A cross-screen human-computer interaction training system for tactical confrontation

Abstract

The present application relates to the field of virtual interaction technology, and proposes a cross-screen human-computer interaction training system for tactical confrontation, which comprises: obtaining an overall correlation index according to the distance relationship between the head position and the outer contour of the building; obtaining a cooperation degree according to the time sequence change of the overall correlation index; obtaining a gaze degree according to the positional relationship of the head position, the hand position and the eye gaze vector; obtaining a tactical command degree in combination with the cooperation degree; further obtaining a tactical deployment probability; obtaining a firing probability according to the overall correlation index and the cooperation degree; judging the training stage of each human body unit according to the tactical deployment probability and the firing probability, and then processing each data stream of each person with corresponding priority. The present application reduces the key information processing delay by processing each data stream with corresponding priority, and improves the simulation credibility of the training system in the dynamic confrontation scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of virtual interaction technology, in particular to a cross-screen human-computer interaction training system for tactical confrontation. BACKGROUND

[0002] In the military field, traditional live training is limited by high loss, high intensity, high risk and insufficient complexity of the battlefield, and it is difficult to achieve high-fidelity tactical simulation, resulting in a gap between training and actual combat needs. The cross-screen human-computer interaction training system based on virtual reality technology breaks through the limitations of physical space and equipment loss by building a full-process closed-loop exercise system, restores the complex battlefield environment with high precision simulation, and realizes all-round immersive training of tactical decision-making, equipment operation and coordinated combat. The system provides low-risk, high-iteration live training scenarios for training through real-time data collection and intelligent evaluation feedback, significantly improves tactical proficiency and battlefield adaptability, and provides data support for equipment effectiveness verification and tactical innovation under the demand of information-based and intelligent warfare. It is a key technical path to break through the bottleneck of traditional training and accelerate the transformation of combat effectiveness generation mode.

[0003] The multi-dimensional dynamic data generated in the training process, such as sensor signals and tactical instructions, have the characteristics of high concurrency, which leads to the inefficiency of existing processing architecture in data stream parallel analysis and other aspects. Especially under the interference of complex battlefield noise, traditional fusion algorithms are difficult to realize adaptive collaborative processing of multi-source data streams due to rigid allocation of computing resources and static feature weights, causing delay in key information extraction and accumulation of synchronization errors, and reducing the simulation credibility of the training system in dynamic confrontation scenarios. SUMMARY

[0004] The present application provides a cross-screen human-computer interaction training system for tactical confrontation to solve the problem of low simulation credibility of existing training systems in dynamic confrontation scenarios. The technical solution adopted is as follows:

[0005] The present application provides a cross-screen human-computer interaction training system for tactical confrontation, which comprises:

[0006] A tactical confrontation training data acquisition module is used to acquire the head position, hand position, eye gaze vector of each human unit and several data streams in the training process, and to acquire the outer contour of several buildings.

[0007] A human body overall correlation index acquisition module is used to obtain the overall correlation index between human units according to the distance relationship between the head position of the human unit and the outer contour of the building.

[0008] The tactical deployment probability analysis module is used to obtain the degree of coordination between human units based on the temporal changes of the overall correlation index between human units; to obtain the degree of gaze between human units based on the positional relationship between the head position, hand position, and eye gaze vector of the human units; and to obtain the tactical command degree of each human unit by combining the degree of coordination between the human units, thereby obtaining the tactical deployment probability of each human unit.

[0009] The crossfire probability analysis module is used to obtain the crossfire probability of each human unit based on the overall correlation index and coordination degree between human units;

[0010] The data stream processing module is used to determine the training stage of each human unit based on the tactical deployment probability and engagement probability of the human unit, and to process each data stream of each human unit with corresponding priority according to the training stage of each human unit.

[0011] Furthermore, the method for obtaining the overall correlation index between human body units based on the distance relationship between the head position of the human body unit and the outer contour of the building includes:

[0012] For any human body unit, obtain the direct contact contour of the human body unit by forming a straight line connecting several building outline points to the head position of the human body unit, and the straight line does not pass through the building outline.

[0013] For any two human body units, if the head positions of the two human body units are connected by a straight line and the straight line does not pass through a building, the two human body units are recorded as a pair of direct human body units.

[0014] If the two human body units have the same direct connection contour, the two human body units are recorded as a pair of related human body units, and the same direct connection contour is recorded as the related contour of the pair of related human body units.

[0015] If there is a connecting straight line between several building outer contour points on the two directly connected contours of the two human body units, and the straight line does not pass through the buildings, the two human body units are recorded as a pair of connected human body units; for any one of the human body units in the pair of connected human body units, the directly connected contour with the smallest distance from the head position of the human body unit among the two directly connected contours is recorded as the connected contour of the human body unit.

