Tactical simulation confrontation training system based on digital twinning and VR

Abstract

The invention discloses a tactical simulation confrontation training system based on digital twinning and VR. The tactical simulation confrontation training system comprises a digital twinning construction and synchronization module, a virtual reality interaction and presentation module, a tactical confrontation simulation and deduction module, a training process monitoring and data acquisition module and a training effect evaluation and intelligent redisk module. According to the invention, the digital twinborn body synchronously mapped with the real combat unit and the environment is constructed, and is deeply integrated with the immersive VR interaction environment, so that a highly vivid and panoramic immersive tactical training scene is provided for trainees, the system supports red and blue parties or multiple parties to carry out intelligent confrontation in a virtual-real combined dynamic environment, and the training efficiency is improved. The multi-dimensional training data is acquired in real time, and based on the data, the system can perform refined and quantitative objective evaluation and intelligent redisk on the tactical actions, decision-making processes and collaborative effects of individuals and teams, so that the authenticity of training, the intelligence of confrontation and the scientificity of evaluation are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of computer system technology, and more specifically to a tactical simulation combat training system based on digital twins and VR. Background Technology

[0002] Traditional tactical training usually relies on physical venues, equipment, and personnel assembly, which has problems such as high cost, complex organization, significant safety risks, limited training scenarios, and difficulty in reproducing complex and ever-changing battlefield environments.

[0003] In recent years, computer-based simulation training systems have alleviated the above problems to some extent, but most systems still have shortcomings such as insufficient simulation realism, weak immersion and presence for trainees, insufficient objectivity and comprehensiveness in the evaluation of training process and results, limited intelligence levels of both adversaries and unnatural interaction.

[0004] Digital twin technology can map physical entities and dynamic processes with high fidelity, while VR technology can provide an immersive experience. Deeply integrating the two and applying them to tactical training is expected to break through existing technological bottlenecks and achieve a more efficient, flexible, and safer modern training model. Summary of the Invention

[0005] To address these issues, this invention provides a tactical simulation combat training system based on digital twins and VR, which solves the problems of insufficient simulation realism, weak immersion and presence for trainees, insufficient objectivity and comprehensiveness in the evaluation of training process and results, and limited intelligence levels and unnatural interaction between opposing sides in the existing technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A tactical simulation combat training system based on digital twins and VR includes:

[0008] The digital twin construction and synchronization module is used to construct and drive a corresponding dynamic digital twin model in virtual space based on preset combat unit and battlefield environment data in the real world, and to ensure that the state of the digital twin model is synchronized and mapped with the state of the entity in the real world or the hypothetical scenario.

[0009] The virtual reality interaction and presentation module is used to receive and process the scene data generated by the dynamic digital twin model, generate a corresponding immersive virtual reality training environment, and present the immersive virtual reality training environment to the trainee through the VR display and interaction device, while collecting the control commands and interactive action data input by the trainee through the VR display and interaction device.

[0010] The tactical confrontation simulation and deduction module is used to simulate the confrontation behavior of the red team, blue team and / or third-party intelligent agents in the immersive virtual reality training environment based on preset tactical rules, physical engine and artificial intelligence algorithm. The confrontation behavior includes maneuver, reconnaissance, attack, defense, communication and interference. It also responds in real time to the control commands and interactive action data input by the trainee, drives the dynamic digital twin model and intelligent agent behavior to perform real-time tactical deduction and interaction, and generates continuous confrontation training process data.

[0011] The training process monitoring and data acquisition module is used to monitor and acquire in real time the state change data of the dynamic digital twin model, the behavioral decision data of the agent, the physiological and operational response data of the trainee, and the event and interaction data in the immersive virtual reality training environment during the adversarial training process, forming a structured multi-source training dataset.

[0012] The training effect evaluation and intelligent debriefing module is used to receive the multi-source training dataset and, based on the preset evaluation index system and evaluation model, to automatically perform quantitative and qualitative analysis on the performance of individual trainees and teams in terms of tactical understanding, decision-making efficiency, action standardization, teamwork, and task achievement. It generates comprehensive evaluation reports and visual analysis charts and supports replay and debriefing based on key event points and time slices.

