A method and system for constructing a detailed battlefield environment based on digital twins

Abstract

The present invention belongs to the technical field of environment construction, and provides a method and system for constructing a fine battlefield environment based on digital twins. The method includes: controlling each drone to transmit the field image data or three-dimensional terrain data of each sub-area of ​​the battlefield area according to a preset patrol strategy; constructing a digital twin sub-model corresponding to each sub-area according to the received field image data or three-dimensional terrain data, aligning the coordinates of each digital twin sub-model, and constructing a preliminary first digital twin model; evaluating the probability of destruction of the drones in each sub-area based on the first digital twin model, determining the prediction step size according to the destruction probability, and predicting an updated digital twin sub-model according to the digital twin sub-model corresponding to the sub-area; adding the updated digital twin sub-model to the first digital twin model to obtain an updated second digital twin model. The present invention realizes the rapid and refined construction of the battlefield environment.

Description

Technical Field

[0001] The present invention relates to the technical field of environment construction, and in particular to a method and system for constructing a fine battlefield environment based on digital twins. Background Art

[0002] In modern warfare, battlefield environment information plays a key role in combat command decisions, troop deployment, and the successful implementation of combat operations. Accurate, real-time, and comprehensive battlefield environment awareness can help combatants understand the battlefield situation in advance and formulate more targeted and effective combat plans, thereby taking the initiative in complex and changing war situations.

[0003] Traditional battlefield environment construction methods mainly rely on satellite remote sensing, aerial reconnaissance, and ground manual surveys. These methods have many limitations. On the one hand, although satellite remote sensing and aerial reconnaissance can obtain battlefield information over a large area, they are insufficient in resolution and real-time performance. For example, it is difficult to accurately capture some concealed and small-scale military targets or terrain details, and the data update cycle is long, which cannot reflect the dynamic changes of the battlefield environment in a timely manner. On the other hand, ground manual surveys are limited by the range of personnel activities, safety risks, and work efficiency. It is not only difficult to achieve comprehensive coverage of a vast battlefield area, but also difficult to implement in a complex and dangerous battlefield environment.

[0004] Therefore, how to achieve rapid and refined construction of the battlefield environment is a technical problem that needs to be solved urgently. Summary of the invention

[0005] In response to the above technical problems, the present invention provides a method, system, electronic device, computer storage medium and computer program product for constructing a fine battlefield environment based on digital twins.

[0006] The present invention discloses a method for constructing a fine battlefield environment based on digital twins, the method comprising the following steps: deploying a number of unmanned aerial vehicles (UAVs) to patrol in each sub-area of ​​the battlefield area, and controlling each UAV to transmit the captured on-site image data or three-dimensional terrain data according to a preset patrol strategy; wherein the patrol strategy includes a transmission strategy and a shooting strategy; constructing a digital twin sub-model corresponding to each sub-area based on the received on-site image data or the three-dimensional terrain data, aligning the coordinates of each digital twin sub-model, and constructing a preliminary first digital twin model; evaluating the destruction probability of the UAVs in each sub-area based on the first digital twin model, determining the prediction step size according to the destruction probability, and predicting an updated digital twin sub-model according to the digital twin sub-model corresponding to the sub-area; adding the updated digital twin sub-model to the first digital twin model to obtain an updated second digital twin model.

[0007] Optionally, if a target drone patrolling in a sub-area receives a dispatch instruction to go to another sub-area, the control of each drone transmits the captured on-site image data or three-dimensional terrain data according to a preset patrol strategy, including: the target drone parses the first patrol strategy from the dispatch instruction; the first patrol strategy is preset by the command center; the target drone determines a second patrol strategy corresponding to at least one intermediate sub-area based on the first patrol strategy; wherein the intermediate sub-area refers to a sub-area that the target drone needs to pass through in the process of going to another sub-area; the second patrol strategy is determined in the following manner: calculating the distance between the intermediate sub-area and another sub-area, determining a first boost coefficient based on the distance, and using the first boost coefficient to enhance and adjust the shooting strategy in the first patrol strategy, that is, adjusting its shooting frequency and shooting resolution to obtain the second patrol strategy.

