Command system based on digital twin electronic sand table

Abstract

The invention relates to the technical field of digital twinning, in particular to a command system based on a digital twinning electronic sand table. The combat unit is used for acquiring the body state and the real-time position of a combat and the real-time image of an enemy; the sand table unit is used for converting the DIS map into a three-dimensional map, displaying the received real-time position of the combat unit and the real-time image of the enemy in the three-dimensional map and updating the real-time position and the real-time image in real time so as to generate the moving track of the enemy; and the command unit inputs a task instruction in the three-dimensional map of the sand table unit according to the movement track of the enemy, selects a combat for task execution, and transmits the task instruction to the combat unit. According to the system, the detection precision of different cameras is considered, and the robustness of the model is improved, so that a more accurate positioning result is provided in a complex environment, efficient battlefield management and command can be realized, and the combat efficiency and the success rate are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twins, and in particular to an electronic sandbox command system based on digital twins. Background Art

[0002] Modern military operations are increasingly dependent on efficient and accurate command and control systems. Traditional command systems usually rely on paper maps, radio communications, and manual reports. These methods have obvious deficiencies in information transmission speed, accuracy, and real-time performance. With the development of information technology, the electronic sandbox command system has emerged and has become an important tool for modern military command. Although the electronic sandbox command system has made significant progress in improving the speed and accuracy of information transmission, it still faces some challenges in practical applications. Especially in complex and changing battlefield environments, the accuracy of target positioning and the robustness of the model are still key issues. These issues not only affect the accuracy of command decisions, but also limit the overall effectiveness of the system. The existing electronic sandbox command system has a large error in the position of the target (such as the enemy, combatants) displayed on the electronic sandbox, which reduces the accuracy of command decisions, and the target positions captured by different cameras are inconsistent, making it difficult to form accurate three-dimensional positioning. Therefore, a digital twin electronic sandbox command system is designed. Summary of the invention

[0003] The purpose of the present invention is to provide a command system based on a digital twin electronic sand table to solve the problem that the position of the target (such as the enemy, combatants) displayed on the electronic sand table has a large error, resulting in reduced accuracy of command decisions, and the target positions captured by different cameras are inconsistent, making it difficult to form accurate three-dimensional positioning.

[0004] To achieve the above object, the present invention provides a digital twin electronic sandbox command system, comprising: A combat unit, the combat unit is used to obtain the combatant's physical state, real-time position and real-time image of the enemy, and transmit the combatant's physical state, real-time position and real-time image of the enemy to the sandbox unit; A sandbox unit, which is used to obtain a GIS map of the target area and convert the DIS map into a three-dimensional map, and at the same time display the received real-time positions of combatants and real-time images of enemies in the three-dimensional map and update them in real time, thereby generating a movement trajectory of the enemy; The sandbox unit includes a map display module, a location determination module, a target tracking module and an analysis and alarm module; The analysis and alarm module is used to receive the physical status of the combatants transmitted by the combat unit and to give an alarm when an abnormal situation occurs; Specifically, the alarm module analyzes the distance between the enemy position and the combatant transmitted by the position determination module, and when the distance is less than the preset safety distance, an alarm is issued, and the position point of the abnormal combatant flashes in the three-dimensional map. At the same time, when the combatant's physical condition is abnormal, an alarm is issued, not only the position point of the abnormal combatant flashes in the three-dimensional map, but also an alarm sound is issued; The command unit inputs the mission instructions into the three-dimensional map of the sandbox unit according to the enemy's movement trajectory, selects the combatants to perform the mission, and transmits them to the combat unit, wherein the mission instructions include the target location, execution route, execution time and alarm signal.

[0005] As a further improvement of the present technical solution, the combat unit includes several positioning modules, and monitoring modules and terminals are provided on the several positioning modules. The positioning modules are used to obtain the real-time position of the combatants, the monitoring modules collect real-time images of the enemy through a camera, and the terminals are used to monitor the physical condition of the combatants and can receive task instructions from the command unit.