[0013] Preferably, the digital twin construction and synchronization module specifically includes:

[0014] The entity modeling unit is used to establish a high-fidelity three-dimensional digital model based on the geometric, physical, functional, and behavioral attributes of the preset combat unit.

[0015] The environment modeling unit is used to construct a virtual battlefield environment model that includes terrain, landforms, weather, and electromagnetic environment based on real geographic information data or hypothetical parameters.

[0016] The data synchronization and driving unit is used to acquire external data source information in real time or near real time through sensor data interface or scenario data input interface, and update and drive the state and behavior of the three-dimensional digital model and the virtual battlefield environment model according to the external data source information, so as to maintain the mapping consistency between the digital twin and the real or scenario entity.

[0017] Preferably, the virtual reality interaction and presentation module specifically includes:

[0018] The scene rendering engine is used to perform real-time 3D graphics rendering and special effects processing on the dynamic digital twin model and the virtual battlefield environment model to generate the visual images of the immersive virtual reality training environment.

[0019] The stereo sound field simulation unit is used to generate corresponding three-dimensional spatial audio based on the spatial relationships and events in the virtual scene;

[0020] The interactive perception and feedback unit is used to capture the trainee's head posture, limb movements and operational intentions through positioning and tracking devices, motion capture devices, force feedback devices and data gloves, and convert them into interactive commands in the virtual environment, while providing tactile and force sensory feedback.

[0021] Preferably, the tactical confrontation simulation and deduction module includes an agent behavior engine, which includes:

[0022] The rule base stores a set of rules based on tactical doctrines and operational procedures;

[0023] The decision model library contains a variety of AI decision models for simulating the autonomous behavior of enemy, friendly, or neutral units.

[0024] The intelligent agent behavior engine loads the corresponding AI decision-making model for the non-player controlled entity in the virtual environment according to the training scenario, and drives the non-player controlled entity to perform autonomous perception, decision-making and action based on the rule base and real-time situation, so as to realize intelligent confrontation or cooperation with the digital twin controlled by the trainee.

[0025] Preferably, the physiological and operational response data of the trainee collected by the training process monitoring and data acquisition module includes at least: eye movement trajectory data, heart rate variability data, skin conductance response data, input timing and accuracy data of the control device, and gaze point and attention distribution data in the VR environment.

[0026] Preferably, the training effect evaluation and intelligent review module specifically includes:

[0027] The indicator calculation unit is used to calculate the values ​​of various underlying indicators in the evaluation indicator system based on the multi-source training dataset. The evaluation indicator system covers task performance indicators, operational skill indicators, decision quality indicators, teamwork indicators, and physiological load indicators.

[0028] The comprehensive analysis model is used to integrate and analyze the underlying indicator values ​​using weighted fusion, pattern recognition or machine learning methods, and output a multi-dimensional comprehensive evaluation result of tactical literacy, psychological quality and team effectiveness.

[0029] The debriefing guidance system is used to automatically identify key decision points, typical errors, and outstanding performance segments in the training process based on the comprehensive evaluation report, provide comparative analysis views and text comments, and support coaches or trainees to select specific perspectives and time ranges for scenario reenactment and process review.

[0030] Preferably, the training effect evaluation and intelligent review module includes a training scheme adaptive optimization unit. The training scheme adaptive optimization unit is used to analyze the weaknesses and shortcomings of the trained individuals or teams based on the comprehensive evaluation results generated by historical training batches, and automatically adjust the difficulty parameters of subsequent training scenarios, the adversarial intensity of the enemy intelligent agent, or the complexity of the training environment accordingly.

[0031] Preferably, it also includes a distributed network support module, which is used to support multiple virtual reality interaction and presentation modules to interconnect through a network, so that trainees located in different physical locations can access the same immersive virtual reality training environment and control their respective digital twins or intelligent agents to conduct collaborative tactical training or remote confrontation exercises.

[0032] Preferably, the distributed network support module adopts a server-based synchronization architecture or a peer-to-peer network synchronization architecture, and includes data compression and differential synchronization mechanisms.