[0008] Optionally, the use of the first boosting coefficient to enhance and adjust the shooting strategy in the first patrol strategy, that is, to adjust its shooting frequency and shooting resolution, includes: the target UAV also extracts the battlefield situation of the certain sub-area and the multiple intermediate sub-areas successively crossed according to the on-site image data it takes, and if the multiple battlefield situation situations indicate that the battlefield situation tends to be intense, then the battlefield situation difference is calculated according to the battlefield situation of the current intermediate sub-area and the certain sub-area, and the second boosting coefficient is calculated according to the size of the battlefield situation difference; the first boosting coefficient and the second boosting coefficient are used to enhance and adjust the shooting strategy in the first patrol strategy, that is, to adjust its shooting frequency and shooting resolution, to obtain the second patrol strategy.

[0009] Optionally, the evaluation of the destruction probability of drones in each sub-area based on the first digital twin model includes: extracting attack object data and shelter distribution data from the first digital twin model; wherein the shelters include trees and buildings; using the first deep prediction model to predict the attack object data to obtain a first destruction probability; using the second deep prediction model to predict the shelter distribution data to obtain a destruction difficulty coefficient; and using the destruction difficulty coefficient to adjust the first destruction probability to obtain a second destruction probability.

[0010] Optionally, predicting an updated digital twin sub-model based on the digital twin sub-model corresponding to the sub-area includes: identifying dynamic objects based on several historical digital twin sub-models corresponding to the sub-area, and obtaining dynamic data sets corresponding to each dynamic object and each historical digital twin sub-model; using the dynamic data sets to predict predicted dynamic data of the dynamic object, and using each predicted dynamic data to predict an updated digital twin sub-model; wherein, no prediction update is performed on non-dynamic objects in the digital twin sub-model.

[0011] The present invention also discloses a fine battlefield environment construction system based on digital twins, the system comprising a deployment module, a digital twin model construction module, and a digital twin model update module; the deployment module is used to deploy a number of drones to patrol in each sub-area of ​​the battlefield area, and control each drone to transmit the captured on-site image data or three-dimensional terrain data according to a preset patrol strategy; wherein the patrol strategy includes a transmission strategy and a shooting strategy; the digital twin model construction module is used to construct a digital twin sub-model corresponding to each sub-area based on the received on-site image data or the three-dimensional terrain data, align the coordinates of each digital twin sub-model, and construct a preliminary first digital twin model; the digital twin model update module is used to: evaluate the destruction probability of the drones in each sub-area based on the first digital twin model, determine the prediction step size according to the destruction probability, and predict an updated digital twin sub-model according to the digital twin sub-model corresponding to the sub-area; add the updated digital twin sub-model to the first digital twin model to obtain an updated second digital twin model.

[0012] Optionally, the digital twin model update module is used to: identify dynamic objects based on several historical digital twin sub-models corresponding to the sub-area, and obtain dynamic data sets corresponding to each dynamic object and each historical digital twin sub-model; use the dynamic data sets to predict the predicted dynamic data of the dynamic object, and use each predicted dynamic data to predict an updated digital twin sub-model; wherein, no prediction update is performed on non-dynamic objects in the digital twin sub-model.

[0013] The present invention also discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement any of the above methods.

[0014] The present invention also discloses a computer storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement any of the above methods.

[0015] The present invention also discloses a computer program product, which includes computer codes. When the computer codes are executed by a processor of an electronic device, any of the above methods is implemented.

[0016] Through the above method, the present invention can at least achieve the following technical effects: 1) Utilizing the data collected by the drone, the digital twin sub-models of each sub-area are quickly constructed and the coordinates are aligned, which can preliminarily complete the digital mapping of the battlefield environment in a short time, allowing the command center to grasp the overall picture of the battlefield in time and provide a basic basis for combat decision-making.

[0017] 2) By dividing the battlefield into sub-areas and arranging drones to patrol according to strategies, it is possible to flexibly obtain on-site image data or 3D terrain data based on the actual conditions of different areas. For example, image data is directly transmitted back to areas with fighting, and extracted 3D terrain data is transmitted back to areas without fighting, which improves the pertinence and efficiency of data collection and accurately reflects the current situation on the battlefield.