[0006] As a further improvement of the technical solution, in the sandbox unit: The map display module is used to obtain the GIS map of the target area through GIS software, and convert the GIS map into a three-dimensional map for display; The position determination module is used to identify the enemy's position through the image collected by the monitoring module, and display the enemy's position and the position of the combatants transmitted by the combat unit on the map display module, and at the same time, monitor the distance between the enemy's position and the combatants; The target tracking module is used to analyze the enemy's movement trajectory based on the received enemy position information and display the generated movement trajectory on the three-dimensional map in real time.

[0007] As a further improvement of the technical solution, the map display module converts the GIS map into a three-dimensional map for display, as follows: Get the two-dimensional data in the GIS map and the elevation data of the target area, where the elevation data is: ; in, is the final elevation value; For the The weight of the layer; For the The elevation value of the layer; is the enhancement factor; is the number of layers; ; Convert 2D coordinate data and elevation data into 3D coordinates, use modeling software to generate 3D models and render them.

[0008] As a further improvement of the technical solution, the value of the enhancement factor is specifically: when the status type of the target area is a city, then ; When the status type of the target area is forest, then ; When the target area is mountainous, ; When the target area's status type is water area, then .

[0009] As a further improvement of the technical solution, the position determination module identifies the enemy's position through the image collected by the monitoring module, as follows: S1, perform target detection on the images taken by each camera and identify the target enemy in the image; S2, converting the two-dimensional image of the enemy detected by each camera into normalized coordinates, converting the normalized coordinates into ray direction vectors, and establishing the ray equation of each camera. In the process of establishing the ray equation of each camera, camera calibration error compensation is introduced for optimization; S3. Use weighted least squares and loss function to minimize the distance between all rays to estimate the enemy's 3D coordinates.

[0010] As a further improvement of this technical solution, S2 is specifically as follows: Convert the two-dimensional image of the enemy detected by each camera into normalized coordinates: ; ; in, , For the The normalized coordinates of the enemies detected by each camera; For the The inverse matrix of the intrinsic parameter matrix of each camera; For the The 2D image coordinates of the enemy detected by each camera; For the The intrinsic parameter matrix of each camera; is the horizontal focal length of the camera; is the vertical focal length of the camera; is the horizontal coordinate of the center of the image; is the ordinate of the center of the image; Convert the normalized coordinates to a ray direction vector: ; in, For the The ray direction vector of the enemy detected by each camera; For the The transpose of the camera rotation matrix; Create the ray equation for each camera: ; in, For the The ray equation for each enemy detected by the camera; For the The world coordinate position of each camera; are the parameters on the ray.

[0011] As a further improvement of the technical solution, in S2, camera calibration error compensation is introduced to optimize the process of establishing the ray equation of each camera, and the optimization is as follows: ; ; ; in, After optimization The intrinsic parameter matrix of each camera; is the calibration error compensation item of the camera intrinsic parameter; After optimization A camera rotation matrix; For the A camera rotation matrix; is the calibration error compensation term of the rotation matrix; After optimization The world coordinate position of each camera; For the The world coordinate position of each camera; is the calibration error compensation term of the position; Convert the two-dimensional image of the enemy detected by each camera into normalized coordinates: ; Convert the normalized coordinates to a ray direction vector: ; Create the ray equation for each camera: ; in, After optimization The inverse matrix of the intrinsic parameter matrix of each camera; After optimization The transpose of the camera rotation matrix.

[0012] As a further improvement of the technical solution, S3 is specifically as follows: Multi-scale feature fusion: ; in, is the fused feature map; is the number of feature maps; For the The number of feature maps; For the feature maps; is the height of the target feature map; is the width of the target feature map; is the upsampling operation; Define the error function: ; in, is the error function; is the number of cameras; For the The weight of each camera; is the three-dimensional coordinate position of the enemy; Define the Huber loss function to optimize the error function: ; in, is the Huber loss function; is the error; is the threshold value; ; in, is the optimized error function; For the Rays of enemies detected by cameras and estimated enemy positions The error between .