[0016] Based on the positional relationship between relevant human body units and their related contours, the overall correlation index of each pair of relevant human body units is obtained;

[0017] Based on the contact profiles of the human body units and each of the human body units, the overall correlation index of each pair of human body units is obtained;

[0018] The overall correlation index of each pair of direct human units is denoted as 1;

[0019] For any two human body units that are not directly human body units, related human body units, or connected human body units, the overall correlation index between the two human body units is recorded as 0.

[0020] Furthermore, the specific method for obtaining the overall correlation index of each pair of related human body units based on the positional relationship between related human body units and their related contours includes:

[0021] For any pair of related human units, obtain the distance from the head position of any one of the human units in the pair to the relevant outline of the pair of related human units, and record the inversely proportional normalized result of the distance as the correlation index between the human unit and the other human unit in the pair of related human units.

[0022] The maximum value of the correlation index corresponding to each of the two human body units in the pair of related human body units is recorded as the overall correlation index of the pair of related human body units.

[0023] Furthermore, the specific method for obtaining the overall correlation index of each pair of human body units based on the connection profiles of the contact human body units and each of the contact human body units is as follows:

[0024] For any pair of connected human units, obtain the distance from the head position of any human unit in the pair to the connection outline of that human unit, and record the inversely proportional normalized result of this distance as the association index between that human unit and the other human unit in the pair.

[0025] The minimum value of the correlation index corresponding to each of the two human body units in the pair of related human body units is denoted as the overall correlation index of the pair of human body units.

[0026] Furthermore, the specific method for obtaining the degree of coordination between human body units based on the time-series changes in the overall correlation index between human body units includes:

[0027] For any two human units, the average of the overall correlation index of the two human units at all moments in the entire training process so far is denoted as the team coordination index of the two human units.

[0028] Define any single human body unit as the target unit;

[0029] For any human unit other than the target unit, the product of the team synergy index of that human unit and the target unit and the current corresponding overall correlation index is recorded as the synergy degree between that human unit and the target unit.

[0030] Furthermore, the method for obtaining the degree of gaze between human units based on the positional relationship between the head position, hand position, and eye gaze vector of the human unit includes the following specific methods:

[0031] For any human body unit other than the target unit, the vector from the head position of the human body unit to the hand position of the target unit is denoted as the gaze sign vector of the human body unit towards the target unit.

[0032] For any human body unit other than the target unit, obtain the cosine of the angle between the eye gaze vector of the human body unit and the gaze marker vector of the human body unit toward the target unit. The mean of the cosine value and 1 is recorded as the degree of gaze of the human body unit toward the target unit.

[0033] Furthermore, the specific methods for obtaining the tactical command level of each human unit include:

[0034] The coordination degree between all human units other than the target unit and the target unit is weighted and normalized. The result is used as the coordination weight between each human unit and the target unit. The gaze degree of each human unit other than the target unit towards the target unit is weighted and summed using the coordination weight. The result is used as the tactical command degree of the target unit.

[0035] Furthermore, the specific method for obtaining the tactical deployment probability of each human unit includes:

[0036] The product of the tactical command level of any human unit and the coordination level between the target unit and the human unit is denoted as the command index of the target unit and the human unit.

[0037] The maximum value of the command index of the target unit and all human units is recorded as the tactical deployment probability of the target unit.

[0038] Furthermore, the specific method for obtaining the firefight probability of each human unit based on the overall correlation index and coordination degree between human units includes:

[0039] Any human unit other than the target unit is recorded as a possible adversary unit of the target unit;

[0040] For any human unit other than the target unit and its potential adversaries, the product of the degree of cooperation between the target unit and the human unit, the inversely proportional normalized value of the degree of cooperation between the target unit and its potential adversaries and the human unit, and the overall correlation index between the target unit's potential adversaries and the human unit is taken as the adversarial probability of the target unit and its potential adversaries relative to the human unit; the maximum value among the adversarial probabilities of the target unit and its potential adversaries relative to all human units is taken as the adversarial index of the target unit and its potential adversaries.

[0041] The maximum value of the confrontation index between the target unit and all human units is denoted as the target unit's engagement probability.

[0042] Furthermore, the specific method for determining the training phase of each human unit based on its tactical deployment probability and engagement probability is as follows:

[0043] Obtain the maximum value between the tactical deployment probability and the engagement probability of the target unit, and record the difference between 1 and this maximum value as the reconnaissance probability of the target unit;

[0044] If the reconnaissance probability of a target unit is greater than both the tactical deployment probability and the engagement probability of the target unit, the target unit is currently in the reconnaissance phase.

[0045] If the tactical deployment probability of the target unit is greater than both the reconnaissance probability and the engagement probability of the target unit, the target unit is currently in the tactical deployment phase.