[0018] 3) Based on the first digital twin model, the probability of the drone being destroyed is evaluated, the prediction step length is determined, and the digital twin sub-model is virtually updated. When the drone is destroyed, the model can be updated for a short period of time to ensure the command center's grasp of the battlefield situation. At the same time, new drones can be arranged to take over the patrol in the future to achieve real model updates and ensure that the digital twin model is always close to the real-time dynamics of the battlefield. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 It is a flow chart of a method for constructing a fine battlefield environment based on digital twins disclosed in an embodiment of the present invention.

[0021] Figure 2 It is a schematic diagram of a scenario in which a target UAV is dispatched to another sub-area disclosed in an embodiment of the present invention.

[0022] Figure 3 It is a structural schematic diagram of a system for building a fine battlefield environment based on digital twins disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The following is a description of the implementation of the present application by specific specific embodiments. People familiar with the technology can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0024] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0025] In response to the above technical issues, such as Figure 1 As shown, an embodiment of the present invention discloses a method for constructing a fine battlefield environment based on digital twins, the method comprising the following steps: S101, deploying a number of drones to patrol in each sub-area of ​​the battlefield area, and controlling each drone to transmit the captured on-site image data or three-dimensional terrain data according to a preset patrol strategy; wherein the patrol strategy includes a transmission strategy and a shooting strategy.

[0026] According to the actual situation of the battlefield and the needs of combat command, the command center divides the battlefield area into multiple sub-areas, and arranges several drones in each sub-area. The drones patrol in their respective sub-areas according to the set patrol strategy. The patrol strategy includes the back-transmission strategy and the shooting strategy. The back-transmission strategy refers to what kind of data is transmitted back to the rear. The data here is the on-site image data or three-dimensional terrain data. The three-dimensional terrain data is extracted from the on-site image data taken by the drone. For example, for sub-areas without fighting, the strategy of extracting and returning three-dimensional terrain data can be adopted. For sub-areas with fighting, the strategy of directly returning on-site image data is adopted, and the command center then extracts three-dimensional terrain data from it.

[0027] The shooting strategy refers to the frequency and resolution at which the drone shoots live images of the sub-area. For example, the sub-area with war is set to shoot with a higher frequency and resolution, and the sub-area without war is set to shoot with a lower frequency and resolution.

[0028] S102, constructing a digital twin sub-model corresponding to each sub-area based on the received on-site image data or the three-dimensional terrain data, aligning the coordinates of each digital twin sub-model, and constructing a preliminary first digital twin model.

[0029] The command center receives on-site image data or three-dimensional terrain data from drones, and uses these data to build digital twin sub-models corresponding to each sub-area, and then aligns and integrates these sub-models in a unified coordinate system to form a preliminary first digital twin model, allowing the battlefield environment to be initially digitally mapped. Among them, the three-dimensional terrain data includes terrain features such as mountains, rivers, roads, bridges, etc., as well as dynamic and static data of various military targets or military objects, such as bunkers, trenches, tanks, soldiers, etc.

[0030] S103: Evaluate the destruction probability of the drones in each sub-area based on the first digital twin model, determine the prediction step size according to the destruction probability, and predict an updated digital twin sub-model based on the digital twin sub-model corresponding to the sub-area.

[0031] As the dynamic situation on the battlefield develops, drones patrolling in various sub-areas may be destroyed. In this way, the first digital twin model cannot be subsequently updated, and the command center cannot grasp the real-time battlefield situation.

[0032] In this regard, the present invention sets the command center to evaluate the battlefield situation of each sub-area, enemy firepower distribution and other factors based on the first digital twin model obtained by the aforementioned preliminary construction, and obtain the probability of destruction of the drone in each sub-area, and determine the appropriate prediction step size according to the probability of destruction. That is, according to the known historical battlefield situation in the first digital twin model (the historical battlefield situation corresponding to each earlier updated node), the change of the battlefield situation in the predicted future period is predicted, and after the drone is destroyed, the corresponding digital twin sub-model is virtually updated according to the prediction result. In this way, even if the drone is destroyed, the command center can also realize a short-term update of the digital twin model in order to grasp the battlefield situation. For example, the first digital twin model is used to use multiple update information in the first hour to predict the advancement situation of the enemy armored vehicle unit from the current to the next 10 minutes.