[0013] As a further improvement of the present technical solution, in the analysis and alarm module, when an alarm is issued when the distance is less than a preset safety distance, the position point of the abnormal fighter will flash in the three-dimensional map; when an alarm is issued when the fighter's physical condition is abnormal, not only will the position point of the abnormal fighter flash in the three-dimensional map, but an alarm sound will also be emitted.

[0014] As a further improvement of the present technical solution, the task instruction includes a target location, an execution route, an execution time and an alarm signal.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. In the digital twin electronic sand table command system, a closed-loop information flow is formed from the combat unit to the sand table unit and then to the command unit, ensuring the timeliness and accuracy of the information. The sand table unit provides decision support for the command unit, and the decision of the command unit will affect the behavior of the combat unit, forming a continuously optimized decision support cycle. Through the collaborative operation of these three units, the entire command system can achieve efficient battlefield management and command, and improve combat efficiency and success rate; 2. In the digital twin electronic sand table command system, camera calibration error compensation is introduced for optimization in the process of establishing the ray equation of each camera, which can correct the internal parameter error and improve the accuracy of target positioning, attitude estimation and position estimation. At the same time, the error function combines the advantages of weighted least squares method and Huber loss function, which not only takes into account the detection accuracy of different cameras, but also improves the robustness of the model, thereby providing more accurate positioning results in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is the overall flow chart of the present invention; The meaning of each number in the figure is: 1. Combat unit; 2. Sand table unit; 21. Map display module; 22. Position determination module; 23. Target tracking module; 24. Analysis and alarm module; 3. Command unit. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example

[0018] See also Figure 1 As shown, a digital twin electronic sand table command system is provided. Digital twin is a technology that connects an entity or system in the physical world with its virtual model. This virtual model is a digital representation of the physical entity or system. Through the continuous updating of real-time data and historical data, it can accurately reflect the state, behavior and performance of the physical entity. The core idea of ​​digital twin is to create a "twin" corresponding to the physical world in the virtual space, so as to realize the monitoring, analysis, prediction and optimization of the physical world, including combat unit 1, sand table unit 2 and command unit 3; wherein: The combat unit 1 is used to obtain the combatant's physical condition, real-time position and real-time image of the enemy, and transmit the combatant's physical condition, real-time position and real-time image of the enemy to the sandbox unit 2; The combat unit 1 includes several positioning modules, and monitoring modules and terminals are provided on the positioning modules. The positioning modules are used to obtain the real-time position of the combatants, the monitoring modules collect real-time images of the enemy through cameras, and the terminals are used to monitor the physical state of the combatants and can receive task instructions from the command unit 3. Each combatant carries a positioning module, and a monitoring module is installed on the combatant's helmet. Real-time monitoring of the combatants' physical condition can help them understand their status in real time, detect abnormal situations in time, and improve their battlefield survival rate. By using the cameras on the terminal and helmet, the enemy's position can be accurately obtained, improving the success rate of tactical operations. Combatants can directly receive task instructions from the command unit 3 through the terminal device, reducing the time and error of information transmission.

[0019] The sandbox unit 2 is used to obtain the GIS map of the target area and convert the DIS map into a three-dimensional map. At the same time, the real-time position of the combatants and the real-time image of the enemy received are displayed in the three-dimensional map and updated in real time, thereby generating the movement trajectory of the enemy; Through the display of three-dimensional maps, commanders can understand the battlefield situation more intuitively and make quick decisions. Sandbox Unit 2 can update the location information of combatants and enemies in real time, helping commanders to keep abreast of the latest battlefield dynamics at any time.

