Online fire-fighting training simulation training system and training method thereof

Abstract

The invention discloses an online fire-fighting training simulation training system and a training method thereof. The system comprises a sensing module, a virtual environment module, an interaction and feedback module and a learning recording and suggestion module. According to the invention, multiple sensory experiences such as smell, temperature, wind feeling and the like are comprehensively introduced into fire-fighting simulation training, the limitation that a traditional VR / AR system purely depends on vision and hearing is broken, the sense of reality and immersion of training are greatly improved, and the coping ability of students in a real fire scene is effectively improved; the state of a trainee is analyzed in real time through an AI algorithm, a self-adaptive threshold intervention mechanism is combined, a virtual AI coach can provide dynamic and personalized guidance in different training scenes, and intelligent intervention based on real-time data greatly improves the fire-fighting training effect; the system not only pays attention to the short-term training effect of the student, but also makes a personalized continuous learning plan through the learning recording and suggestion module, and promotes the long-term progress of the student.

Description

Technical Field

[0001] The present invention relates to the technical field related to fire training, and in particular to an online fire training simulation training system and a training method thereof. Background Art

[0002] As society pays more and more attention to fire safety, fire training has become one of the essential skills training for practitioners in all walks of life. Traditional fire training methods usually rely on physical simulation training venues, such as fire simulators, fire extinguishing drills, etc. However, the implementation cost of these physical simulation trainings is high, the maintenance and consumption of venues and equipment are huge, and the number and scenes of training are also subject to physical limitations, making it difficult to achieve a high degree of realistic restoration and repeated practice. In addition, it is difficult for traditional training to cover different types of fire scenes in a short period of time, resulting in limited training effects.

[0003] In recent years, virtual reality (VR) and augmented reality (AR) technologies have been gradually applied to the field of fire training, attempting to make up for the shortcomings of traditional training methods. Such technologies can construct realistic virtual fire scenes, and through three-dimensional vision and spatial sound effects, help trainees conduct realistic fire simulation training in a safe environment. However, the existing system lacks multi-sensory deep immersion experience, such as smell, temperature, touch, etc., which makes the training lack of realism and immersion; in addition, the existing VR / AR training system still has significant deficiencies in real-time interactivity, personalized feedback, and trainee status perception. For example, the existing system often lacks real-time monitoring of trainees' emotions and physical conditions (such as heart rate), and cannot dynamically adjust training content and feedback, which may cause trainees to behave abnormally under high-pressure environments, and the system fails to intervene in time for effective guidance, affecting the effectiveness of training. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] Therefore, in order to solve the above technical problems, the present invention provides the following technical solutions: an online fire training simulation training system, comprising a perception module, a virtual environment module, an interaction and feedback module, and a learning record and suggestion module;

[0006] Among them, the perception module is closely integrated with the virtual environment module to capture the trainees' movements and voice inputs in real time, and drive the changes of events in the virtual environment; the virtual environment module and the interaction and feedback module work together to enhance the trainees' sense of immersion and interaction through the feedback of the virtual AI coach and the dynamic changes of the environment; the interaction and feedback module relies on the data provided by the perception module, combines the AI ​​algorithm for real-time analysis, and provides personalized improvement suggestions through the learning record and suggestion module; the learning record and suggestion module not only records the trainees' performance, but also shares data with the interaction and feedback module to generate personalized learning plans and promote the continuous progress of trainees.

[0007] As a preferred solution of the online fire training simulation training system described in the present invention, the perception module is responsible for collecting the trainees' movements, voice and physiological data for real-time feedback and interaction in a virtual environment; the perception module includes motion capture equipment (such as gloves, cameras, head-mounted display devices), heart rate monitoring equipment, microphones, etc.; this module mainly collects the trainees' postures, movements, voice commands and physiological reactions (such as heart rate, breathing rate, etc.) during the training process as a basis for the system to judge the trainees' behavior and training effects.

[0008] As a preferred solution of the online fire training simulation training system described in the present invention, the virtual environment module (virtual reality + augmented reality layer) is used to provide a highly realistic fire simulation environment, supporting multi-sensory input such as vision, hearing, touch, and smell; the virtual environment module includes VR or AR head-mounted devices, environmental sensing equipment (such as odor generators, temperature regulators, fans, etc.), high-fidelity audio systems, and 3D environment simulation engines, physical simulation engines (used to simulate dynamic effects such as flames and smoke), and multi-sensory stimulation control systems; this module is used to construct a highly realistic fire environment, including combustion scenes, smoke diffusion, temperature changes, odor simulation, etc., and at the same time combines multi-sensory stimulation such as sound and light effects to enhance students' perception of fire scenes.