[0033] In addition, after a drone is destroyed, a new drone can be dispatched or transferred from other sub-areas to take over the patrol, so that the corresponding digital twin sub-model can be truly updated as soon as possible.

[0034] 104. Add the updated digital twin sub-model to the first digital twin model to obtain an updated second digital twin model.

[0035] The predicted and updated digital twin sub-model is integrated into the first digital twin model to form a second digital twin model with richer content and more in line with the real-time dynamics of the battlefield, providing more accurate battlefield environment information for combat decision-making. It should be noted that the update of the first digital twin model is continuous and multiple, and multiple second digital twin models will be obtained.

[0036] Through the above method, the present invention can at least achieve the following technical effects: 1) Utilizing the data collected by the drone, the digital twin sub-models of each sub-area are quickly constructed and the coordinates are aligned, which can preliminarily complete the digital mapping of the battlefield environment in a short time, allowing the command center to grasp the overall picture of the battlefield in time and provide a basic basis for combat decision-making.

[0037] 2) By dividing the battlefield into sub-areas and arranging drones to patrol according to strategies, it is possible to flexibly obtain on-site image data or 3D terrain data based on the actual conditions of different areas. For example, image data is directly transmitted back to areas with fighting, and extracted 3D terrain data is transmitted back to areas without fighting, which improves the pertinence and efficiency of data collection and accurately reflects the current situation on the battlefield.

[0038] 3) Based on the first digital twin model, the probability of the drone being destroyed is evaluated, the prediction step length is determined, and the digital twin sub-model is virtually updated. When the drone is destroyed, the model can be updated for a short period of time to ensure the command center's grasp of the battlefield situation. At the same time, new drones can be arranged to take over the patrol in the future to achieve real model updates and ensure that the digital twin model is always close to the real-time dynamics of the battlefield.

[0039] Optionally, if a target drone patrolling in a sub-area receives a dispatch instruction to go to another sub-area, the control of each drone transmits the captured on-site image data or three-dimensional terrain data according to a preset patrol strategy, including: the target drone parses the first patrol strategy from the dispatch instruction; the first patrol strategy is preset by the command center; the target drone determines a second patrol strategy corresponding to at least one intermediate sub-area based on the first patrol strategy; wherein the intermediate sub-area refers to a sub-area that the target drone needs to pass through in the process of going to another sub-area; the second patrol strategy is determined in the following manner: calculating the distance between the intermediate sub-area and another sub-area, determining a first boost coefficient based on the distance, and using the first boost coefficient to enhance and adjust the shooting strategy in the first patrol strategy, that is, adjusting its shooting frequency and shooting resolution to obtain the second patrol strategy.

[0040] In this embodiment, if Figure 2 As shown, when a drone fails or is shot down (see Figure 2 3), the Command Center will be from the adjacent sub-area (see Figure 2 Neutron Area 1) dispatches some target drones (see Figure 2The drone with an arrow at sub-area 2 goes to that area (i.e. the other sub-area mentioned above) to take over shooting. The command center determines the appropriate first patrol strategy in advance according to the fighting situation (intensity of the fighting) in the other sub-area. For example, when the fighting intensity in the other sub-area is high, the back-transmission strategy is set to back-transmission of on-site image data, and the shooting strategy is high shooting frequency and high resolution; and when the fighting intensity in the other sub-area is low, the back-transmission strategy is set to back-transmission of on-site image data, and the shooting strategy is low shooting frequency and low resolution.

[0041] At the same time, the dispatched target drone may need to cross at least one intermediate sub-area to reach the above-mentioned other sub-area. In this process, since there are other drones patrolling normally in the intermediate sub-area, the target drone does not need to shoot or only needs to shoot at a low level to save power. At the same time, as the target drone gradually approaches the above-mentioned other sub-area, the target drone calculates the distance between the current intermediate sub-area and the other sub-area, and decides on a suitable first boost coefficient based on the size of the distance. The first boost coefficient is used to gradually increase the shooting frequency and shooting resolution in the first patrol strategy set by the command center, that is, as the target drone gradually approaches the other sub-area, the target drone gradually increases the shooting frequency and shooting resolution to achieve the early and quick shooting of the on-site image data of the other sub-area (that is, make preparations for shooting in advance) and transmit it back, so that the command center can obtain the latest digital twin model of the other sub-area in a more timely manner.