[0020] The sandbox unit 2 includes a map display module 21, a position determination module 22, a target tracking module 23 and an analysis and alarm module 24; The map display module 21 is used to obtain the GIS map of the target area through GIS software, and convert the GIS map into a three-dimensional map for display; the three-dimensional map is displayed in real time on the electronic sand table, providing rich geographic information and visual effects to help commanders better understand and analyze the battlefield situation.

[0021] The map display module 21 converts the GIS map into a three-dimensional map for display, as follows: Get the two-dimensional data in the GIS map and the elevation data of the target area, where the elevation data is: ; in, is the final elevation value; For the The weight of the layer; For the The elevation value of the layer; is the enhancement factor; is the number of layers; ; When the status type of the target area is city, then ; When the status type of the target area is forest, then ; When the target area is mountainous, ; When the target area's status type is water area, then ; Convert 2D coordinate data and elevation data into 3D coordinates, use modeling software to generate 3D models and render them; The position determination module 22 is used to identify the enemy's position through the image collected by the monitoring module, display the enemy's position and the position of the combatant transmitted by the combat unit 1 on the map display module 21, and monitor the distance between the enemy's position and the combatant.

[0022] The position determination module 22 identifies the enemy's position through the images collected by the monitoring module, as follows: S1, perform target detection on the images taken by each camera and identify the target enemy in the image; S2, converting the two-dimensional image of the enemy detected by each camera into normalized coordinates, converting the normalized coordinates into ray direction vectors, and establishing the ray equation of each camera. In the process of establishing the ray equation of each camera, camera calibration error compensation is introduced for optimization; S3, using weighted least squares and loss function to minimize the distance between all rays to estimate the enemy's three-dimensional coordinates; S2 is as follows: Convert the two-dimensional image of the enemy detected by each camera into normalized coordinates: ; ; in, , For the The normalized coordinates of the enemies detected by each camera; For the The inverse matrix of the intrinsic parameter matrix of each camera; For the The 2D image coordinates of the enemy detected by each camera; For the The intrinsic parameter matrix of each camera; is the horizontal focal length of the camera; is the vertical focal length of the camera; is the horizontal coordinate of the center of the image; is the ordinate of the center of the image; Convert the normalized coordinates to a ray direction vector: ; in, For the The ray direction vector of the enemy detected by each camera; For the The transpose of the camera rotation matrix represents the camera posture. The rotation matrix is ​​a 3×3 orthogonal matrix. Create the ray equation for each camera: ; in, For the The ray equation of the enemy detected by the camera is expressed as a three-dimensional vector ; For the The world coordinate position of the camera, represented as a three-dimensional vector ; It is the parameter on the ray, indicating the position from the camera distance to the enemy's position; In the process of establishing the ray equation of each camera, the camera calibration error compensation is introduced for optimization, which can reduce the positioning deviation caused by the camera intrinsic calibration error. The optimization is as follows: ; ; ; in, After optimization The intrinsic parameter matrix of each camera; is the calibration error compensation item of the camera intrinsic parameter; After optimization A camera rotation matrix; For the A camera rotation matrix; is the calibration error compensation term of the rotation matrix; After optimization The world coordinate position of each camera; For the The world coordinate position of each camera; is the calibration error compensation term of the position; Internal parameter matrix Contains parameters such as the focal length of the camera and the coordinates of the principal point. If these parameters are inaccurate, it will lead to conversion errors from image coordinates to normalized coordinates. By introducing the internal reference calibration error compensation term , it can correct the internal parameter error and improve the accuracy of target positioning; Rotation Matrix Describes the posture of the camera. If the posture is inaccurate, it will cause the conversion error from the normalized coordinates to the ray direction vector. The rotation matrix calibration error compensation term is introduced , the attitude error can be corrected to improve the accuracy of attitude estimation; The world coordinate position of the camera If it is inaccurate, it will cause the starting position error of the ray equation. By introducing the position calibration error compensation term , the position error can be corrected and the accuracy of position estimation can be improved.