[0009] As a preferred solution of the online fire training simulation training system described in the present invention, the interaction and feedback module (analysis and guidance layer) integrates a virtual AI coach and an AI analysis system, and dynamically analyzes the training process according to the real-time behavior and physiological data of the trainees; the trainees ask questions through voice, and the virtual AI coach gives voice or text guidance according to the situation, providing real-time interaction and intelligent feedback; the virtual AI coach system collects data including trainee behavior data, physiological data, voice input and environment; specifically, through sensors and motion capture devices, the trainees' action data in the virtual training environment, such as body posture, steps, arm movements, fire extinguishing operations, etc., are obtained; the trainees' physiological reaction data, such as heart rate, respiratory rate, sweat conditions, etc., are collected through heart rate monitoring equipment, temperature sensors and other equipment. These data can reflect the physiological changes of trainees in a tense environment and help evaluate psychological coping ability; the trainees' voice commands or inquiries are collected through microphones and other equipment to ensure that the virtual AI coach can recognize the questions raised by the trainees and respond; scene information, fire progress, smoke diffusion path and other data in the virtual environment also need to be input into the virtual AI coach so that it can provide guidance to the trainees according to the dynamic changes of the environment.

[0010] As a preferred solution of the online fire training simulation training system of the present invention, the virtual AI coach collects the trainee's physiological and behavioral data (such as action accuracy, reaction time, heart rate, etc.) and uses machine learning algorithms and dynamic feedback rules to adjust its guidance content in real time;

[0011] The specific steps are as follows: Define the following variables and parameters: Trainee training status : Indicates the training status of the trainee at time t; movement accuracy : Indicates the accuracy of the student's action at time t; reaction time : Indicates the trainees’ emergency response time to fire environment; heart rate changes : Indicates the real-time heart rate changes of trainees during training; baseline heart rate : Indicates the heart rate under normal conditions. The fluctuation of heart rate reflects the physiological load of the students; Anxiety index : Indicates the anxiety level of the trainees; the complexity of the fire state : Indicates the complexity of the current fire simulation environment; task completion : Indicates the degree of completion of the student in the task; the student's training status The calculation formula is as follows: F ( t ) = ∫ 0 t [ A ( t ) ⋅ e − l R ( t ) + 1 1 + e − α [ H ( t ) − H 0 ] ⋅ 1 n ( 1 + x ( t ) ) ] ⋅ [ 1 + sin ( π 2 F ( t ) ) ] d t ∫ 0 t [ C ( t ) ⋅ e − c ( t − t ) ] d t ;in, :This item reflects the student's action performance and reaction time at time t; action accuracy directly affects the feedback result calculated by the formula, and the reaction time By exponential decay factor The intensity of feedback that affects the feedback result calculated by this formula. Specifically, the faster the response, the more timely the feedback. 1 1 + e − α [ H ( t ) − H 0 ] ⋅ 1 n ( 1 + x ( t ) ) : This is a complex interaction term that takes into account the student's physiological state; in this formula, It is an adjustment parameter used to control the intensity of the impact of heart rate changes on overall anxiety. Relative to baseline heart rate The sigmoid function is used to control the range of influence and the anxiety index The growth is then harmonized by a logarithmic function to prevent over-amplification; [ 1 + sin ( π 2 F ( t ) ) ] :This item introduces the complexity of the fire simulation environment and modulates the fire complexity through a sine function, which facilitates dynamic adjustment of the feedback intensity. For example, when the fire is more dangerous, the coach's feedback will be stronger; :This item is in the denominator, indicating the weight of the student's task completion, using an exponential decay factor Reduce the impact of early task performance and emphasize students' task performance close to the current time; The formula range is: The value range of represents the training status of the students, which in turn determines the intensity and frequency of the virtual AI coach's real-time feedback; a larger value means that the virtual AI coach needs to give stronger feedback and correct the students' mistakes in time; a smaller value means that the virtual AI coach needs to reduce intervention and allow students to play freely.

[0012] As a preferred solution of the online fire training simulation training system of the present invention, the threshold is set based on the initial ability and training goals of different trainees. ,when > , it means that the trainee may have serious problems in training (such as slow reaction, inaccurate movements, and excessive anxiety). At this time, the system determines that the trainee's training status requires intervention, and the virtual AI coach will strengthen feedback or provide specific guidance; when ≤ , it indicates that the trainee is in good condition, and the system determines that the virtual system needs to reduce intervention, allowing the trainee to continue training in the current way.