[0042] Optionally, the use of the first boosting coefficient to enhance and adjust the shooting strategy in the first patrol strategy, that is, to adjust its shooting frequency and shooting resolution, includes: the target UAV also extracts the battlefield situation of the certain sub-area and the multiple intermediate sub-areas successively crossed according to the on-site image data it takes, and if the multiple battlefield situation situations indicate that the battlefield situation tends to be intense, then the battlefield situation difference is calculated according to the battlefield situation of the current intermediate sub-area and the certain sub-area, and the second boosting coefficient is calculated according to the size of the battlefield situation difference; the first boosting coefficient and the second boosting coefficient are used to enhance and adjust the shooting strategy in the first patrol strategy, that is, to adjust its shooting frequency and shooting resolution, to obtain the second patrol strategy.

[0043] In this embodiment, the target drone also photographs the original sub-region and multiple intermediate sub-regions it crosses in sequence while flying to another sub-region, and extracts the battlefield situation of each region from the captured on-site image data. For example, the battlefield situation can be judged by identifying information such as the intensity of military activities, weapon use, and personnel mobilization in the image, and these situation data can be comprehensively evaluated to obtain an evaluation value that can reflect the severity of the battlefield situation.

[0044] If multiple evaluation values ​​obtained at different times show that the battlefield situation is becoming intense, then the target drone may be more likely to head to another sub-area where intense fighting is taking place (i.e., the direction of intense fighting). In this case, it is necessary to further determine the second enhancement coefficient in order to further appropriately enhance the first patrol strategy obtained above.

[0045] Specifically, the target drone calculates the difference in battlefield situation between the current middle sub-area and the original sub-area. For example, if the military activity intensity of a sub-area is 20%, and the military activity intensity of the current middle sub-area is 40%, then the difference in battlefield situation is 20%. Then, the second boost coefficient is calculated based on the size of the battlefield situation difference. The larger the battlefield situation difference, the more drastic the battlefield situation change, and the larger the second boost coefficient is set. For example, when the battlefield situation difference is 10%-20%, the second boost coefficient is 1.2; when the difference is 20%-30%, the second boost coefficient is 1.5.

[0046] At this point, the first boost coefficient and the second boost coefficient are derived based on different considerations, and the first boost coefficient and the second boost coefficient are used to comprehensively improve and adjust the shooting strategy in the first patrol strategy. For example, the shooting frequency in the first patrol strategy is once every 10 minutes, the shooting resolution is high-definition, the first boost coefficient is 1.3, and the second boost coefficient is 1.2. The adjusted shooting frequency becomes once every (10÷(1.3×1.2)) ≈ 6.4 minutes, and the shooting resolution is further improved on the basis of high-definition (the quantization level is set for the resolution in advance, and the corresponding level is increased according to the coefficient).

[0047] After the above adjustments, a second patrol strategy adapted to the current situation in the middle sub-area is obtained, enabling the target UAV to dynamically adjust its shooting behavior according to the actual situation on the battlefield, and prepare to obtain and transmit on-site image data of another sub-area as early as possible, so that the command center can obtain the latest and complete digital twin model of the area more timely.

[0048] It should be noted that the image data or three-dimensional terrain data captured by the target UAV in the middle sub-area can also be used by the command center. For example, when there is a shooting blind spot in a certain area in the middle sub-area, the target UAV happens to pass through the blind spot. In this way, there will be no blind spots or the blind spots will be reduced in the digital twin model.

[0049] Optionally, the evaluation of the destruction probability of drones in each sub-area based on the first digital twin model includes: extracting attack object data and shelter distribution data from the first digital twin model; wherein the shelters include trees and buildings; using the first deep prediction model to predict the attack object data to obtain a first destruction probability; using the second deep prediction model to predict the shelter distribution data to obtain a destruction difficulty coefficient; and using the destruction difficulty coefficient to adjust the first destruction probability to obtain a second destruction probability.