[0023] Convert the two-dimensional image of the enemy detected by each camera into normalized coordinates: ; Convert the normalized coordinates to a ray direction vector: ; Create the ray equation for each camera: ; in, After optimization The inverse matrix of the intrinsic parameter matrix of each camera; After optimization The transpose of the camera rotation matrix.

[0024] S3 is as follows: Multi-scale feature fusion: ; in, is the fused feature map; is the number of feature maps; For the The number of feature maps; For the feature maps; is the height of the target feature map; is the width of the target feature map; It is an upsampling operation, which is used to adjust the feature map to the same size; Through multi-scale feature fusion, targets of different sizes can be better processed and the accuracy of target detection can be improved. Feature maps of different scales can capture detail information at different levels. Weighted fusion can enhance feature representation, improve the detection performance of the model, and ensure that the two-dimensional image coordinates of the enemy detected by each camera are more accurate.

[0025] Define the error function: ; in, is the error function; is the number of cameras; For the The weight of each camera can be determined based on the camera's detection accuracy or other factors; is the enemy's three-dimensional coordinate position, expressed as a three-dimensional vector; By introducing weight ,Considering the detection accuracy of different cameras, the positioning accuracy is improved.,High-precision cameras will be given higher weights, thus occupying a larger proportion in the error function.

[0026] Define the Huber loss function to optimize the error function: ; in, is the Huber loss function, which is used to reduce the influence of outliers; is the error; is the threshold, used to distinguish the linear part from the quadratic part; The Huber loss function combines the advantages of square loss and absolute loss. It uses square loss for small errors and linear loss for large errors, thereby reducing the impact of outliers on the error function and improving the robustness of the model. ; in, For the optimized error function, the weighted least squares method and Huber loss function are combined; For the Rays of enemies detected by cameras and estimated enemy positions The error between Optimized error function It combines the advantages of weighted least squares method and Huber loss function, taking into account the detection accuracy of different cameras and improving the robustness of the model, thus providing more accurate positioning results in complex environments; The target tracking module 23 is used to analyze the enemy's movement trajectory according to the received enemy position information, and display the generated movement trajectory on the three-dimensional map in real time; The analysis and alarm module 24 is used to receive the physical condition of the combatant transmitted by the combat unit 1, and to give an alarm when an abnormal situation occurs; the analysis and alarm module 24 also receives the distance between the enemy position and the combatant transmitted by the position determination module 22, and gives an alarm when the distance is less than the preset safety distance, and flashes the position point of the abnormal combatant in the three-dimensional map; In the analysis and alarm module 24, when an alarm is issued when the distance is less than the preset safety distance, the position point of the abnormal fighter will flash in the three-dimensional map; when an alarm is issued when the fighter's physical condition is abnormal, not only the position point of the abnormal fighter will flash in the three-dimensional map, but also an alarm sound will be emitted.

[0027] In summary, the map display module 21 provides a basic map display platform for other modules, ensuring that all information can be displayed and analyzed on the same three-dimensional map; the position determination module 22 determines the position of the enemy and the position of the combatants through image acquisition and target detection, and transmits this position information to the map display module 21 and the target tracking module 23, and cooperates with the analysis and alarm module 24 to monitor the distance between the enemy and the combatants and provide distance data for safe distance alarm; the target tracking module 23 receives the enemy position information transmitted by the position determination module 22, analyzes the enemy's movement trajectory, and displays the trajectory information in real time in the map display module 21, and cooperates with the position determination module 22 and the analysis and alarm module 24 to provide the enemy's dynamic information to help commanders make decisions; the analysis and alarm module 24 receives the combatants' physical condition data transmitted by the combat unit 1, monitors abnormal situations and issues alarms, and also receives the enemy's position and the distance data between the combatants transmitted by the position determination module 22, monitors the safe distance and issues alarms, and displays the alarm information in the three-dimensional map to help commanders quickly identify and handle abnormal situations; Through the collaborative work of these four modules, Sandbox Unit 2 can efficiently process and display battlefield information, provide real-time and accurate situational awareness and decision support. Each module has its unique functions and roles, and together they constitute a powerful command and control system, enhancing the ability of battlefield management and command.