[0013] As a preferred solution of the online fire training simulation system of the present invention, the threshold The calculation method is as follows:

[0014] Define the following variables and parameters: : Defined as the baseline value of the initial anxiety index, indicating the normal anxiety level without excessive psychological burden; : defined as the attenuation coefficient of the anxiety index, used to control the time effect of the anxiety state; T: defined as the total training time of the trainees; sensitivity parameter : Used to control the impact of heart rate fluctuations on the threshold. A higher sensitivity parameter value means that heart rate fluctuations have a stronger impact on the threshold. Complexity influence coefficient : Used to adjust the impact of complex environments on the threshold; then, the threshold The calculation formula is as follows: i = ∫ 0 T [ 1 1 + e − c ( H ( t ) − H 0 ) ⋅ 1 1 + β ⋅ F ( t ) + x 0 ⋅ e − d ⋅ x ( t ) ] d t ; In the above formula, the threshold The value range is between.

[0015] As a preferred solution of the online fire training simulation training system described in the present invention, the learning record and suggestion module (storage and optimization layer) is responsible for recording the trainees' training data and providing personalized learning suggestions and continuous improvement guidance after the training; the module includes a learning analysis system and a personalized recommendation algorithm. The learning analysis system is used to record each training data of the trainees and generate a personalized evaluation report. The personalized recommendation algorithm is used to provide personalized improvement suggestions and the next training plan based on the trainees' performance to help the trainees better master their skills.

[0016] The present invention also provides a training method for the above-mentioned online fire training simulation training system, which specifically includes the following steps: S1: the system constructs a fire simulation environment for trainees through VR / AR equipment; S2: the training task is started, and the trainees perform task operations in the virtual environment, such as finding a fire extinguisher, extinguishing the flame, rescuing trapped persons, etc.; during this period, the system provides dynamic feedback (such as changes in fire intensity, smoke diffusion, temperature rise, etc.), and uses multi-sensory simulation (smell, heat sensation, wind sensation, etc.) to enhance the sense of reality; S3: the trainees ask questions to the virtual AI coach through voice, such as how to use a fire extinguisher, how to deal with a certain fire, etc.; the virtual AI coach analyzes the problems in real time based on the AI ​​algorithm and gives answers, and provides timely voice guidance according to the trainees' performance (whether the movements are correct, whether they are nervous, etc.); S4: after the training is completed, the system automatically evaluates the trainees' performance, generates a training report (including the trainees' reaction speed, task completion status, error analysis, etc.), and makes improvement suggestions (such as improving fire extinguishing posture, increasing escape speed, etc.) and training improvement suggestions based on the training data and the trainees' performance.

[0017] As a preferred embodiment of the online fire training simulation training method of the present invention, at the beginning of the training, the trainees perform well, the values ​​of the trainees' heart rate changes and anxiety index are low, and the complexity of the fire scene is relatively small. At this time, the threshold Higher, assuming The value is also low. The virtual AI coach allows students to complete tasks independently without excessive intervention. As the complexity of the fire scene increases, the fire spreads rapidly, and the heart rate changes and anxiety index values ​​of the students increase. The threshold Start to decline, if at this time Rising and exceeding the threshold , indicating that the trainee's performance has fallen below expectations, the virtual AI coach will immediately intervene and provide specific guidance, such as prompting the trainee to breathe calmly or providing route optimization suggestions; in the final stage of training, the trainee may gradually recover through the feedback of the system, the trainee's heart rate change and anxiety index values ​​begin to decline, and the fire complexity also decreases. If at this time Below the threshold again , the virtual AI coach will reduce intervention and allow trainees to complete tasks independently.

[0018] Beneficial effects of the present invention: 1. The perception module of the present invention can capture the trainee's movements, voice input, heart rate and emotional changes in real time. Through close integration with the virtual environment module, the system can dynamically adjust the scene complexity and the feedback of the virtual AI coach according to the trainee's state, ensuring that the trainee is effectively monitored and guided at every step in the training. For example, if the heart rate change and the anxiety index of the trainee are detected to rise sharply, the system will automatically reduce the complexity of the fire scene, or start the virtual AI coach to provide coping strategies and guidance. This real-time perception and dynamic feedback function significantly improves the personalization and effectiveness of training. 2. The present invention introduces multi-sensory simulation technologies such as smell, temperature, and wind sensation, and through comprehensive sensory stimulation, enables trainees to have a more realistic experience in the fire scene; for example, trainees can smell the smoke and feel the heat changes brought by the flame. This multi-sensory design greatly enhances the realism of the scene and the immersion of the training, and effectively improves the trainees' coping ability in real fire scenes. 3. The interaction and feedback module of the present invention analyzes the performance of the trainees in real time through the AI ​​algorithm, and provides personalized voice guidance and feedback through the virtual AI coach. The virtual AI coach can not only answer questions raised by trainees during training, but also adjust teaching content and intervention measures in real time according to the trainees' correctness of movements, operation fluency, psychological state, etc. For example, when a trainee shows anxiety or operates improperly, the virtual AI coach can intervene immediately, adjust the teaching strategy in real time according to the trainee's state, and provide appropriate and more targeted help and guidance to help the trainee adjust his state. And the threshold formula , realizing an adaptive intervention mechanism; when the trainee's state is stable, the virtual AI coach will reduce intervention and allow the trainee to complete the task independently; when the trainee's state deteriorates, the system will actively intervene to provide guidance. This flexible intervention mechanism helps to improve the efficiency of training, reduce the mistakes of trainees under high-pressure conditions, and avoid unnecessary intervention to interrupt the trainees' autonomous operations, thereby improving learning effects. 5. The learning record and suggestion module of the present invention generates detailed training reports and provides personalized improvement suggestions by recording the trainees' training performance. This module can formulate a learning plan for the next stage for the trainees based on the trainees' historical performance and current training data, and promote the trainees' continuous improvement in skills. For example, the system can recommend specific training modules to some trainees who are slow to respond or operate improperly to strengthen their weaknesses. Compared with the prior art, the present invention not only focuses on the real-time performance of the trainees, but also takes into account the long-term learning needs of the trainees, realizing personalized continuous training. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0020] Figure 1 FIG. 4 is a diagram of the overall system architecture of the present invention.