[0033] This invention has the following advantages: By constructing a digital twin that is synchronously mapped with real combat units and environments, and deeply integrating it with an immersive VR interactive environment, this invention provides trainees with a highly realistic, panoramic, and immersive tactical training scenario. The system supports intelligent confrontation between red and blue teams or multiple parties in a dynamic environment combining virtual and real elements, and collects multi-dimensional training data in real time. Based on this data, the system can conduct refined and quantitative objective evaluation and intelligent review of individual and team tactical actions, decision-making processes, and collaborative effects, significantly improving the realism of training, the intelligence of confrontation, and the scientific nature of evaluation. This not only greatly reduces the dependence of training on physical resources and training risks, but also supports repetitive and customizable training in complex battlefield environments, thereby effectively improving command and decision-making capabilities, tactical execution capabilities, and team collaborative combat capabilities. Attached Figure Description

[0034] To more intuitively illustrate the prior art and this application, exemplary drawings are provided below. It should be understood that the specific shapes and structures shown in the drawings should not generally be regarded as limiting conditions for implementing this application; for example, based on the technical concept disclosed in this application and the exemplary drawings, those skilled in the art are able to easily make conventional adjustments or further optimizations to the addition / reduction / classification, specific shapes, positional relationships, connection methods, size ratios, etc. of certain units (components).

[0035] Figure 1 A block diagram of a tactical simulation combat training system based on digital twins and VR provided in an embodiment of this application. Detailed Implementation

[0036] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these embodiments are merely for further explanation of the present invention and should not be construed as limiting the scope of protection of the present invention. Technical engineers in the field can make some non-essential improvements and adjustments to the present invention based on the above-described content. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Please see Figure 1 A tactical simulation combat training system based on digital twins and VR includes:

[0038] The digital twin construction and synchronization module is used to construct and drive a corresponding dynamic digital twin model in virtual space based on preset combat unit and battlefield environment data in the real world, and to ensure that the state of the digital twin model is synchronized and mapped with the state of the entity in the real world or the hypothetical scenario.

[0039] The virtual reality interaction and presentation module is connected to the digital twin construction and synchronization module. It is used to receive and process the scene data generated by the dynamic digital twin model, generate a corresponding immersive virtual reality training environment, and present the immersive virtual reality training environment to the trainee through the VR display and interaction device. At the same time, it collects the control commands and interactive action data input by the trainee through the VR display and interaction device.

[0040] The tactical confrontation simulation and deduction module is connected to the digital twin construction and synchronization module and the virtual reality interaction and presentation module, respectively. It is used to simulate the confrontation behavior of the red team, blue team and / or third-party intelligent agents in the immersive virtual reality training environment based on preset tactical rules, physical engine and artificial intelligence algorithm. The confrontation behavior includes maneuvering, reconnaissance, attack, defense, communication and interference. It also responds in real time to the control commands and interactive action data input by the trainee, drives the dynamic digital twin model and intelligent agent behavior to perform real-time tactical deduction and interaction, and generates continuous confrontation training process data.

[0041] The training process monitoring and data acquisition module is used to monitor and acquire in real time the state change data of the dynamic digital twin model, the behavioral decision data of the agent, the physiological and operational response data of the trainee, and the event and interaction data in the immersive virtual reality training environment during the adversarial training process, forming a structured multi-source training dataset.

[0042] The training effect evaluation and intelligent review module is connected to the training process monitoring and data acquisition module. It is used to receive the multi-source training dataset and, based on the preset evaluation index system and evaluation model, to automatically perform quantitative and qualitative analysis on the performance of individual trainees and teams in terms of tactical understanding, decision-making efficiency, action standardization, teamwork, and task achievement. It generates comprehensive evaluation reports and visual analysis charts and supports replay and review based on key event points and time slices.

[0043] In implementing this invention, the digital twin construction and synchronization module first constructs a dynamic, high-fidelity digital twin model in virtual space based on high-precision data of real combat units (such as individual soldiers, vehicles, and weapon platforms) and the battlefield environment. This model not only includes external geometry but also integrates physical characteristics and functional logic, and maintains synchronous mapping with its entity or hypothetical state through continuous data inflow (from real sensors or scenario data). This process lays a highly realistic virtual foundation for the entire training system, closely related to the physical world or tactical scenarios, enabling the training environment to accurately reflect complex and ever-changing real battlefield conditions. This ensures the high degree of simulation and practical relevance of the training, providing a reliable data source and interactive object for all subsequent training stages.