[0028] The command unit 3 inputs the task instruction in the three-dimensional map of the sandbox unit 2 according to the enemy's movement trajectory, selects the combatants to perform the task, and transmits it to the combat unit 1, wherein the task instruction includes the target location, execution route, execution time and alarm signal; Based on the rich information provided by Sand Table Unit 2, commanders can plan combat plans more scientifically and reasonably, respond quickly to battlefield changes, and adjust tactical strategies in a timely manner. By specifying specific target locations, execution routes, execution times, and alarm signals, they can ensure that combatants can complete their tasks accurately.

[0029] A closed-loop information flow is formed from combat unit 1 to sand table unit 2 and then to command unit 3, ensuring the timeliness and accuracy of information. Sand table unit 2 provides decision support for command unit 3, and the decision of command unit 3 will affect the behavior of combat unit 1, forming a continuously optimized decision support cycle. Through the collaborative operation of these three units, the entire command system can achieve efficient battlefield management and command, and improve combat efficiency and success rate.

[0030] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only preferred examples of the present invention, and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A digital twin electronic sandbox command system, characterized in that: include: A combat unit (1), the combat unit (1) being used to obtain the combatant's physical state, real-time position and real-time image of the enemy, and transmit the combatant's physical state, real-time position and real-time image of the enemy to the sandbox unit (2); A sandbox unit (2), the sandbox unit (2) being used to obtain a GIS map of the target area and convert the DIS map into a three-dimensional map, while displaying the received real-time positions of combatants and real-time images of enemies in the three-dimensional map and updating them in real time, thereby generating a movement trajectory of the enemy; The sandbox unit (2) comprises a map display module (21), a position determination module (22), a target tracking module (23) and an analysis and alarm module (24); The analysis and alarm module (24) is used to receive the physical condition of the combatants transmitted by the combat unit (1) and to generate an alarm when an abnormal condition occurs; Specifically, the analysis alarm module (24) receives the distance between the enemy position and the combatant transmitted by the position determination module (22), and when the distance is less than a preset safety distance, an alarm is issued, and the position point of the abnormal combatant flashes in the three-dimensional map; when an alarm is issued when the combatant's physical condition is abnormal, not only the position point of the abnormal combatant flashes in the three-dimensional map, but also an alarm sound is emitted; A command unit (3) inputs a mission instruction into a three-dimensional map of a sandbox unit (2) according to the enemy's movement trajectory, selects combatants to perform the mission, and transmits the mission instruction to a combat unit (1), wherein the mission instruction includes a target location, an execution route, an execution time, and an alarm signal.

2. The digital twin electronic sandbox command system according to claim 1 is characterized in that: The combat unit (1) comprises a plurality of positioning modules, on which are provided monitoring modules and terminals. The positioning modules are used to obtain the real-time position of combatants, the monitoring modules collect real-time images of the enemy through cameras, and the terminals are used to monitor the physical condition of combatants and are capable of receiving task instructions from the command unit (3).

3. The digital twin electronic sandbox command system according to claim 2 is characterized in that: In the sandbox unit (2): The map display module (21) is used to obtain a GIS map of the target area through GIS software, and convert the GIS map into a three-dimensional map for display; The position determination module (22) is used to identify the position of the enemy through the image collected by the monitoring module, and display the position of the enemy and the position of the combatant transmitted by the combat unit (1) on the map display module (21), and at the same time, monitor the distance between the enemy position and the combatant; The target tracking module (23) is used to analyze the enemy's movement trajectory based on the received enemy position information, and to display the generated movement trajectory on a three-dimensional map in real time.