[0021] Figure 2 It is the work flow chart of the present invention. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0025] Example 1

[0026] Reference Figure 1~2 , which is the first embodiment of the present invention, provides an online fire training simulation system, including a perception module, a virtual environment module, an interaction and feedback module, and a learning record and suggestion module;

[0027] The perception module is responsible for collecting the trainees' movements, voice and physiological data for real-time feedback and interaction in the virtual environment; the perception module includes motion capture equipment (such as gloves, cameras, head-mounted display devices), heart rate monitoring equipment, microphones, etc.; this module mainly collects the trainees' postures, movements, voice commands and physiological reactions (such as heart rate, breathing rate, etc.) during the training process as the basis for the system to judge the trainees' behavior and training effects.

[0028] The virtual environment module (virtual reality + augmented reality layer) is used to provide a highly realistic fire simulation environment, supporting multi-sensory inputs such as vision, hearing, touch, and smell; the virtual environment module includes VR or AR head-mounted devices, environmental sensing devices (such as odor generators, temperature regulators, fans, etc.), high-fidelity audio systems, as well as 3D environment simulation engines, physical simulation engines (used to simulate dynamic effects such as flames and smoke), and multi-sensory stimulation control systems; this module is used to build a highly realistic fire environment, including combustion scenes, smoke diffusion, temperature changes, odor simulation, etc., and combines multi-sensory stimulations such as sound and light effects to enhance trainees' perception of fire scenes;

[0029] The interaction and feedback module (analysis and guidance layer) integrates the virtual AI coach and AI analysis system to provide real-time interaction and intelligent feedback. The module includes a built-in AI chip or cloud computing processor, a speech recognition and synthesis system, a virtual AI coach system (virtual AI coach), and a real-time data analysis system. It dynamically analyzes the training process of the trainees based on their real-time behavior and physiological data. The virtual AI coach gives voice or text guidance based on the situation. At the same time, trainees can ask questions through voice, and the system will answer their questions in real time and give targeted feedback and suggestions based on their performance. The data collected by the virtual AI coach include trainee behavior data, physiological data, voice input, and environmental data. Specifically, sensors and motion capture devices are used to obtain trainees' action data in the virtual training environment, such as body posture, steps, arm movements, fire-fighting operations, etc. The heart rate monitoring equipment, temperature sensors and other equipment are used to collect trainees' physiological response data, such as heart rate, breathing rate, sweat conditions, etc. These data can reflect the physiological changes of trainees in stressful environments and help evaluate their psychological coping abilities. The trainees’ voice commands or inquiries are collected through microphones and other devices to ensure that the virtual AI coach can recognize and respond to the questions raised by the trainees. The scene information, fire progress, smoke diffusion path and other data in the virtual environment also need to be input into the virtual AI coach so that it can provide guidance to the trainees based on the dynamic changes of the environment.

[0030] The virtual AI coach collects students' physiological and behavioral data (such as movement accuracy, reaction time, heart rate, etc.) and uses machine learning algorithms and dynamic feedback rules to adjust its guidance content in real time. The specific steps are as follows:

[0031] Define the following variables and parameters: Trainee training status : indicates the training status of the trainee at time t; Action accuracy: indicates the accuracy of the trainee's action at time t; Reaction time : Indicates the trainees’ emergency response time to fire environment; heart rate changes : Indicates the real-time heart rate changes of trainees during training; baseline heart rate : Indicates the heart rate under normal conditions. The fluctuation of heart rate reflects the physiological load of the students; Anxiety index : Indicates the anxiety level of the trainees; the complexity of the fire state : Indicates the complexity of the current fire simulation environment; task completion : Indicates the degree of completion of the trainee in the task (such as whether the fire was extinguished successfully, escaped, etc.);