[0044] Subsequently, the virtual reality interaction and presentation module transforms the aforementioned dynamic digital twin model into an immersive environment that trainees can directly perceive through their senses. It utilizes a high-performance graphics rendering engine to generate a three-dimensional visual scene, combined with spatial audio and force feedback devices, creating a strong sense of presence for the trainees. Trainees, through VR devices, are placed in this environment, and their head, hand, and eye movements are captured in real time and translated into control commands for the digital twin. This deep immersion and natural interaction greatly enhances trainees' participation and concentration, making their physiological and psychological state closer to real combat scenarios, creating prerequisites for effective training.

[0045] In the core training phase, the tactical confrontation simulation and deduction module begins operation. Based on built-in tactical rules and artificial intelligence algorithms, it drives the blue team or a third-party intelligent agent to conduct autonomous and intelligent confrontational behaviors (such as maneuvering and flanking maneuvers, and fire suppression) within the environment. Simultaneously, this module processes the trainee's (red team's) control commands in real time, driving the digital twin controlled by the trainee to dynamically interact and deduce from the intelligent agent. This process creates a highly flexible and challenging intelligent confrontation environment, forcing trainees to constantly observe, judge, and make decisions as if on a real battlefield, effectively honing their tactical thinking and rapid response capabilities.

[0046] The training process monitoring and data acquisition module records multi-dimensional data streams seamlessly and objectively throughout the entire process. It not only records macro-level situational data such as state changes of entities (digital twins and intelligent agents) and combat events in the virtual environment, but also simultaneously collects detailed operational data of individual trainees (such as aiming trajectories and firing timing) as well as physiological response data such as eye movements and heart rate. This transforms the originally subjective and vague training process into quantifiable and traceable structured data, providing a solid data foundation for subsequent objective and accurate assessments. This allows the analysis results to more realistically and comprehensively reflect the trainees' overall abilities and status.

[0047] After training, the training effectiveness evaluation and intelligent debriefing module conducts in-depth analysis based on the massive multi-source training dataset. Through a pre-set evaluation model, it not only calculates traditional performance indicators such as task completion time and hit rate, but also analyzes the rationality of the decision-making process, the efficiency of team communication and collaboration, and even the level of physiological stress. The system then automatically generates a visual evaluation report, intelligently locating key decision points and typical mistakes, and supporting multi-angle replay and debriefing. This process elevates the training effect from a subjective perception to a scientifically quantifiable level. The personalized, data-driven feedback and debriefing provided accurately pinpoint trainees' weaknesses, guiding them to make targeted improvements, greatly enhancing the scientific rigor, closed-loop nature, and iterative optimization efficiency of the training.

[0048] The digital twin construction and synchronization module specifically includes:

[0049] The entity modeling unit is used to establish a high-fidelity three-dimensional digital model based on the geometric, physical, functional, and behavioral attributes of the preset combat unit.

[0050] The environment modeling unit is used to construct a virtual battlefield environment model that includes terrain, landforms, weather, and electromagnetic environment based on real geographic information data or hypothetical parameters.

[0051] The data synchronization and driving unit is used to acquire external data source information in real time or near real time through sensor data interface or scenario data input interface, and update and drive the state and behavior of the three-dimensional digital model and the virtual battlefield environment model based on the information, so as to maintain the mapping consistency between the digital twin and the real or scenario entity.

[0052] The solid modeling unit is responsible for creating 3D models with detailed geometric appearance and inherent physical properties (such as mass, inertia, and material properties). The environment modeling unit utilizes GIS data or manual editing to construct a virtual battlefield that includes complex terrain undulations, vegetation, buildings, and dynamic atmospheric effects. The data synchronization and actuation unit is crucial to this solution. For example, when simulating a tank, the data synchronization and actuation unit can receive and process data such as the tank's position, speed, and ammunition status set in the scenario, driving the virtual tank model to make corresponding movements and state changes in real time. This layered and refined modeling and synchronization mechanism ensures high credibility in the behavior and performance of the digital twin.