[0046] If the probability of a target unit engaging in combat is greater than both the probability of tactical deployment and the probability of reconnaissance, the target unit is currently in the engagement phase.

[0047] The beneficial effects of this invention are as follows: When the system host server performs fusion processing on the collected multi-dimensional dynamic data, due to the high concurrency of the multi-dimensional dynamic data, it is necessary to classify the multi-dimensional dynamic data into multiple data streams. The processing priority of each data stream is determined according to the needs of the training phase. This invention obtains the overall correlation index between human units by measuring the distance between the head position of a human unit and the outer contour of a building, thus initially determining the correlation between the positions of each pair of human units. In tactical combat training, team-based collaborative operations are mostly conducted. This invention, by obtaining the tactical deployment probability of each human unit, determines the likelihood of tactical hand signal communication between team members during collaborative operations. Tactical combat involves group collaborative operations, so team members will communicate about the battle situation. This invention, through the overall correlation index and degree of collaboration between human units, obtains the engagement probability of each human unit, thus determining the likelihood of an imminent engagement for each human unit. Thus, this invention determines the training stage of each human unit by the tactical deployment probability and engagement probability of each human unit, and then processes each data stream of each person with corresponding priority, reducing the processing delay of key information and improving the simulation credibility of the training system in dynamic confrontation scenarios. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a structural block diagram of a cross-screen human-computer interaction training system for tactical confrontation provided in one embodiment of the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. 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.

[0051] Please see Figure 1 The diagram illustrates a flowchart of a cross-screen human-computer interaction training system for tactical confrontation, provided by an embodiment of the present invention. The system includes:

[0052] Tactical confrontation training data acquisition module 101: Acquires the head position, hand position, eye gaze vector and several data streams of each human unit during the training process; acquires the outer contours of several buildings.

[0053] It should be noted that in a cross-screen human-computer interaction training system for tactical confrontation, the data processing flow mainly includes multi-source data acquisition, data fusion processing, and distribution and display of processing results. In this process, front-end sensors and interactive terminals acquire high-concurrency heterogeneous data in real time and transmit the data to the back-end. The back-end, based on a multi-modal fusion engine, completes cross-screen alignment, semantic association, and feature extraction of the data stream. It then generates tactical evaluation results and adversarial feedback instructions through a real-time analysis model. Finally, the optimized instructions are distributed to various terminal devices through a distributed interface, driving real-time interaction between the virtual battlefield environment and the training characters. This process requires the coordinated interaction of hardware and software; therefore, the hardware facilities are first deployed according to requirements to collect the data needed for interaction.

[0054] Specifically, in a cross-screen human-computer interaction training system for tactical confrontation, the hardware facilities in the multi-source data acquisition stage include, but are not limited to, VR headsets, tactical uniforms, tactical gloves, depth cameras, laser simulation emitters, laser bullet point acquisition hosts, and system host servers.

[0055] The VR headset includes: a tracking module comprising an inertial measurement unit consisting of a three-axis gyroscope, accelerometer, and magnetometer, which accurately tracks head position and posture to update the virtual environment in real time and ensure synchronization between the scene and user actions; a head-mounted display module for providing visual output and tracking user head movements to achieve dynamic visual updates; a stereo module to enhance immersion by providing stereo sound in conjunction with the visual output; a dual-microphone array for directional sound pickup; and an eye-tracking module for acquiring high-precision eye gaze vectors in real time.

[0056] The tactical uniform is equipped with accelerometers and gyroscopes at the shoulders, elbows, knees, ankles, and torso to collect detailed data on human movement.

[0057] The tactical glove has a finger tracking function to recognize hand gestures of human units during combat.

[0058] The specific arrangement of the depth cameras is as follows: several depth cameras are evenly arranged on the rigid truss at the top of the training field to acquire the position of each human body unit and the three-dimensional skeleton model in real time.

[0059] The laser simulator is used to simulate firearms and firing conditions;

[0060] The laser bullet point acquisition host is used to acquire laser positioning data in real time.

[0061] The system's host server is used for multimodal data fusion, adversarial calculations, and real-time rendering of the battlefield environment.

[0062] It should be noted that when using the system host server to fuse and process the collected multi-dimensional dynamic data, due to the high concurrency of the multi-dimensional dynamic data, in order to overcome the real-time processing bottleneck of the multi-dimensional dynamic data, it is necessary to classify the multi-modal data and then prioritize the classified multi-dimensional dynamic data according to the real-time requirements of each scenario.

[0063] Specifically, the data to be processed by the cross-screen human-computer interaction training system for tactical combat will be divided into data streams as follows:

[0064] Tactical command flow: including but not limited to voice signals and gestures;

[0065] Battlefield situation flow: including but not limited to 3D modeling data of human units, ammunition trajectory, and health information;

[0066] Equipment status flow: including but not limited to weapon ammunition quantity and medical equipment status;

[0067] Physiological behavioral flow: including but not limited to heart rate, eye movement trajectory;

[0068] Environmental simulation flow: including but not limited to physical collision feedback.