[0032] Trainee training status The calculation formula is as follows: F ( t ) = ∫ 0 t [ A ( t ) ⋅ e − l R ( t ) + 1 1 + e − α [ H ( t ) − H 0 ] ⋅ 1 n ( 1 + x ( t ) ) ] ⋅ [ 1 + sin ( π 2 F ( t ) ) ] d t ∫ 0 t [ C ( t ) ⋅ e − c ( t − t ) ] d t ;

[0033] in, :This item reflects the student's action performance and reaction time at time t; action accuracy directly affects the feedback result calculated by the formula, and the reaction time By exponential decay factor The intensity of feedback that affects the feedback result calculated by this formula. Specifically, the faster the response, the more timely the feedback. 1 1 + e − α [ H ( t ) − H 0 ] ⋅ 1 n ( 1 + x ( t ) ) : This is a complex interaction term that takes into account the student's physiological state; in this formula, It is an adjustment parameter used to control the intensity of the impact of heart rate changes on overall anxiety. Relative to baseline heart rate The sigmoid function is used to control the range of influence and the anxiety index The growth is then harmonized by a logarithmic function to prevent over-amplification; [ 1 + sin ( π 2 F ( t ) ) ] :This item introduces the complexity of the fire simulation environment and modulates the fire complexity through a sine function, which facilitates dynamic adjustment of the feedback intensity. For example, when the fire is more dangerous, the coach's feedback will be stronger; :This item is in the denominator, indicating the weight of the student's task completion, using an exponential decay factor Reduce the impact of early task performance and emphasize students' task performance close to the current time; The formula range is: The value range of represents the training status of the students, and then determines the intensity and frequency of the virtual AI coach's real-time feedback; a larger value means that the virtual AI coach needs to give stronger feedback and correct the students' mistakes in time; a smaller value means that the virtual AI coach needs to reduce intervention and allow students to play freely. Specifically: set thresholds based on the initial abilities and training goals of different students ,when > , it means that the trainee may have serious problems in training (such as slow reaction, inaccurate movements, and excessive anxiety). At this time, the system determines that the trainee's training status requires intervention, and the virtual AI coach will strengthen feedback or provide specific guidance; when ≤ When , it means that the trainee is in good condition, the system determines that the virtual system needs to reduce intervention and allows the trainee to continue training in the current way; among them, the threshold The calculation method is as follows: Define the following variables and parameters: : Defined as the baseline value of the initial anxiety index, indicating the normal anxiety level without excessive psychological burden; : defined as the attenuation coefficient of the anxiety index, used to control the time effect of the anxiety state; T: defined as the total training time of the trainees; sensitivity parameter : Used to control the impact of heart rate fluctuations on the threshold. A higher sensitivity parameter value means that heart rate fluctuations have a stronger impact on the threshold. Complexity influence coefficient : Used to adjust the impact of complex environments on thresholds; threshold The calculation formula is as follows: i = ∫ 0 T [ 1 1 + e − c ( H ( t ) − H 0 ) ⋅ 1 1 + β ⋅ F ( t ) + x 0 ⋅ e − d ⋅ x ( t ) ] d t ; In the above formula, the threshold The value range is between.

[0034] The learning record and suggestion module (storage and optimization layer) is responsible for recording the trainees' training data and providing personalized learning suggestions and continuous improvement guidance after the training. This module includes a learning analysis system and a personalized recommendation algorithm. The learning analysis system is used to record each training data of the trainees and generate personalized evaluation reports. The personalized recommendation algorithm is used to provide personalized improvement suggestions and next training plans based on the trainees' performance, so as to help the trainees better master their skills.

[0035] Among them, the perception module is closely integrated with the virtual environment module to capture the trainees' movements and voice inputs in real time, and drive the changes of events in the virtual environment; the virtual environment module and the interaction and feedback module work together to enhance the trainees' sense of immersion and interaction through the feedback of the virtual AI coach and the dynamic changes of the environment; the interaction and feedback module relies on the data provided by the perception module, combines the AI ​​algorithm for real-time analysis, and provides personalized improvement suggestions through the learning record and suggestion module; the learning record and suggestion module not only records the trainees' performance, but also shares data with the interaction and feedback module to generate personalized learning plans and promote the continuous progress of trainees.

[0036] The present invention also provides a training method for the above-mentioned online fire training simulation system, which specifically includes the following steps:

[0037] S1: The system builds a fire simulation environment for students through VR / AR devices;

[0038] S2: The training task is initiated, and trainees perform task operations in the virtual environment, such as finding a fire extinguisher, extinguishing the flames, and rescuing trapped people. During this period, the system provides dynamic feedback (such as changes in fire, smoke diffusion, temperature rise, etc.), and uses multi-sensory simulation (smell, heat, wind, etc.) to enhance the sense of reality.

