combat damage assessment tool obtains the corresponding combat damage assessment area, including its detailed geographic coordinates, based on the assigned combat damage assessment task number. When the distance between the combat damage assessment tool and the circular area exceeds... It is assumed that the battle results and damage assessment tool has reached the area boundary; assume that the battle results and damage assessment tool can obtain the current absolute coordinates and calculate the new direction of movement by generating a new turning angle; assume that the battle results and damage assessment tool does not need to go through acceleration and deceleration phases during its movement, avoids hardware limitations such as communication conflicts, and has no turning errors. Secondly, the unmanned reconnaissance tools use a greedy algorithm to plan their battle damage assessment routes. This means they prioritize visiting the battle damage assessment tasks closest to their current location, then those further away, until all battle damage assessment tasks have been visited. If the distance between the battle damage assessment tools is less than [a certain value], [they will proceed as planned]. 1.2 times that of the unmanned combat equipment edge computing platform and without receiving aggregating requests, the edge computing platform dynamically adjusts combat missions and runs path planning algorithms on the battle results and damage assessment tool. The battle results and damage assessment tool will trigger an obstacle avoidance mechanism, causing adjacent battle results and damage assessment tools to generate new movement directions and paths. Once the battle results and damage assessment tool enters the communication range allocated by the edge computing aggregator of unmanned combat equipment, it will be able to obtain all information and participate in the battle results and damage assessment mission. Furthermore, the edge computing aggregate of unmanned combat equipment can set high-density and high-frequency battle result and damage assessment tasks for important battle result and damage assessment areas or targets. That is, multiple unmanned battle result and damage assessment tools need to be densely clustered in a certain important position to repeatedly assess battle results and damage and work together. When multiple unmanned battle result and damage assessment tools enter an important battle result and damage assessment position, they form a specific unmanned combat equipment edge computing aggregate at that position. The unmanned battle result and damage assessment tools within the aggregate can communicate with each other, share their respective battle result and damage assessment information and their respective statuses. The unmanned combat equipment edge computing aggregate can continuously optimize the aggregate's battle result and damage assessment tasks until the high-density and high-frequency battle result and damage assessment tasks can meet the operational requirements.

10. The system according to claim 7, characterized in that, In step 3-3, the following steps are adopted: Sub-step 3-3-1: Initialization of the battle results and damage assessment tool aggregate; Assuming the edge computing aggregate of unmanned combat equipment is composed of It consists of a set of tools for assessing battle results and damage, which include An aggregate; The probability of the battle results and damage assessment tool being added to a cluster, under roaming conditions, is: ； When the number of group members does not exceed At the threshold, The value is greater than zero, otherwise it is equal to zero; where, This refers to unmanned combat results and damage assessment tools. In time The state space of time, This refers to unmanned combat results and damage assessment tools. In time The state space of time; The probability of the battle results and damage assessment tool leaving the cluster is: [The probability of the battle results and damage assessment tool leaving the cluster under clustered conditions is:] ； Sub-step 3-3-2: The battle results and damage assessment tool conducts a round of battle results and damage assessments on the assigned battle results and damage assessment tasks and collects battle results and damage assessment data. The unmanned combat results and damage assessment tool conducts a patrol-based combat results and damage assessment based on the combat results and damage assessment tasks assigned in sub-step 3-2-3 and the combat results and damage assessment areas in sub-step 3-2-4. It ensures that the patrol-based combat results and damage assessment route covers the entire combat results and damage assessment area, and uses cameras, infrared imaging equipment, radar, and listening equipment to collect combat results and damage assessment data in the combat results and damage assessment area during the patrol-based combat results and damage assessment. Within a cluster, the unmanned combat results and damage assessment tool can obtain the number of unmanned combat results and damage assessment tools, the number of unmanned combat tools, and probability parameters in the cluster through local information transmission between adjacent unmanned combat results and damage assessment tools. ； The constant represents the "willingness" of the battle results and damage assessment tool to leave the cluster; the battle results and damage assessment tool in the cluster will determine whether to leave the cluster every second, and share and forward the battle results and damage assessment data, thereby