[0050] In this embodiment, two types of important data related to the evaluation of the probability of the drone being destroyed are extracted from the first digital twin model that is initially constructed (in subsequent updates, this corresponds to the second digital twin model), namely, attack target data and shelter distribution data. Among them, shelters include objects such as trees and buildings that can provide certain protection or shielding for drones in a battlefield environment. The attack target data covers information related to various enemy equipment, facilities or personnel with attack capabilities, such as the deployment location, quantity, range, launch frequency, type, etc. of enemy air defense weapons; the shelter distribution data specifically describes the location, quantity, height, distribution density, and other information of trees, buildings, etc. in each sub-area.

[0051] The first deep prediction model and the second deep prediction model are pre-built, and both preferably use general large models, such as DeepSeek, GPT, etc., to avoid the high workload defect of vertical model construction. The first deep prediction model is used to predict the probability of the drone being destroyed when facing these attack targets based on the extracted attack target data. For example, the first deep prediction model can calculate the possibility of the drone being hit and destroyed by air defense weapons in a specific area based on the deployment location of the enemy's equipment with drone strike capability and the flight route of the drone, thereby deriving the first probability of destruction. The first probability of destruction reflects the risk of destruction faced by the drone due to the attack target without considering the influence of obstructions.

[0052] Then, the second deep prediction model analyzes the impact of the distribution of obstructions on the drone destruction process. For example, in a sub-area, if the buildings are dense and high, it will be more difficult for the enemy's attack equipment to accurately hit the drone when the drone flies in this area. Based on these characteristics of the obstructions, the second deep prediction model calculates the destruction difficulty coefficient to indicate the difficulty of destroying the drone under the distribution of obstructions. The larger the destruction difficulty coefficient, the more difficult it is to destroy the drone due to the existence of obstructions.

[0053] Finally, the first destruction probability is adjusted using the calculated destruction difficulty coefficient. If the destruction difficulty coefficient is large, it means that the shelter has a good protective effect on the drone. Then, after considering the influence of the shelter, the actual destruction probability of the drone will be reduced, so the first destruction probability will be reduced accordingly; conversely, if the destruction difficulty coefficient is small, it means that the protection of the shelter is limited, and the first destruction probability may be appropriately increased. In this way, the influence of the attack object and the shelter is comprehensively considered to obtain the second destruction probability that more accurately reflects the actual destruction probability of the drone in each sub-area. For example, the second destruction probability = the first destruction probability * (1-destruction difficulty coefficient).

[0054] Optionally, predicting an updated digital twin sub-model based on the digital twin sub-model corresponding to the sub-area includes: identifying dynamic objects based on several historical digital twin sub-models corresponding to the sub-area, and obtaining dynamic data sets corresponding to each dynamic object and each historical digital twin sub-model; using the dynamic data sets to predict predicted dynamic data of the dynamic object, and using each predicted dynamic data to predict an updated digital twin sub-model; wherein, no prediction update is performed on non-dynamic objects in the digital twin sub-model.

[0055] In this embodiment, during the construction of the digital twin sub-model corresponding to the sub-area, several historical digital twin sub-models will be generated over time. These models record the battlefield environment information of the sub-area at different times. By analyzing and identifying these historical digital twin sub-models, the dynamic objects therein are found. Dynamic objects refer to elements whose status, position, attributes, etc. in the battlefield environment will change over time, such as combat troops (soldiers, tanks, etc.), flying missiles, moving ships, etc. Elements that will not change significantly in the short term, such as mountains and fixed buildings, are non-dynamic objects.

[0056] For each dynamic object identified, the corresponding dynamic data is extracted from each historical digital twin model to form a dynamic data set. These dynamic data may include the object's position coordinates, movement speed, movement direction, weapon status (such as whether it is fired, remaining ammunition) and other information that changes over time. For example, for a tank, its dynamic data set may contain data such as its latitude and longitude position, driving speed, and turret pointing angle at different times.