[0069] It should be noted that in the cross-screen human-computer interaction training process for tactical confrontation, the entire training process is mainly divided into the reconnaissance phase, the tactical deployment phase, and the engagement phase. Each phase has different requirements for the real-time processing of each multimodal data classification, so it is necessary to first determine the current phase of the training process. Determining the training phase requires collecting the data upon which this determination is based.

[0070] Specifically, obtain the CAD environment model of the training site used for this training, and extract the outer contours of all buildings from the CAD environment model;

[0071] Using a depth camera combined with HMR human body mesh reconstruction technology, the head and hand positions of each human body unit are acquired at a frequency of 30Hz; HMR human body mesh reconstruction technology is a well-known technology, and the specific method will not be described here.

[0072] The eye gaze vector for each human body unit is acquired at a frequency of 30Hz using the eye-tracking module in the VR headset.

[0073] Human Body Overall Correlation Index Acquisition Module 102: Based on the distance relationship between the head position of the human body unit and the outer contour of the building, the overall correlation index between human body units is obtained.

[0074] It should be noted that the following analysis only considers a specific moment in the tactical combat training process to determine the training phase at that moment. Since the training primarily focuses on player-to-player combat, when two human units are close enough to potentially engage in firefight, it's highly likely they are at a scene transition point. Situations where firefight is imminent or ongoing are categorized as follows: 1. The two human units are directly connected by a straight line, with no obstacles; 2. Both units can simultaneously observe the edge of the same building; 3. The edges of the buildings observed by both units can be directly connected by a straight line. Therefore, this logic is used to establish the interactive relationship between every two human units.

[0075] Specifically, for any human body unit, a building outline with several building outline points that are connected to the head position of the human body unit by a straight line, and the straight line does not pass through the building outline, is recorded as the direct contact outline of the human body unit.

[0076] For any two human body units, if the head positions of the two human body units are connected by a straight line and the straight line does not pass through a building, the two human body units are recorded as a pair of direct human body units.

[0077] If the two human body units have the same direct connection contour, the two human body units are recorded as a pair of related human body units, and the same direct connection contour is recorded as the related contour of the pair of related human body units.

[0078] If there is a connecting straight line between several building outer contour points on the two directly connected contours of the two human body units, and the straight line does not pass through the buildings, the two human body units are recorded as a pair of connected human body units; for any one of the human body units in the pair of connected human body units, the directly connected contour with the smallest distance from the head position of the human body unit among the two directly connected contours is recorded as the connected contour of the human body unit.

[0079] It should be noted that when a pair of related human units are enemies, if one of the human units in the pair is relatively close to the relevant outline, it means that the human unit is more likely to cross the corresponding relevant outline and become a direct human unit. That is, after crossing the corresponding relevant outline, the human unit may be directly exposed to the field of vision of the other human unit in the pair, and there is a high probability of engaging in combat.

[0080] Specifically, for any pair of related human units, the distance from the head position of any one of the human units in the pair to the relevant contour of the pair of related human units is obtained, and the inversely proportionally normalized result of this distance is recorded as the correlation index between the human unit and the other human unit in the pair of related human units.

[0081] The maximum value of the correlation index corresponding to each of the two human body units in the pair of related human body units is recorded as the overall correlation index of the pair of related human body units.

[0082] It should be noted that when either human unit in a pair of related human units crosses the corresponding related outline, that is, when the correlation index between either human unit and the other human unit in the pair of related human units is large, the two sides can engage in combat.

[0083] It should be noted that in combat training, scenarios often arise where opposing sides confront each other using cover. When both sides are located at the edges of cover, and the edges of the cover can be connected in a straight line without obstruction through the central passage, even if both sides are themselves covered by cover, they may still engage in firefights through the open area of ​​the passage. Therefore, when a pair of contact units are enemies, if each unit in the pair is relatively close to its respective contact outline, the contact outline may be the cover outline of both sides, thus increasing the likelihood of a firefight.

[0084] Specifically, for any pair of connected human units, the distance from the head position of any one of the human units in the pair to the connection outline of that human unit is obtained, and the inversely proportionally normalized result of this distance is recorded as the association index between that human unit and the other human unit in the pair.

[0085] The minimum value of the correlation index corresponding to each of the two human body units in the pair of related human body units is denoted as the overall correlation index of the pair of human body units.

[0086] It should be noted that when the two human units in a pair of contact human units are relatively close to their respective contact outlines, it indicates that the pair of contact human units may be behind corresponding cover and engaging in combat.