transmitting it back to the data center of its own unmanned combat system. Sub-step 3-3-3: The unmanned combat results and damage assessment tool returns due to insufficient power or malfunction and the unfinished combat results and damage assessment task is handed over. When a battle damage assessment tool detects insufficient power or a malfunction, it will reserve power for a return trip. If the battle damage assessment mission has been completed, it will return directly. If the mission has not been completed, it will exchange information with other unmanned battle damage assessment tools in the cluster to obtain relevant information about nearby available unmanned battle damage assessment tools. It will then calculate the probability of a failed mission handover, i.e., the probability of a nearby available unmanned battle damage assessment tool joining the cluster. The available unmanned combat damage assessment tools in the vicinity need to calculate the possibility of handing over unfinished combat damage assessment tasks based on their remaining power, their own mission completion status, combat damage assessment capabilities, combat damage assessment range, and current location, and make a decision to join the cluster based on the calculation results. If a combat damage assessment tool within the perception range does not join the cluster, it will be prohibited from joining the cluster for a certain period of time. The edge computing cluster of unmanned combat equipment will then dispatch a suitable, fully powered unmanned combat damage assessment tool to take over the unfinished combat damage assessment task, ensuring... The effectiveness of the battle results and damage assessment tool was incorporated into the aggregate. The probability is: ； For the current moment, the aggregate Number of members; For aggregates The threshold; It is the largest aggregate that the system can form; When there are multiple aggregates in a group system This ensures fair competition among the various clusters; that is, the battle results and damage assessment tool tends to include clusters with higher thresholds in order to maintain the balance among the clusters. Sub-step 3-3-4: Enhance the assessment of combat results and damage for important targets; For important areas or targets, relying on a single unmanned combat results and damage assessment tool is far from sufficient. Often, two or more unmanned combat results and damage assessment tools are needed to conduct continuous and intensive combat results and damage assessments of the assessment area from multiple angles and over a long period of time using various methods such as cameras, infrared imaging equipment, radar, and listening equipment. In unit time Within this context, the probability of any two battle result / damage assessment tools meeting is... The probability that each battle result / damage assessment tool encounters any other battle result / damage assessment tool is: Therefore: ； The area of ​​the enclosed region; The average speed of the movement of the tool for assessing battle results and damage; The perception radius of the battle results and damage assessment tool; for The area covered by the battle damage assessment tool within a given time period; The Markov process of the unit battle results and damage assessment tool constitutes the dynamic aggregation process of the edge computing aggregate of unmanned combat equipment; by using the probability parameters of the battle results and damage assessment tool, we can obtain the average number of members in each aggregate at any time point; the change in the average number of members in the aggregate is affected by other aggregates in the system and state transition parameters. Sub-step 3-3-5: Optimization of the distribution of battle results and damage assessment tools within the cluster; At any given time, the distribution of unmanned combat vehicles within each aggregate of edge computing clusters in the system can be described by difference equations: ； The tool for assessing battle results and damage per unit time encounters an aggregate. and the average number added; The number of tools for assessing combat results and damages representing the roaming status per unit of time. This indicates that the battle results and damage assessment tool has been incorporated into the cluster. The probability of; Indicates an aggregate The initial state; This indicates the aggregate per unit time. The average number of members leaving the cluster; When the system reaches a steady state, we can obtain: ； By substituting the parameters of the battle results and losses assessment tool into the difference equation, the ratio of member quotas of any two aggregates at system equilibrium can be calculated. ； Sub-step 3-3-6: Calculate the battle damage to our own defensive targets and the results of the enemy's attack on the targets; The edge computing aggregate of unmanned combat equipment calculates the target losses in the friendly defense zone and the enemy's strike zone based on the battle results and damage assessment tasks assigned in sub-step 3-2-3 and the battle results and damage assessment areas in sub-step 3-2-4, i.e., the damage to targets before and after the start and end of the combat mission; it runs artificial intelligence algorithms to identify targets in the friendly defense zone and the enemy's strike zone from the battle results and damage assessment data collected by cameras, infrared imaging equipment, radar, and listening equipment, as well as targets approaching or intruding into the friendly defense