[0057] Using the dynamic data set obtained above, prediction algorithms such as time series analysis and machine learning prediction models are used to predict the future state of the dynamic object, thereby obtaining the predicted dynamic data of the dynamic object. For example, based on the speed and direction of the vehicle in the past period of time, predict the location it may reach in the next period of time; based on the launch trajectory and speed of the missile, predict the time and place of its hitting the target, etc. Preferably, the above prediction algorithm also considers the historical movement law of the dynamic object, the current state, and various influencing factors of the battlefield environment (such as terrain, enemy interference, etc.) to predict its future dynamics as accurately as possible.

[0058] The predicted dynamic data of each dynamic object is integrated into the digital twin sub-model of the sub-area. Since the state of non-dynamic objects remains basically unchanged in the short term, no prediction update is performed in this process, and their state in the original digital twin sub-model is maintained. By integrating the predicted dynamic data into the model, the digital twin sub-model can reflect the possible state of the battlefield environment at a certain moment in the future. Even if the corresponding drone is destroyed, the command center can still obtain an updated digital twin sub-model when a new drone is dispatched to take over.

[0059] like Figure 3 As shown, an embodiment of the present invention also discloses a fine battlefield environment construction system based on digital twins, the system comprising a deployment module, a digital twin model construction module, and a digital twin model update module; the deployment module is used to deploy a number of drones to patrol in each sub-area of ​​the battlefield area, and control each drone to transmit the captured on-site image data or three-dimensional terrain data according to a preset patrol strategy; wherein the patrol strategy includes a transmission strategy and a shooting strategy; the digital twin model construction module is used to construct a digital twin sub-model corresponding to each sub-area based on the received on-site image data or the three-dimensional terrain data, align the coordinates of each digital twin sub-model, and construct a preliminary first digital twin model; the digital twin model update module is used to: evaluate the destruction probability of the drones in each sub-area based on the first digital twin model, determine the prediction step size according to the destruction probability, and predict an updated digital twin sub-model according to the digital twin sub-model corresponding to the sub-area; add the updated digital twin sub-model to the first digital twin model to obtain an updated second digital twin model.

[0060] An embodiment of the present invention further discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method described in the above embodiment.

[0061] An embodiment of the present invention further discloses a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is executed by a processor to implement the method described in the above embodiment.

[0062] The computer-readable storage medium described above may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the above. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media may include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0063] An embodiment of the present invention further discloses a computer program product, which includes computer code. When the computer code is executed by a processor of an electronic device, the method described in the above embodiment is implemented.

[0064] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0065] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a fine battlefield environment based on digital twins, characterized in that: The method comprises the following steps: deploying a number of drones to patrol in each sub-area of ​​the battlefield area, and controlling each drone to transmit the captured on-site image data or three-dimensional terrain data according to a preset patrol strategy; wherein the patrol strategy includes a transmission strategy and a shooting strategy; constructing a digital twin sub-model corresponding to each sub-area according to the received on-site image data or the three-dimensional terrain data, aligning the coordinates of each digital twin sub-model, and constructing a preliminary first digital twin model; evaluating the destruction probability of the drones in each sub-area based on the first digital twin model, determining the prediction step size according to the destruction probability, and predicting an updated digital twin sub-model according to the digital twin sub-model corresponding to the sub-area; adding the updated digital twin sub-model to the first digital twin model to obtain an updated second digital twin model; The destruction probability of the drone in each sub-area is evaluated based on the first digital twin model, including: extracting attack object data and shelter distribution data from the first digital twin model; wherein the shelters include trees and buildings; using the first deep prediction model to predict the first destruction probability of the drone when facing these attack objects based on the extracted attack object data; using the second deep prediction model to predict and process the shelter distribution data to obtain a destruction difficulty coefficient; and using the destruction difficulty coefficient to adjust the first destruction probability to obtain a second destruction probability.

2. According to claim 1, a method for constructing a fine battlefield environment based on digital twins is characterized in that: If a target UAV patrolling in a certain sub-area receives a dispatch instruction to go to another sub-area, the control of each UAV transmits the captured on-site image data or three-dimensional terrain data according to a preset patrol strategy, including: the target UAV parses the dispatch instruction to obtain a first patrol strategy; the first patrol strategy is preset by the command center; the target UAV determines a second patrol strategy corresponding to at least one intermediate sub-area according to the first patrol strategy; wherein the intermediate sub-area refers to a sub-area that the target UAV needs to pass through in the process of going to another sub-area; the second patrol strategy is determined in the following manner: calculating the distance between the intermediate sub-area and another sub-area, determining a first boost coefficient according to the distance, and using the first boost coefficient to enhance and adjust the shooting strategy in the first patrol strategy, that is, adjusting its shooting frequency and shooting resolution to obtain the second patrol strategy.