[0053] The virtual reality interaction and presentation module specifically includes:

[0054] The scene rendering engine is used to perform real-time 3D graphics rendering and special effects processing on the dynamic digital twin model and the virtual battlefield environment model to generate the visual images of the immersive virtual reality training environment.

[0055] The stereo sound field simulation unit is used to generate corresponding three-dimensional spatial audio based on the spatial relationships and events in the virtual scene;

[0056] The interactive perception and feedback unit is used to capture the trainee's head posture, limb movements and operational intentions through positioning and tracking devices, motion capture devices, force feedback devices and data gloves, and convert them into interactive commands in the virtual environment, while providing tactile and force sensory feedback.

[0057] The scene rendering engine is responsible for transforming digital twin models and battlefield environment models into high-frame-rate, realistic visual images. The stereo sound field simulation unit, based on a sound propagation model, simulates spatial audio such as bullet whistles and explosion locations to enhance immersion. The interactive perception and feedback unit captures trainee movements through hardware (such as VR controllers and force feedback vests) and can convert impacts received in the virtual environment (such as being hit) into tactile vibration feedback. For example, when a trainee performs a concealment maneuver, the system can accurately capture their crouching posture and synchronously display it in the virtual world; the force feedback gloves can also simulate the rough feel of touching the surface of cover.

[0058] The tactical confrontation simulation and deduction module integrates an agent behavior engine, which includes:

[0059] The rule base stores a set of rules based on tactical doctrines and operational procedures;

[0060] The decision model library contains a variety of AI decision models for simulating the autonomous behavior of enemy, friendly, or neutral units.

[0061] The intelligent agent behavior engine loads the corresponding AI decision-making model for the non-player controlled entity in the virtual environment according to the training scenario, and drives the non-player controlled entity to perform autonomous perception, decision-making and action based on the rule base and real-time situation, so as to realize intelligent confrontation or cooperation with the digital twin controlled by the trainee.

[0062] The rule base encodes basic combat rules and physical constraints. The decision model library contains AI models of varying complexity, based on finite state machines, behavior trees, and even deep reinforcement learning, to simulate opponent behavior at different training levels (e.g., novice, expert). For example, in an urban warfare scenario, an enemy agent can be loaded with an "ambush" behavior model, enabling it to autonomously choose advantageous ambush locations, determine the timing of firing, and conduct simple tactical coordination with allied agents, thus providing trainees with a highly challenging and varied combat experience.

[0063] The physiological and operational response data of the trainees collected by the training process monitoring and data acquisition module include at least: eye movement trajectory data, heart rate variability data, skin conductance response data, input timing and accuracy data of the control device, and gaze point and attention distribution data in the VR environment.

[0064] Eye-tracking data can analyze trainees' attention allocation and observation habits in complex scenarios (whether they have overlooked threats in certain corners); heart rate variability data can indirectly assess their psychological stress and fatigue levels; and manipulation input timing data can accurately measure their decision-making and operational speed from target detection to firing reaction. Collecting this detailed, individualized data allows assessments to transcend traditional outcome evaluations, delving into cognitive processes and stress responses, providing objective evidence for assessing psychological qualities and operational proficiency.

[0065] The training performance evaluation and intelligent review module specifically includes:

[0066] The indicator calculation unit is used to calculate the values ​​of various underlying indicators in the evaluation indicator system based on the multi-source training dataset. The evaluation indicator system covers task performance indicators, operational skill indicators, decision quality indicators, teamwork indicators, and physiological load indicators.

[0067] The comprehensive analysis model is used to integrate and analyze the underlying indicator values ​​using weighted fusion, pattern recognition or machine learning methods, and output a multi-dimensional comprehensive evaluation result of tactical literacy, psychological quality and team effectiveness.

[0068] The debriefing guidance system is used to automatically identify key decision points, typical errors, and outstanding performance segments in the training process based on the comprehensive evaluation report, provide comparative analysis views and text comments, and support coaches or trainees to select specific perspectives and time ranges for scenario reenactment and process review.

