Fire rescue stress scene VR simulation system and method
By using a VR simulation system for fire and rescue emergency scenarios, and leveraging virtual experience decision-making and real-time data optimization, the problem of insufficient scenario realism in fire and rescue training has been solved, achieving efficient and safe emergency training results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANGHAI FIRE RES INST OF MEM
- Filing Date
- 2026-03-23
- Publication Date
- 2026-06-19
AI Technical Summary
The existing fire and rescue emergency training scenarios lack realism, and the training process is rigid and uncontrollable, affecting firefighters' daily lives and work efficiency.
This invention provides a VR simulation system for fire rescue emergency scenarios. It enables personalized decision-making through a virtual experience decision-making module, and combines a virtual interaction module and an experience scheme optimization module. It utilizes a head-mounted VR display device to construct an immersive virtual environment, obtain user feedback and monitoring data in real time, and dynamically optimize training schemes.
Precise and efficient stress training was achieved in an absolutely safe virtual environment, which improved the relevance and safety of the training, avoided the risk of overstimulation, and enhanced the emergency response capabilities of firefighters.
Smart Images
Figure CN122244392A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of VR technology, and in particular to a VR simulation system and method for fire rescue emergency scenarios. Background Technology
[0002] With the rapid development of the economy and society and the increasing activity of production, fire and rescue tasks are becoming increasingly heavy. Firefighters bear the heavy responsibility of fire prevention, firefighting, and handling and rescuing various types of disasters and accidents. They are in a high-pressure, high-risk working environment for a long time, which can easily induce trauma and stress-related psychological disorders, bringing lasting negative impacts on their daily lives, interpersonal relationships, and work duties. At present, there are technical problems in related technologies, such as insufficient realism of fire and rescue stress training scenarios and rigid and uncontrollable training processes. Summary of the Invention
[0003] This application provides a VR simulation system and method for fire rescue emergency scenarios, solving the technical problems of insufficient scenario realism and rigid, uncontrollable training processes in existing fire rescue emergency training. It achieves the technical effect of precise and efficient emergency training for fire rescue personnel in an absolutely safe virtual environment through a dynamically adjustable immersive experience.
[0004] This application provides a VR simulation system for fire rescue emergency scenarios. The system includes: a virtual experience decision-making module, used to make personalized virtual experience decisions based on the user's basic information and emergency scenario experience needs, and obtain an emergency scenario experience plan; a virtual interaction module, used to guide the user into a virtual interaction scenario through a head-mounted VR display device based on the emergency scenario experience plan, and simultaneously obtain user feedback data and user monitoring data; an experience plan optimization module, used to optimize the emergency scenario experience plan based on the user feedback data and the user monitoring data, and obtain a virtual experience optimization plan; and an experience optimization module, used to guide the user to conduct an immersive optimization experience based on the virtual experience optimization plan.
[0005] Preferably, the virtual experience decision-making module includes: a vector construction module, used to clean the data based on the user's basic information and the stress scenario experience requirements, and construct the experiencer's feature vector; a history processing module, used to process the data based on the historical virtual experience decision record set, and obtain a historical experience decision dataset; a network construction module, used to train a virtual experience decision network based on the historical experience decision dataset; and an experience scheme acquisition module, used to input the experiencer's feature vector into the virtual experience decision network to obtain the stress scenario experience scheme.
[0006] Preferably, the network construction module includes: a data partitioning module for partitioning the historical experience decision dataset to obtain an experience decision training set and an experience decision test set; a supervised training module for supervising the training of the BP neural network based on the experience decision training set to obtain an experience decision base network; a network testing module for testing the experience decision base network based on the experience decision test set to obtain experience decision accuracy; and a reinforcement learning module for performing reinforcement learning on the experience decision base network based on the experience decision accuracy and a predetermined experience decision accuracy to generate the virtual experience decision network.
[0007] Preferably, the experience optimization module includes: an experience scoring module, used to score the VR experience based on the user feedback data and the user monitoring data, and obtain a current experience score; a scoring comparison module, used to compare the current experience score with a predetermined experience score, and obtain an experience judgment tag; and an adjustment module, used to adaptively adjust the stress scenario experience scheme based on the experience judgment tag, and generate the virtual experience optimization scheme.
[0008] Preferably, the optimized experience module includes: a scheme parsing module, used to parse the virtual experience optimization scheme and obtain structured adjustment instructions; and a closed-loop optimization module, used to perform VR simulation dynamic closed-loop optimization according to the structured adjustment instructions.
[0009] Preferably, the system includes a communication module for interactive communication between the virtual experience decision module, the virtual interaction module, the experience scheme optimization module, and the optimized experience module.
[0010] This application also provides a VR simulation method for fire rescue emergency scenarios. The method is applied to a VR simulation system for fire rescue emergency scenarios. The method includes: making personalized virtual experience decisions based on the user's basic information and emergency scenario experience needs to obtain an emergency scenario experience plan; guiding the user into a virtual interactive scenario through a head-mounted VR display device based on the emergency scenario experience plan, while simultaneously acquiring user feedback data and user monitoring data; optimizing the emergency scenario experience plan based on the user feedback data and user monitoring data to obtain a virtual experience optimization plan; and guiding the user to undergo an immersive optimized experience based on the virtual experience optimization plan.