[0039] S3: Students ask questions to the virtual AI coach through voice, such as how to use a fire extinguisher, how to deal with a certain fire, etc. The virtual AI coach analyzes the questions in real time based on the AI ​​algorithm and gives answers. At the same time, it provides timely voice guidance based on the student's performance (whether the action is correct, whether he is nervous, etc.);

[0040] S4: After the training is completed, the system automatically evaluates the trainees’ performance and generates a training report (including the trainees’ reaction speed, task completion status, error analysis, etc.), and makes suggestions for improvement based on the training data and the trainees’ performance (such as improving fire-fighting posture, increasing escape speed, etc.) and training improvement suggestions, including training modules to strengthen weaknesses.

[0041] Application examples:

[0042] Assume that a trainee is taking an online fire training simulation and a fire occurs in the simulated environment. During the training, the system continuously collects the trainee's physiological (such as heart rate), psychological (such as anxiety index) and behavioral data (such as movement accuracy and reaction time). The system uses the formula Calculate the overall performance score of the student at each moment, and the threshold To determine whether intervention is needed;

[0043] Initial state: At the beginning of the training, the trainees’ heart rate variability and anxiety index are low, and the complexity of the fire scene is relatively small; at this time, the threshold Higher; Assume The value of is also low (the trainee performs well), so the virtual AI coach will allow the trainee to complete the task autonomously without intervening too much;

[0044] State change: As the complexity of the fire scene increases, the fire spreads rapidly, and the trainees' heart rate changes and anxiety index values ​​increase; Threshold Start to descend. If Rise and exceed , indicating that the student's performance has fallen below expectations, the virtual AI coach will immediately intervene and provide specific guidance, such as prompting the student to breathe calmly or providing route optimization suggestions;

[0045] End state: In the final stage of training, the trainees may gradually recover through the feedback of the system, the trainees' heart rate changes and anxiety index values ​​begin to drop, and the fire complexity also decreases; if at this time Again below , the virtual AI coach will reduce intervention and allow trainees to complete tasks independently.

[0046] The present invention comprehensively introduces multi-sensory experiences such as smell, temperature, and wind into fire simulation training, breaking the limitation of traditional VR / AR systems that rely solely on vision and hearing, and greatly enhancing the realism and immersion of training; through real-time analysis of trainee status by AI algorithm, and combined with adaptive threshold intervention mechanism, virtual AI coaches can provide dynamic and personalized guidance in different training scenarios. This intelligent intervention based on real-time data greatly improves the effect of fire training; the present invention not only focuses on the short-term training effect of trainees, but also formulates personalized continuous learning plans through learning records and suggestion modules to promote the long-term progress of trainees. This system design that combines real-time feedback and long-term planning is extremely forward-looking.

[0047] Example 2

[0048] This is the second embodiment of the present invention. This embodiment is different from the first embodiment in that, in order to verify its beneficial effects, experimental comparison data between the present invention and the prior art are provided.

[0049] This embodiment aims to verify the dynamic threshold based on the existing technology and the technology of the present invention. And the state evaluation formula The system has the advantages of: the simulation environment includes a fire scene and 10 trainees with basic firefighting knowledge but different experiences. The system will collect the trainees' physiological data (such as heart rate), psychological data (such as anxiety index), and behavioral performance (such as reaction time and operation accuracy) in real time; through the threshold formula and state formula designed by the present invention, the system can adjust the intervention frequency and method for the trainees in real time.

[0050] Specifically, physiological data collection equipment: such as heart rate belts, which collect heart rate in real time; anxiety index measurement equipment: which evaluates the anxiety index in real time based on the trainees' performance in different complex scenarios; fire scene simulation system: the complexity changes dynamically, including fire, smoke spread, obstacles, etc.

[0051] The simulation training process is as follows:

[0052] Trainees enter the simulated training environment, and the system will Start training;

[0053] The initial training phase is a simple scenario with a minor fire situation, and the trainees’ heart rate and anxiety index Maintain at a low level;

[0054] As time goes by, the scene complexity As the fire intensified and the difficulty of escape increased, the heart rate changes and anxiety index values ​​of the trainees gradually increased;

[0055] The system is based on the formula Calculate the intervention threshold and use the state formula Assess the current status of the learners;

[0056] when > When the system thinks that the trainee is under excessive stress, it triggers the virtual AI coach to intervene and give escape instructions or psychological adjustment tips;

[0057] The control group used existing technology to decide whether to intervene based on static assessment criteria, without considering real-time changes in heart rate, anxiety index, etc.;