4. The digital twin electronic sandbox command system according to claim 3 is characterized in that: The map display module (21) converts the GIS map into a three-dimensional map for display, as follows: Get the two-dimensional data in the GIS map and the elevation data of the target area, where the elevation data is: ； in, is the final elevation value; For the The weight of the layer; For the The elevation value of the layer; is the enhancement factor; is the number of layers; ; Convert 2D coordinate data and elevation data into 3D coordinates, use modeling software to generate 3D models and render them.

5. The digital twin electronic sandbox command system according to claim 4 is characterized in that: The value of the enhancement factor is specifically: when the status type of the target area is a city, then ; When the target area's status type is forest, ; When the target area is mountainous, ; When the target area is a water area, .

6. The digital twin electronic sandbox command system according to claim 5 is characterized in that: The position determination module (22) identifies the enemy's position through the image collected by the monitoring module, as follows: S1, perform target detection on the images taken by each camera and identify the target enemy in the image; S2, converting the two-dimensional image of the enemy detected by each camera into normalized coordinates, converting the normalized coordinates into ray direction vectors, and establishing the ray equation of each camera. In the process of establishing the ray equation of each camera, camera calibration error compensation is introduced for optimization; S3. Use weighted least squares and loss function to minimize the distance between all rays to estimate the enemy's 3D coordinates.

7. The digital twin electronic sandbox command system according to claim 6 is characterized in that: The S2 is specifically as follows: Convert the two-dimensional image of the enemy detected by each camera into normalized coordinates: ； ； in, , For the The normalized coordinates of the enemies detected by each camera; For the The inverse matrix of the intrinsic parameter matrix of each camera; For the The 2D image coordinates of the enemy detected by each camera; For the The intrinsic parameter matrix of each camera; is the horizontal focal length of the camera; is the vertical focal length of the camera; is the horizontal coordinate of the center of the image; is the ordinate of the center of the image; Convert the normalized coordinates to a ray direction vector: ； in, For the The ray direction vector of the enemy detected by each camera; For the The transpose of the camera rotation matrix; Create the ray equation for each camera: ； in, For the The ray equation for each enemy detected by the camera; For the The world coordinate position of each camera; are the parameters on the ray.

8. The digital twin electronic sandbox command system according to claim 7 is characterized in that: In S2, camera calibration error compensation is introduced to optimize the process of establishing the ray equation of each camera. The optimization is as follows: ； ； ； in, After optimization The intrinsic parameter matrix of each camera; is the calibration error compensation item of the camera intrinsic parameter; After optimization A camera rotation matrix; For the A camera rotation matrix; is the calibration error compensation term of the rotation matrix; After optimization The world coordinate position of each camera; For the The world coordinate position of each camera; is the calibration error compensation term of the position; Convert the two-dimensional image of the enemy detected by each camera into normalized coordinates: ； Convert the normalized coordinates to a ray direction vector: ； Create the ray equation for each camera: ； in, After optimization The inverse matrix of the intrinsic parameter matrix of each camera; After optimization The transpose of the camera rotation matrix.

9. The digital twin electronic sandbox command system according to claim 8 is characterized in that: The S3 is as follows: Multi-scale feature fusion: ； in, is the fused feature map; is the number of feature maps; For the The number of feature maps; For the feature maps; is the height of the target feature map; is the width of the target feature map; is the upsampling operation; Define the error function: ； in, is the error function; is the number of cameras; For the The weight of each camera; is the three-dimensional coordinate position of the enemy.

10. The digital twin electronic sandbox command system according to claim 9 is characterized in that: The Huber loss function is defined in the error function for optimization, and the optimization is as follows: ； in, is the Huber loss function; is the error; is the threshold value; ； in, is the optimized error function; For the Rays of enemies detected by cameras and estimated enemy positions The error between .

Images