[0087] It should be further explained that for each pair of direct human units, since both sides are exposed to each other's line of sight, the possibility of firefights is extremely high.

[0088] Specifically, the overall correlation index for each pair of direct human body units is denoted as 1.

[0089] It should be noted that for any two human units that are not directly human units, related human units, or connected human units, firefights cannot be conducted due to the excessive distance and the presence of buildings obstructing the view.

[0090] Specifically, for any two human body units that do not belong to direct human body units, related human body units, or connected human body units, the overall correlation index between the two human body units is recorded as 0.

[0091] Tactical Deployment Probability Analysis Module 103: Based on the temporal changes in the overall correlation index between human units, the degree of coordination between human units is obtained; based on the positional relationship between the head position, hand position, and eye gaze vector of human units, the degree of gaze between human units is obtained; combined with the degree of coordination between human units, the degree of tactical command of each human unit is obtained, and thus the tactical deployment probability of each human unit is obtained.

[0092] It should be noted that since tactical combat training generally involves small-team coordinated operations, hand gestures are required during coordination. Data from hand gestures often has a high priority during the tactical deployment phase. For example, when preparing to launch an attack, the squad leader will issue instructions to the team members using predefined attack gestures, thereby enabling the entire squad to attack simultaneously.

[0093] It should be further explained that during tactical confrontation, since squad members with the same temporary combat objective are often close to each other during squad-level coordinated operations, the overall correlation index between squad members is consistently high. Therefore, the possibility of each pair of human units having the same temporary combat objective is first determined based on the change in the overall correlation index between each pair of human units.

[0094] Specifically, for any two human units, the average of the overall correlation index of the two human units at all moments in the entire training process so far is denoted as the team coordination index of the two human units.

[0095] It should be noted that the higher the squad coordination index of two human units, the greater the likelihood that the two human units share the same temporary combat objective.

[0096] It should be noted that tactical communication will occur during squad-level coordinated operations. Gestures are the main means of tactical communication. During combat, since each human unit's gestures involve gripping combat equipment, they generally do not provide valuable assistance in the fight. However, during the tactical deployment phase, gesture commands represent issuing orders and need to be processed in real time. Therefore, it is necessary to accurately identify the tactical deployment phase.

[0097] It should be further explained that the hand gesture is a signal issued by the team leader to the team members. Therefore, the human unit that receives the signal is the team member of the team leader. So, whether the team leader issues a hand gesture is determined by the actions of the team members. Therefore, it is necessary to first confirm the degree of coordination between the human units.

[0098] Specifically, any human unit is designated as the target unit. For any human unit other than the target unit, the product of the team coordination index of the human unit and the target unit and the current overall correlation index is recorded as the coordination degree between the human unit and the target unit. It should be noted that, except for the team coordination index acquisition process which needs to consider all moments up to now, subsequent processing is based on the data at the current corresponding moment.

[0099] It should be noted that the higher the team coordination index between a human unit and a target unit, the more likely that the human unit and the target unit belong to the same team with the same temporary combat objective. At the same time, the higher the overall correlation index between the human unit and the target unit, the more likely that the human unit and the target unit are currently in the same scene location. The greater the degree of coordination between the human unit and the target unit, the more likely that the human unit and the target unit have a cooperative relationship at this time.

[0100] It should be further explained that when the squad leader gives a hand gesture, the squad members will face the squad leader and focus their eyes on the squad leader's hand gesture.

[0101] Specifically, for any human body unit other than the target unit, the vector from the head position of the human body unit to the hand position of the target unit is denoted as the gaze sign vector of the human body unit towards the target unit.

[0102] For any human body unit other than the target unit, obtain the cosine of the angle between the eye gaze vector of the human body unit and the gaze marker vector of the human body unit toward the target unit. The mean of the cosine value and 1 is recorded as the degree of gaze of the human body unit toward the target unit.

[0103] As an example, the degree of fixation can be expressed by the formula:

[0104]

[0105] In the formula, A a θ represents the degree of fixation of the a-th individual human unit (excluding the target unit) towards the target unit; a Let be the angle between the eye gaze vector of the a-th human unit (excluding the target unit) and the gaze marker vector of the a-th human unit (excluding the target unit) towards the target unit; cos() is the cosine function.

[0106] It should be noted that A a The larger the value, the more likely the a-th human unit other than the target unit is to be looking at the target unit's hand position, and the more likely the target unit is to issue a gesture command.

[0107] It should be noted that if the human units that are cooperating with the target unit at this time all have a high degree of attention to the target unit, it indicates that the target unit may be the captain who issued the hand gesture command. Therefore, this can be used to judge the tactical deployment phase.

[0108] Specifically, the coordination degree between all human units other than the target unit and the target unit is weighted and normalized. The result is used as the coordination weight between each human unit and the target unit. The coordination weight is then used to weight and sum the gaze degree of each human unit other than the target unit towards the target unit. The result is used as the tactical command degree of the target unit.