zone, including infrared signals of combat personnel, equipment, and military dogs, firepower configurations, field fortifications, buildings, landmines, camouflage, bridges, images and radar signals of unmanned combat equipment, and suspicious sound signals. The process involves calculating the probability of loss for both defensive and offensive targets. First, a comparison of these probabilities is performed; targets with higher probabilities have a higher loss rate and are more severely impacted by attacks, while those with lower probabilities have a better defensive effect. Second, priorities are calculated based on the size and threat level of the identified targets; higher-priority targets are more important, and high-priority defensive targets should have the lowest possible probability of loss (i.e., lower damage), while high-priority offensive targets should have the highest possible probability of loss (i.e., higher results). Finally, it checks whether all targets in the results and damage assessment task have been calculated. If so, the results and damage assessment calculation is terminated; otherwise, the results and damage assessment calculation is performed sequentially, with supplementary results and damage assessments conducted for targets with low identification probabilities that are difficult to identify. Sub-step 3-3-7, virtual battle results and damage assessment exercise to confirm and optimize the battle results and damage assessment task; The edge computing aggregate of unmanned combat equipment can initially assess the results and losses on its own. That is, based on the relevant reconnaissance data in step 1 and the relevant combat data in step 2, it compares the results and losses assessment task results in sub-step 3-3-6 to initially calculate the results and losses assessment data. The edge computing aggregate of unmanned combat equipment can perform virtual battle result and damage assessment exercises on the edge computing aggregate platform. This means that it verifies the rationality and feasibility of battle result and damage assessment task allocation through online calculations, identifying and correcting errors and deviations before executing the assessment tasks. To test the system's stability, accuracy, and scalability, the edge computing aggregate of unmanned combat equipment can select different model parameters for calculation and comparative analysis, and can also adjust the parameters... To analyze the relationship between time and the number of members in the unmanned combat results and damage assessment tool; when the number of members in the combat results and damage assessment tool is... AND equals the cluster threshold Adjusting model parameters while keeping the number of simulations constant. The study investigated the trend of the average number of clusters over time; and examined the changes in the number of battle results and damage assessment tools in the system. And adjust the aggregate threshold accordingly. It can obtain the relationship between the average number of clusters and time under different numbers of battle results and damage assessment tools; The edge computing aggregate of unmanned combat equipment can perform calculations and optimize the execution of combat result and damage assessment tasks on the edge computing aggregate of unmanned combat equipment. That is, it can calculate and optimize the aggregation and balance of multiple unmanned combat result and damage assessment tools online, that is, optimize the arrangement within the working area. The unmanned aerial vehicle (UAV) combat results and damage assessment tool was reassigned, and an aggregation request was initiated when the time threshold for completing the combat results and damage assessment task was more optimal; by modifying the threshold and model parameters... To conduct simulation experiments, namely virtual battle results and damage assessment exercises, can optimize the average size and quota ratio of the aggregate at different times before the battle results and damage assessment task is completed, better accelerate the group convergence process of multiple unmanned battle results and damage assessment tools, and provide decision-making basis for the execution prediction of unmanned reconnaissance battle results and damage assessment tasks, the group convergence characteristics of unmanned battle results and damage assessment tools, and the determination of the parameters of the group system. Sub-step 3-3-8, the assessment of battle results and losses is completed; After each unmanned combat damage assessment tool completes all combat damage assessment tasks, it reports the combat damage assessment data and task completion status to the unmanned combat equipment edge computing aggregate. The unmanned combat equipment edge computing aggregate determines whether the status of friendly defensive targets and enemy attack targets meets the user's combat instructions. If the user's combat instructions are met, and there are no other new tasks, it returns to its own base, and the operation ends. If the user's combat instructions are not met, and the unmanned combat equipment edge computing aggregate assigns a new reconnaissance task, it jumps to step 1 to execute the new reconnaissance task; if the unmanned combat equipment edge computing aggregate assigns a new combat task, it jumps to step 2 to execute the new combat task; if the unmanned combat equipment edge computing aggregate assigns a new combat damage assessment task, it jumps to step 3 to execute the new combat damage assessment task.

Images