[0011] This application proposes a VR simulation system and method for fire rescue emergency scenarios. The system utilizes a virtual experience decision-making module to personalize the virtual experience based on the user's basic information and emergency scenario experience needs, obtaining an emergency scenario experience plan. A virtual interaction module guides the user into a virtual interaction scenario using a head-mounted VR display device, simultaneously acquiring user feedback data and user monitoring data. An experience plan optimization module optimizes the emergency scenario experience plan based on the user feedback and monitoring data, obtaining an optimized virtual experience plan. An optimized experience module guides the user through an immersive optimized experience based on the optimized virtual experience plan. This solves the technical problems of insufficient scenario realism and rigid, uncontrollable training processes in existing fire rescue emergency training. It achieves the technical effect of precise and efficient emergency training for fire rescue personnel in an absolutely safe virtual environment through a dynamically adjustable immersive experience. Attached Figure Description
[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.
[0013] Figure 1 This is a schematic diagram of the structure of a VR simulation system for fire rescue emergency scenarios provided in an embodiment of this application.
[0014] Figure 2 This is a flowchart illustrating a VR simulation method for fire rescue emergency scenarios provided in an embodiment of this application.
[0015] Explanation of reference numerals in the attached diagram: Virtual experience decision-making module 10, virtual interaction module 20, experience solution optimization module 30, optimized experience module 40, and communication module 50. Detailed Implementation
[0016] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.
[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0018] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.
[0019] Example 1: This application provides a VR simulation system for fire rescue emergency scenarios, such as... Figure 1 As shown, the system includes: The virtual experience decision-making module 10 is used to make personalized virtual experience decisions based on the user's basic information and stress scenario experience needs, and to obtain stress scenario experience solutions. The virtual interaction module 20 is used to guide the user into a virtual interaction scene through a head-mounted VR display device based on the stress scenario experience scheme, and simultaneously acquire user feedback data and user monitoring data. The experience optimization module 30 is used to optimize the stress scenario experience plan based on the user feedback data and the user monitoring data, and obtain a virtual experience optimization plan. The optimized experience module 40 is used to guide the user to have an immersive optimized experience according to the virtual experience optimization scheme.
[0020] The virtual experience decision-making module includes: a vector construction module, used to clean data based on the user's basic information and the stress scenario experience requirements to construct an experiencer feature vector; a history processing module, used to process data based on a historical virtual experience decision record set to obtain a historical experience decision dataset; a network construction module, used to train a virtual experience decision network based on the historical experience decision dataset; and an experience scheme acquisition module, used to input the experiencer feature vector into the virtual experience decision network to obtain the stress scenario experience scheme.
[0021] The network construction module includes: a data partitioning module for partitioning the historical experience decision dataset to obtain an experience decision training set and an experience decision test set; a supervised training module for supervising the training of the BP neural network based on the experience decision training set to obtain an experience decision base network; a network testing module for testing the experience decision base network based on the experience decision test set to obtain the experience decision accuracy; and a reinforcement learning module for performing reinforcement learning on the experience decision base network based on the experience decision accuracy and a predetermined experience decision accuracy to generate the virtual experience decision network.
[0022] The experience optimization module includes: an experience scoring module, used to score the VR experience based on the user feedback data and the user monitoring data, and obtain the current experience score; a scoring comparison module, used to compare the current experience score with a predetermined experience score, and obtain an experience judgment tag; and an adjustment module, used to adaptively adjust the stress scenario experience scheme based on the experience judgment tag, and generate the virtual experience optimization scheme.
[0023] The optimized experience module includes: a scheme parsing module, used to parse the virtual experience optimization scheme and obtain structured adjustment instructions; and a closed-loop optimization module, used to perform VR simulation dynamic closed-loop optimization according to the structured adjustment instructions.
[0024] The system includes a communication module 50, which facilitates communication between the virtual experience decision module 10, the virtual interaction module 20, the experience scheme optimization module 30, and the optimized experience module 40. The communication module 50 enables remote control, management, and updates of the system and scenes. The communication module 50 is based on a wireless local area network (WLAN). Specifically, after the control terminal (tablet computer) and the VR experience terminal (headset) are connected to the same wireless network, the control terminal discovers the device via UDP broadcast and establishes a persistent connection with the selected VR experience terminal via TCP. The control terminal sends structured control commands to the VR experience terminal, which parses the commands and executes the corresponding operations, achieving precise remote control. The control terminal has a content management module. When scene video or sound effect files need to be updated, the operator imports the new files into the control terminal's local storage. Through the device connection interface, the target VR device can be selected, and the new files can be transferred to the specified directory on the VR device via a TCP connection. Simultaneously, the system supports firmware updates: update packages are pushed to the VR device, and the device automatically installs the update when idle.
[0025] The VR simulation system for fire rescue emergency scenarios provided in this embodiment of the invention can execute the VR simulation method for fire rescue emergency scenarios provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
[0026] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.