[0109] As an example, the level of tactical command can be expressed by the formula:

[0110]

[0111] In the formula, B represents the tactical command level of the target unit; D represents the number of human units participating in the combat training; C a The degree of coordination between the a-th human unit (excluding the target unit) and the target unit; A a Let C be the degree of gaze of the a-th human body unit (excluding the target unit) toward the target unit; softmax() is the weight normalization function, which normalizes the C values ​​of all human body units except the target unit. a .

[0112] It should be noted that if the target unit has a high degree of tactical command, it means that the target unit leader is commanding the team members, so the team members are more likely to be commanded, that is, the team members and the leader are more likely to be in the tactical deployment stage.

[0113] Specifically, the product of the tactical command level of any human unit and the coordination level between the target unit and the human unit is recorded as the command index of the target unit and the human unit.

[0114] The maximum value of the command index of the target unit and all human units is recorded as the tactical deployment probability of the target unit.

[0115] As an example, the probability of tactical deployment can be expressed by the formula:

[0116] G = MAX(C e ×B e )

[0117] In the formula, G represents the tactical deployment probability of the target unit; C e B represents the degree of coordination between the e-th individual unit and the target unit. e The tactical command level of the e-th human unit; MAX(C e ×B e) is for all human body units (C e ×B e The maximum value of ).

[0118] It should be noted that (C) e ×B e ) represents the command index of the target unit and the e-th individual unit, (C) e ×B e The larger C is, the greater the probability that the target unit is being tactically commanded by the e-th human unit; when the e-th human unit is the target unit itself, C e Choose 1, meaning the target unit itself is a possible commander.

[0119] Firefight Probability Analysis Module 104: Based on the overall correlation index and coordination degree between human units, the firefight probability of each human unit is obtained.

[0120] It should be noted that the overall correlation index between human units is calculated based on the relative positional relationship between two human units. Since tactical combat involves group coordinated operations, members of a single squad will communicate the battle situation. When the overall correlation index between any human unit in the squad and an enemy human unit is relatively high, then the probability of each human unit in that squad engaging in combat with an enemy human unit should be relatively high. This improves the processing efficiency of the data generated by relevant human units when the two sides engage in combat.

[0121] Specifically, any human unit other than the target unit is recorded as a possible adversary unit of the target unit;

[0122] For any human unit other than the target unit and its potential adversaries, the product of the degree of cooperation between the target unit and the human unit, the inversely proportional normalized value of the degree of cooperation between the target unit and the human unit and its potential adversaries, and the overall correlation index between the target unit and the human unit is taken as the adversarial probability of the target unit and its potential adversaries relative to the human unit; the maximum value among the adversarial probabilities of the target unit and its potential adversaries relative to all human units is taken as the adversarial index of the target unit and its potential adversaries.

[0123] As an example, the adversarial index is expressed by the formula:

[0124] E = MAX(C d ×exp(-C d ′ )×F d )

[0125] In the formula, E is the adversarial index between the target unit and its potential adversaries; C dC represents the degree of coordination between the target unit and the d-th human unit, excluding the target unit and its potential adversaries; d ′ F represents the degree of cooperation between the target unit's potential adversaries and the d-th human unit excluding the target unit and its potential adversaries; d MAX(C) is the overall correlation index between the target unit's potential adversaries and the d-th human unit excluding the target unit and its potential adversaries; d ×exp(-C d ′ )×F d ) is the target unit and its potential adversaries relative to all human units other than the target unit and its potential adversaries (C d ×exp(-C d ′ )×F d The maximum value is given by exp(), which is an exponential function with the natural constant as its base.

[0126] It should be noted that C d The larger the value, the more likely the target unit and the d-th human unit (excluding the target unit and its potential adversaries) are members of the same squad, exp(-C d ′ The larger the value of F, the more likely the target unit's potential adversaries and the d-th human unit (excluding the target unit and its potential adversaries) are not members of the same squad; d The larger E is, the more likely the target unit's potential adversaries will encounter the dth human unit other than the target unit and its potential adversaries. The larger E is, the more likely the target unit's squad will encounter the squads of its potential adversaries.

[0127] It should be noted that during tactical confrontation training, when two squads meet, regardless of whether they are in a confrontational relationship, the battlefield situation flow must be elevated to the highest priority. This is because the meeting itself will significantly change the battlefield dynamics, and the enemy's coordinates and weapon status need to be updated in real time to support rapid decision-making. By processing the data with the highest priority, the system can ensure low latency and high reliability of the data exchanged between the two sides, providing key support for the prevention of sudden conflicts and tactical coordination.

[0128] Furthermore, the maximum value of the confrontation index between the target unit and all human units is denoted as the target unit's engagement probability.