3. A method for constructing a fine battlefield environment based on digital twins according to claim 2, characterized in that: The method of using the first boosting coefficient to improve and adjust the shooting strategy in the first patrol strategy, that is, adjusting its shooting frequency and shooting resolution, includes: the target UAV also extracts the battlefield situation of the certain sub-area and multiple intermediate sub-areas successively crossed according to the on-site image data it has taken, and if the multiple battlefield situation situations indicate that the battlefield situation tends to be intense, then the battlefield situation difference is calculated according to the battlefield situation of the current intermediate sub-area and the certain sub-area, and the second boosting coefficient is calculated according to the size of the battlefield situation difference; using the first boosting coefficient and the second boosting coefficient to improve and adjust the shooting strategy in the first patrol strategy, that is, adjusting its shooting frequency and shooting resolution, to obtain the second patrol strategy.

4. The method for constructing a fine battlefield environment based on digital twins according to claim 1, characterized in that: The method of predicting an updated digital twin sub-model based on the digital twin sub-model corresponding to the sub-area includes: identifying dynamic objects based on several historical digital twin sub-models corresponding to the sub-area, and obtaining dynamic data sets corresponding to each dynamic object and each historical digital twin sub-model; using the dynamic data sets to predict predicted dynamic data of the dynamic object, and using each predicted dynamic data to predict an updated digital twin sub-model; wherein, no prediction update is performed on non-dynamic objects in the digital twin sub-model.

5. A detailed battlefield environment construction system based on digital twins, characterized by: The system includes a deployment module, a digital twin model construction module, and a digital twin model update module; The deployment module is used to deploy a number of drones to patrol in each sub-area of ​​the battlefield area, and control each drone to transmit the captured on-site image data or three-dimensional terrain data according to a preset patrol strategy; wherein the patrol strategy includes a transmission strategy and a shooting strategy; The digital twin model construction module is used to construct a digital twin sub-model corresponding to each sub-area according to the received on-site image data or the three-dimensional terrain data, align the coordinates of each digital twin sub-model, and construct a preliminary first digital twin model; The digital twin model update module is used to: evaluate the destruction probability of the drones in each sub-area based on the first digital twin model, determine the prediction step size according to the destruction probability, and predict an updated digital twin sub-model according to the digital twin sub-model corresponding to the sub-area; add the updated digital twin sub-model to the first digital twin model to obtain an updated second digital twin model; The destruction probability of the drone in each sub-area is evaluated based on the first digital twin model, including: extracting attack object data and shelter distribution data from the first digital twin model; wherein the shelters include trees and buildings; using the first deep prediction model to predict the first destruction probability of the drone when facing these attack objects based on the extracted attack object data; using the second deep prediction model to predict and process the shelter distribution data to obtain a destruction difficulty coefficient; and using the destruction difficulty coefficient to adjust the first destruction probability to obtain a second destruction probability.

6. A digital twin-based fine battlefield environment construction system according to claim 5, characterized in that: The digital twin model update module is used to: identify dynamic objects according to several historical digital twin sub-models corresponding to the sub-area, and obtain dynamic data sets corresponding to each dynamic object and each historical digital twin sub-model; use the dynamic data sets to predict the predicted dynamic data of the dynamic object, and use each predicted dynamic data to predict an updated digital twin sub-model; wherein, no prediction update is performed on non-dynamic objects in the digital twin sub-model.

7. An electronic device comprising: At least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method according to any one of claims 1 to 4.

8. A computer storage medium storing a computer program, characterized in that: The computer program is executed by a processor to implement the method according to any one of claims 1 to 4.

9. A computer program product, characterized in that: The computer program product includes computer codes, and when the computer codes are executed by a processor of an electronic device, the method according to any one of claims 1 to 4 is implemented.

Images