[0058] After the training, the reaction time, task completion rate, number of interventions and other data of each trainee in the training are recorded to evaluate the effectiveness of the system;

[0059] The following table records the performance data of 10 trainees under different systems, including key indicators such as heart rate fluctuation, anxiety index, fire complexity, number of interventions by virtual AI coaches, and task completion rate. The specific situation is shown in the following table:

[0060]

[0061] It can be clearly seen from the table data that the system of the present invention has significant improvements in multiple key indicators compared with the prior art, specifically:

[0062] Heart rate control: Under the system of the present invention, the average heart rate of the students is generally low; for example, the average heart rate of student No. 1 dropped from 125 beats / minute to 110 beats / minute, and the heart rates of other students also showed similar declines; this shows that the system effectively reduces the physiological stress of students by dynamically adjusting the intervention threshold in real time.

[0063] Anxiety index: The system of the present invention can better control the anxiety level of students; for example, the anxiety index of student No. 2 dropped from 0.50 to 0.35, and the anxiety indexes of other students also decreased accordingly; this shows that the dynamic threshold can intervene more promptly to prevent students from being in a state of high anxiety for a long time.

[0064] Number of virtual AI coach interventions: The system of the present invention dynamically adjusts according to the physiological and psychological states of the trainees, reducing the number of unnecessary interventions; for example, the number of interventions for trainee No. 1 was reduced from 5 times to 3 times, a reduction of 40%; this shows that the system avoids excessive intervention while providing effective guidance, thereby improving the training experience.

[0065] Task completion rate: The trainees using the system of the present invention generally showed a higher task completion rate; the task completion rate of trainee No. 1 increased from 80% to 90%, and that of trainee No. 2 increased from 75% to 85%; this proves that the system can improve the trainees' operating efficiency and task completion through appropriate intervention.

[0066] Reaction time: After using the system of the present invention, the reaction time of the students is generally shortened; for example, the average reaction time of student No. 3 is shortened from 5.4 seconds to 4.1 seconds, which improves the decision-making speed of the students in a high-pressure environment, indicating that the system can help students make correct action decisions in a shorter time.

[0067] In summary, through the above data analysis, compared with the prior art, the present invention introduces a dynamic threshold And the state evaluation formula , can adjust the intervention frequency and intensity in real time according to the students' physiological and psychological state, thereby effectively reducing the students' physiological stress, lowering the anxiety index, and improving the task completion rate and reaction speed; this adaptive intervention method avoids excessive intervention and improves the training effect and the overall experience of the students.

[0068] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An online fire training simulation system, characterized by: It includes a perception module, a virtual environment module, an interaction and feedback module, and a learning record and suggestion module; among them, the perception module is closely integrated with the virtual environment module to capture the trainees' movements and voice input in real time, and drive the changes of events in the virtual environment; the virtual environment module and the interaction and feedback module work together to enhance the trainees' sense of immersion and interaction through the feedback of the virtual AI coach and the dynamic changes of the environment; the interaction and feedback module relies on the data provided by the perception module, combines the AI ​​algorithm for real-time analysis, and provides personalized improvement suggestions through the learning record and suggestion module; the learning record and suggestion module not only records the trainees' performance, but also shares data with the interaction and feedback module to generate personalized learning plans and promote the continuous progress of the trainees.

2. The online fire training simulation system according to claim 1, characterized in that: The perception module is responsible for collecting the trainees' movements, voices and physiological data, which serves as the basis for the system to judge the trainees' behavior and training effects, and is used for real-time feedback and interaction in the virtual environment. The perception module includes motion capture equipment, heart rate monitoring equipment and microphones.

3. The online fire training simulation system according to claim 2, characterized in that: The virtual environment module is used to provide a highly realistic fire simulation environment and supports multi-sensory input such as vision, hearing, touch, and smell; the virtual environment module includes VR or AR headsets, environmental sensing equipment, high-fidelity audio systems, as well as 3D environment simulation engines, physical simulation engines, and multi-sensory stimulation control systems.

4. The online fire training simulation system according to claim 3, characterized in that: The interaction and feedback module integrates a virtual AI coach and an AI analysis system to dynamically analyze the trainees' training process based on their real-time behavior and physiological data. Trainees ask questions through voice, and the virtual AI coach gives voice or text guidance based on the situation, providing real-time interaction and intelligent feedback.