[0069] The indicator calculation unit extracts specific indicators from the raw data, such as "average decision-making time," "team command response delay," and "effective reconnaissance information sharing rate." The comprehensive analysis model uses multi-indicator weighted scoring or cluster analysis to provide comprehensive comments such as "excellent tactical awareness but insufficient decision-making stability under pressure." The debriefing guidance system can automatically break down a failed assault operation into multiple key event points, such as "disorganized formation during engagement" and "inappropriate assault route selection," and supports comparative playback from the commander's overhead view or the team members' first-person perspective, greatly improving the relevance and efficiency of the debriefing guidance.

[0070] The training effect evaluation and intelligent review module is also connected to a training scheme adaptive optimization unit. This training scheme adaptive optimization unit is used to analyze the weaknesses and shortcomings of the trained individuals or teams based on the comprehensive evaluation results generated by historical training batches, and automatically adjust the difficulty parameters of subsequent training scenarios, the adversarial intensity of the enemy intelligent agent, or the complexity of the training environment to generate personalized progressive training schemes.

[0071] It also includes a distributed network support module, which supports multiple virtual reality interaction and presentation modules to interconnect through a network, enabling trainees located in different physical locations to access the same immersive virtual reality training environment and control their respective digital twins or intelligent agents to conduct collaborative tactical training or remote confrontation exercises, while ensuring the state synchronization and real-time interaction between each node.

[0072] The distributed network support module allows multiple geographically dispersed VR training terminals (such as commanders, pilots, and armored soldiers located at different bases) to access the same unified virtual battlefield environment. For example, a joint exercise can be organized in different locations, where one side plays the role of an attacking force conducting a ground assault at location A, while the other side plays the role of an air force providing air support at location B. The two sides interact and cooperate in real time within a shared virtual space.

[0073] The distributed network support module adopts a synchronization architecture based on an authoritative server or a peer-to-peer network synchronization architecture, and includes data compression and differential synchronization mechanisms to optimize network bandwidth usage and reduce interaction latency. At the same time, the module also provides support for training scenario distribution, user session management, training resource scheduling, and training process recording and playback.

[0074] Employing an authoritative server architecture ensures consistency of state across all clients, preventing disputes caused by network latency and asynchrony. Data compression and differential synchronization mechanisms (such as transmitting only changed entity states rather than full-scene data) effectively reduce network bandwidth requirements, making large-scale, multi-user scenario training possible under normal network conditions.

[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A tactical simulation counter-training system based on digital twin and VR, characterized in that, The method comprises the following steps: a digital twin construction and synchronization module is used to construct and drive a dynamic digital twin model corresponding to a preset combat unit and battlefield environment data in the real world in a virtual space, and ensure that the state of the digital twin model is data-synchronized and mapped with the entity state in the real world or a scenario; a virtual reality interaction and presentation module is used to receive and process scenario data generated by the dynamic digital twin model, generate a corresponding immersive virtual reality training environment, and present the immersive virtual reality training environment to a trainee through a VR display and interaction device, while collecting control instructions and interaction action data input by the trainee through the VR display and interaction device; a tactical confrontation simulation and deduction module is used to simulate confrontation behaviors of red, blue and / or third parties in the immersive virtual reality training environment based on preset tactical rules, a physics engine and artificial intelligence algorithms, the confrontation behaviors including maneuvering, reconnaissance, attack, defense, communication and interference, and the confrontation behaviors are driven to respond to the control instructions and interaction action data input by the trainee in real time, so that real-time tactical deduction and interaction of the dynamic digital twin model and the behaviors of the intelligent agents are generated, and continuous confrontation training process data is generated; a training process monitoring and data collection module is used to monitor and collect state change data of the dynamic digital twin model, behavior decision data of the intelligent agents, physiological and operation response data of the trainee, and event and interaction data in the immersive virtual reality training environment in real time during the confrontation training process, and form a structured multi-source training data set; a training effect evaluation and intelligent review module is used to receive the multi-source training data set, and automatically quantitatively and qualitatively analyze the performance of individual trainees and teams in terms of tactical understanding, decision-making efficiency, action specification, coordination, and task completion degree based on a preset evaluation index system and evaluation model, generate a comprehensive evaluation report and visual analysis chart, and support replay and review based on key event points and time slices.