[0129] Data stream processing module 105: Determines the training stage of each human unit based on the tactical deployment probability and engagement probability of each human unit, and processes each data stream of each human unit with corresponding priority according to the training stage of each human unit.

[0130] It should be noted that when using the system host server to fuse and process the collected multi-dimensional dynamic data, due to the high concurrency of the multi-dimensional dynamic data, different data streams need to be processed with different priorities according to the needs of different scenarios.

[0131] It should be further explained that, due to the different divisions of squads and the different scenarios during tactical combat training, the battlefield stage of each human unit will vary. Therefore, it is necessary to determine the battlefield stage of each human unit based on its tactical deployment probability and engagement probability.

[0132] Specifically, obtain the maximum value between the tactical deployment probability and the engagement probability of the target unit, and record the difference between 1 and this maximum value as the reconnaissance probability of the target unit;

[0133] If the reconnaissance probability of a target unit is greater than both the tactical deployment probability and the engagement probability of the target unit, the target unit is currently in the reconnaissance phase.

[0134] If the tactical deployment probability of the target unit is greater than both the reconnaissance probability and the engagement probability of the target unit, the target unit is currently in the tactical deployment phase.

[0135] If the probability of a target unit engaging in combat is greater than both the probability of tactical deployment and the probability of reconnaissance, the target unit is currently in the engagement phase.

[0136] The priority order of all data streams in the reconnaissance phase, tactical deployment phase, and engagement phase is as follows:

[0137] Reconnaissance Phase: Environmental simulation flow, physiological behavior flow, tactical command flow, battlefield situation flow, equipment status flow;

[0138] Tactical deployment phase: tactical command flow, battlefield situation flow, equipment status flow, physiological behavior flow, and environmental simulation flow;

[0139] Engagement phase: Battlefield situation flow, equipment status flow, tactical command flow, physiological behavior flow, and environmental simulation flow.

[0140] Based on the above method, the stage of each human unit at each moment in the tactical confrontation training is determined. According to the data flow priority of each human unit at each moment, the system host server processes the collected data in real time, thereby improving the data processing efficiency of the system host server at different battlefield stages.

[0141] This embodiment uses the exp(-MX) model to represent the inverse proportional relationship and normalization processing. MX is the input of the model, and the implementer can set the inverse proportional function and normalization function according to the actual situation.

[0142] 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, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A cross-screen human-computer interaction training system for tactical combat, characterized in that, The system includes: The tactical combat training data acquisition module is used to acquire the head position, hand position, eye gaze vector, and several data streams of each human unit during the training process; and to acquire the outer contours of several buildings. The overall human body correlation index acquisition module is used to obtain the overall correlation index between human body units based on the distance relationship between the head position of the human body unit and the outer contour of the building. The tactical deployment probability analysis module is used to obtain the degree of coordination between human units based on the temporal changes of the overall correlation index between human units; to obtain the degree of gaze between human units based on the positional relationship between the head position, hand position, and eye gaze vector of the human units; and to obtain the tactical command degree of each human unit by combining the degree of coordination between the human units, thereby obtaining the tactical deployment probability of each human unit. The crossfire probability analysis module is used to obtain the crossfire probability of each human unit based on the overall correlation index and coordination degree between human units; The data stream processing module is used to determine the training stage of each human unit based on the tactical deployment probability and engagement probability of the human unit, and to process each data stream of each human unit with corresponding priority according to the training stage of each human unit. The method for obtaining the overall correlation index between human body units based on the distance relationship between the head position of the human body unit and the outer contour of the building includes: For any human body unit, obtain the direct contact contour of the human body unit by forming a straight line connecting several building outline points to the head position of the human body unit, and the straight line does not pass through the building outline. For any two human body units, if the head positions of the two human body units are connected by a straight line and the straight line does not pass through a building, the two human body units are recorded as a pair of direct human body units. If the two human body units have the same direct connection contour, the two human body units are recorded as a pair of related human body units, and the same direct connection contour is recorded as the related contour of the pair of related human body units. If there is a connecting straight line between several building outer contour points on the two directly connected contours of the two human body units, and the straight line does not pass through the buildings, the two human body units are recorded as a pair of connected human body units; for any one of the human body units in the pair of connected human body units, the directly connected contour with the smallest distance from the head position of the human body unit among the two directly connected contours is recorded as the connected contour of the human body unit. Based on the positional relationship between relevant human body units and their related contours, the overall correlation index of each pair of relevant human body units is obtained; Based on the contact profiles of the human body units and each of the human body units, the overall correlation index of each pair of human body units is obtained; The overall correlation index of each pair of direct human units is denoted as 1; For any two human body units that are not directly human body units, related human body units, or connected human body units, the overall correlation index between the two human body units is recorded as 0.