5. The online fire training simulation system according to claim 4, characterized in that: The virtual AI coaching system collects students' physiological and behavioral data, determines the students' training status based on the AI ​​analysis system, and then adjusts its guidance intensity in real time. The specific steps are as follows: Define the following variables and parameters: Student training status : Indicates the training status of the trainee at time t; movement accuracy : Indicates the accuracy of the student's action at time t; reaction time : Indicates the trainees’ emergency response time to fire environment; heart rate changes : Indicates the real-time heart rate changes of trainees during training; baseline heart rate : Indicates the heart rate of the trainees in a normal state. The fluctuation of heart rate reflects the physiological load of the trainees. Anxiety Index : Indicates the anxiety level of the trainees; the complexity of the fire state : Indicates the complexity of the current fire simulation environment; task completion : Indicates the degree of completion of the student in the task; the student's training status The calculation formula is as follows: ;in, :This item reflects the student's action performance and reaction time at time t; action accuracy directly affects the feedback result calculated by the formula, and the reaction time By exponential decay factor The intensity of feedback that affects the feedback result calculated by this formula. Specifically, the faster the response, the more timely the feedback. : This is a complex interaction term that takes into account the physiological state of the students; In this formula, It is an adjustment parameter used to control the intensity of the impact of heart rate changes on overall anxiety. Relative to baseline heart rate The sigmoid function is used to control the range of influence and the anxiety index The growth is harmonized by a logarithmic function to prevent over-amplification; :This item introduces the complexity of the fire simulation environment and modulates the fire complexity through a sine function, which facilitates dynamic adjustment of the feedback intensity. For example, when the fire is more dangerous, the coach's feedback will be stronger; :This item is in the denominator, indicating the weight of the student's task completion, using an exponential decay factor Reduce the impact of early task performance and emphasize students' task performance close to the current time; The formula range is: The value range of represents the training status of the students, which in turn determines the intensity and frequency of the virtual AI coach's real-time feedback; a larger value means that the virtual AI coach needs to give stronger feedback and correct the students' mistakes in time; a smaller value means that the virtual AI coach needs to reduce intervention and allow students to play freely.

6. The online fire training simulation system according to claim 5, characterized in that: The virtual AI coaching system sets thresholds based on the initial abilities and training goals of different trainees ,when > , it means that the trainee may have serious problems in training. At this time, the system determines that the trainee's training status needs intervention, and the virtual AI coach will strengthen feedback or provide specific guidance; ≤ , it indicates that the trainee is in good condition, and the system determines that the virtual system needs to reduce intervention, allowing the trainee to continue training in the current way.

7. The online fire training simulation system according to claim 6, characterized in that: Threshold The calculation method is as follows: Define the following variables and parameters: : Defined as the baseline value of the initial anxiety index, indicating the normal anxiety level without excessive psychological burden; : defined as the attenuation coefficient of the anxiety index, used to control the temporal influence of the anxiety state; T: defined as the total duration of the trainees’ training; : Defined as a sensitivity parameter, which is used to control the degree of influence of heart rate fluctuation on the threshold. A higher sensitivity parameter value means that heart rate fluctuation has a stronger influence on the threshold. : Defined as the complexity influence coefficient, it is used to adjust the impact of complex environment on the threshold; then, the threshold The calculation formula is as follows: ; In the above formula, the threshold The value range is between.

8. The online fire training simulation system according to claim 7, characterized in that: The learning record and suggestion module is responsible for recording students' training data, providing personalized learning suggestions and continuous improvement guidance after training, and helping students better master skills.

9. The training method of the online fire training simulation system as claimed in claim 8, characterized in that: The system includes the following specific steps: S1: The system builds a fire simulation environment for trainees through VR or AR devices; S2: The training task is started, and trainees perform task operations in the virtual environment. At the same time, the system provides dynamic feedback and uses multi-sensory simulation to enhance the sense of reality; S3: Trainees ask questions to the virtual AI coach through voice, and the virtual AI coach analyzes the questions and gives answers in real time based on the AI ​​algorithm, and provides timely voice guidance based on the trainees' performance; S4: After the training is completed, the system automatically evaluates the trainee’s performance, generates a training report, and provides improvement suggestions and training improvement recommendations based on the training data and the trainee’s performance.

10. The training method of the online fire training simulation system as claimed in claim 9, characterized in that: At the beginning of the training, the trainees performed well, their heart rate changes and anxiety index values ​​were low, and the complexity of the fire scene was relatively small. At this time, the threshold Higher, assuming The value is also low. The virtual AI coach allows students to complete tasks independently without excessive intervention. As the complexity of the fire scene increases, the fire spreads rapidly, and the heart rate changes and anxiety index values ​​of the students increase. The threshold Start to decline, if at this time Rising and exceeding the threshold , indicating that the trainee's performance has fallen below expectations, the virtual AI coach will immediately intervene and provide specific guidance, such as prompting the trainee to breathe calmly or providing route optimization suggestions; in the final stage of training, the trainee may gradually recover through the feedback of the system, the trainee's heart rate change and anxiety index values ​​begin to decline, and the fire complexity also decreases. If at this time Below the threshold again , the virtual AI coach will reduce intervention and allow trainees to complete tasks independently.

Images