2. The digital twin and VR based tactical simulation adversarial training system of claim 1, wherein, The digital twin construction and synchronization module specifically comprises: an entity modeling unit is used to establish a high-fidelity three-dimensional digital model according to geometric, physical, functional and behavior attributes of the preset combat unit; an environment modeling unit is used to construct a virtual battlefield environment model including terrain, topography, weather, and electromagnetic environment according to real geographic information data or scenario parameters; a data synchronization and driving unit is used to acquire external data source information in real time or quasi-real time through a sensor data interface or a scenario data input interface, and update and drive the state and behavior of the three-dimensional digital model and the virtual battlefield environment model according to the external data source information, so as to maintain the mapping consistency between the digital twin and the real or scenario entity.

3. The digital twin and VR based tactical simulation adversarial training system of claim 1, wherein, The virtual reality interaction and presentation module specifically comprises: a scene rendering engine is used to perform real-time three-dimensional graphic rendering and special effect processing on the dynamic digital twin model and the virtual battlefield environment model, and generate visual pictures of the immersive virtual reality training environment; A stereo field simulation unit is configured to generate corresponding three-dimensional spatial audio according to spatial relationships and events in a virtual scene; An interactive perception and feedback unit is configured to capture the head posture, limb movement and operation intention of the trainee through a positioning tracking device, a motion capture device, a force feedback device and a data glove, and convert them into interactive instructions in a virtual environment, while providing sensory feedback of touch and force.

4. The digital twin and VR based tactical simulation counter-play training system of claim 1, wherein, The tactical confrontation simulation and deduction module includes an agent behavior engine, which includes: A rule base storing a rule set based on tactical orders and combat regulations; A decision model base containing a plurality of AI decision models for simulating autonomous behaviors of enemy, friendly or neutral units; The agent behavior engine loads corresponding AI decision models for non-player-controlled entities in the virtual environment according to the training scenario, and drives the non-player-controlled entities to perform autonomous perception, decision-making and action according to the rule base and real-time situation, realizing intelligent confrontation or cooperation between the non-player-controlled entities and the digital twin controlled by the trainee.

5. The digital twin and VR based tactical simulation adversarial training system of claim 1, wherein, The physiological and operation response data of the trainee collected by the training process monitoring and data acquisition module include at least eye movement trajectory data, heart rate variability data, skin electric response data, input timing and accuracy data of the control device, and gaze point and attention distribution data in the VR environment.

6. The digital twin and VR based tactical simulation counter-play training system of claim 1, wherein, The training effect evaluation and intelligent review module specifically includes: An index calculation unit configured to calculate the values of each underlying index in the evaluation index system according to the multi-source training data set, the evaluation index system covering task performance indicators, operation skill indicators, decision quality indicators, team cooperation indicators and physiological load indicators; A comprehensive analysis model configured to integrate and analyze the values of each underlying index using a weighted fusion, pattern recognition or machine learning method, and output a multi-dimensional comprehensive evaluation result of tactical literacy, psychological quality and team effectiveness; A review guidance system configured to automatically identify key decision points, typical errors and excellent performance segments in the training process based on the comprehensive evaluation report, provide comparative analysis views and written comments, and support the trainer or trainee to select a specific perspective and time range for scene replay and process review.

7. The digital twin and VR based tactical simulation adversarial training system of claim 6, wherein, The training effect evaluation and intelligent review module includes a training scheme adaptive optimization unit configured to analyze the weaknesses and capability gaps of the trainee or team based on the comprehensive evaluation results generated by historical training batches, and automatically adjust the difficulty parameters of the subsequent training scenario, the confrontation intensity of the enemy agent, or the complexity of the training environment.

8. The digital twin and VR based tactical simulation adversarial training system of claim 7, wherein, A distributed network support module is also included, which is configured to support multiple virtual reality interaction and presentation modules to be interconnected through a network, so that trainees located in different physical locations can access the same immersive virtual reality training environment and control their own digital twins or agents respectively for collaborative tactical training or remote confrontation exercises.

9. The digital twin and VR based tactical simulation adversarial training system of claim 8, wherein, The distributed network support module adopts a synchronization architecture based on an authoritative server or a peer-to-peer network, and includes a data compression and differential synchronization mechanism.

Images