2. The cross-screen human-computer interaction training system for tactical confrontation as described in claim 1, characterized in that, The method for obtaining the overall correlation index of each pair of related human body units based on the positional relationship between related human body units and their related contours includes: For any pair of related human units, obtain the distance from the head position of any one of the human units in the pair to the relevant outline of the pair of related human units, and record the inversely proportional normalized result of the distance as the correlation index between the human unit and the other human unit in the pair of related human units. The maximum value of the correlation index corresponding to each of the two human body units in the pair of related human body units is recorded as the overall correlation index of the pair of related human body units.

3. The cross-screen human-computer interaction training system for tactical confrontation as described in claim 1, characterized in that, The method for obtaining the overall correlation index of each pair of human body units based on the connection profile of each human body unit includes: For any pair of connected human units, obtain the distance from the head position of any human unit in the pair to the connection outline of that human unit, and record the inversely proportional normalized result of this distance as the association index between that human unit and the other human unit in the pair. The minimum value of the correlation index corresponding to each of the two human body units in the pair of related human body units is denoted as the overall correlation index of the pair of human body units.

4. The cross-screen human-computer interaction training system for tactical confrontation as described in claim 1, characterized in that, The method for obtaining the degree of coordination between human body units based on the time-series changes in the overall correlation index between human body units includes the following specific methods: For any two human units, the average of the overall correlation index of the two human units at all moments in the entire training process so far is denoted as the team coordination index of the two human units. Define any single human body unit as the target unit; For any human unit other than the target unit, the product of the team synergy index of that human unit and the target unit and the current corresponding overall correlation index is recorded as the synergy degree between that human unit and the target unit.

5. A cross-screen human-computer interaction training system for tactical confrontation as described in claim 4, characterized in that, The method for determining the degree of gaze between human units based on the positional relationship between the head, hand, and eye gaze vectors of the human unit includes the following specific methods: For any human body unit other than the target unit, the vector from the head position of the human body unit to the hand position of the target unit is denoted as the gaze sign vector of the human body unit towards the target unit. For any human body unit other than the target unit, obtain the cosine of the angle between the eye gaze vector of the human body unit and the gaze marker vector of the human body unit toward the target unit. The mean of the cosine value and 1 is recorded as the degree of gaze of the human body unit toward the target unit.

6. The cross-screen human-computer interaction training system for tactical confrontation as described in claim 5, characterized in that, The specific methods for obtaining the tactical command level of each human unit are as follows: The coordination degree between all human units other than the target unit and the target unit is weighted and normalized. The result is used as the coordination weight between each human unit and the target unit. The gaze degree of each human unit other than the target unit towards the target unit is weighted and summed using the coordination weight. The result is used as the tactical command degree of the target unit.

7. A cross-screen human-computer interaction training system for tactical confrontation as described in claim 6, characterized in that, The specific method for obtaining the tactical deployment probability of each human unit is as follows: The product of the tactical command level of any human unit and the coordination level between the target unit and the human unit is denoted as the command index of the target unit and the human unit. The maximum value of the command index of the target unit and all human units is recorded as the tactical deployment probability of the target unit.

8. A cross-screen human-computer interaction training system for tactical confrontation as described in claim 4, characterized in that, The method for obtaining the firefight probability of each human unit based on the overall correlation index and coordination degree between human units includes the following specific methods: Any human unit other than the target unit is recorded as a possible adversary unit of the target unit; For any human unit other than the target unit and its potential adversaries, the product of the degree of cooperation between the target unit and the human unit, the inversely proportional normalized value of the degree of cooperation between the target unit and its potential adversaries and the human unit, and the overall correlation index between the target unit's potential adversaries and the human unit is taken as the adversarial probability of the target unit and its potential adversaries relative to the human unit; the maximum value among the adversarial probabilities of the target unit and its potential adversaries relative to all human units is taken as the adversarial index of the target unit and its potential adversaries. The maximum value of the confrontation index between the target unit and all human units is denoted as the target unit's engagement probability.

9. A cross-screen human-computer interaction training system for tactical confrontation as described in claim 4, characterized in that, The method for determining the training phase of each human unit based on its tactical deployment probability and engagement probability is as follows: Obtain the maximum value between the tactical deployment probability and the engagement probability of the target unit, and record the difference between 1 and this maximum value as the reconnaissance probability of the target unit; If the reconnaissance probability of a target unit is greater than both the tactical deployment probability and the engagement probability of the target unit, the target unit is currently in the reconnaissance phase. If the tactical deployment probability of the target unit is greater than both the reconnaissance probability and the engagement probability of the target unit, the target unit is currently in the tactical deployment phase. If the probability of a target unit engaging in combat is greater than both the probability of tactical deployment and the probability of reconnaissance, the target unit is currently in the engagement phase.

Images