[0027] Example 2: This application provides a VR simulation method for fire rescue emergency scenarios. The method is applied to a VR simulation system for fire rescue emergency scenarios, such as... Figure 2 As shown, the method includes: Step A100: Based on the user's basic information and stress scenario experience requirements, make personalized virtual experience decisions and obtain stress scenario experience solutions. In one possible implementation, step A100 further includes steps A110 to A140. Step A110: Perform data cleaning based on the user's basic information and stress scenario experience requirements to construct a user feature vector. User basic information includes the user's age, gender, years of service, rank and position, as well as data such as the type, number, and completion rate of previously completed VR training scenarios. Stress scenario experience requirements include information on the type of stress scenario corresponding to the user, such as a car accident scene, a fire scene, a rescue scene, or other scenes. Data cleaning includes handling missing values, format standardization, and outlier handling for the user's basic information and stress scenario experience requirements. After data cleaning, combine the processed structured data into a multi-dimensional feature vector. This multi-dimensional feature vector is the user feature vector, providing data support for subsequent personalized virtual experience decisions for the user. Missing value handling includes filling missing fields with default values (e.g., setting the missing "Years of Service" to 0), the median, or by inference based on other fields. Format standardization includes converting text information (e.g., job title "firefighter") into predefined enumeration values (e.g., code 1) to ensure all data is either numeric or categorical. Outlier handling includes identifying and correcting obviously erroneous data, such as "Age" being 200 years old.
[0028] Step A120: Organize the data based on the historical virtual experience decision record set to obtain the historical experience decision dataset. Each VR simulation of a fire rescue emergency scenario generates a complete record. The historical virtual experience decision record set includes multiple historical virtual experience decision records. Each historical virtual experience decision record includes the historical virtual experience time, historical user basic information, historical emergency scenario experience requirements, and historical emergency scenario experience solutions. The system periodically runs a data organization task to process the historical virtual experience decision record set into a well-organized historical experience decision dataset. This transforms the "operation logs" generated during system operation into a "historical experience decision dataset" with analytical value, providing a solid data foundation for subsequent training of the virtual experience decision network. Data organization includes data extraction and cleaning of the historical virtual experience decision record set, handling missing values and outliers to ensure data quality.
[0029] Step A130: Train the virtual experience decision network based on the historical experience decision dataset. In one possible implementation, step A130 further includes steps A131 to A134. Step A131: Divide the historical experience decision dataset to obtain an experience decision training set and an experience decision test set; Step A132: Perform supervised training on the BP neural network based on the experience decision training set to obtain the experience decision base network; Step A133: Test the experience decision base network based on the experience decision test set to obtain the experience decision accuracy; Step A134: Based on the experience decision accuracy, perform reinforcement learning on the experience decision base network according to a predetermined experience decision accuracy to generate the virtual experience decision network.
[0030] The historical experience decision-making dataset is divided according to a predetermined ratio (e.g., 7:3). 70% of the data in the historical experience decision-making dataset is used as the experience decision-making training set for subsequent model training. The remaining 30% of the data is used as the experience decision-making test set for testing and evaluating the trained model. This division ensures that both the training and test sets are representative and can cover various possible experience decision-making scenarios and data characteristics.
[0031] This application selects a backpropagation (BP) neural network as the basic model structure. BP neural networks possess excellent nonlinear mapping and self-learning capabilities, enabling them to adapt to complex experience-decision relationship modeling. The experience-decision training set is input into the BP neural network for supervised training. During training, appropriate training parameters, such as the learning rate and the number of iterations, are set. The learning rate determines the magnitude of weight adjustment in each iteration; an excessively large learning rate may prevent convergence, while an excessively small learning rate will slow down the training process. The number of iterations controls the number of training rounds and needs to be set appropriately based on the size and complexity of the dataset. By continuously adjusting the weights and biases of the BP neural network, the error between the network output and the expected result is gradually reduced, thus obtaining the basic experience-decision network. This basic experience-decision network already possesses a certain ability to make experience decisions based on the input data.
[0032] The trained experience decision-making foundation network is tested using an experience decision test set. Data from the experience decision test set is input into the experience decision-making foundation network to obtain the network output. The network output is compared and analyzed with the true labels in the experience decision test set, and appropriate evaluation metrics are used to measure the performance of the experience decision-making foundation network, thereby obtaining the experience decision accuracy. Experience decision accuracy includes the output accuracy of the experience decision-making foundation network. The predetermined experience decision accuracy is a threshold set according to actual application needs, representing the expected performance level of the model. If the experience decision accuracy does not reach the predetermined experience decision accuracy, it indicates that the experience decision-making foundation network has room for improvement. At this time, reinforcement learning algorithms are used to optimize the experience decision-making foundation network. Reinforcement learning allows the model to learn the optimal strategy through interaction with the environment. In the experience decision scenario, the accuracy of experience decisions can be used as a reward signal to guide the network to continuously adjust its parameters to improve decision performance. After the reinforcement learning process, when the experience decision accuracy reaches or exceeds the predetermined experience decision accuracy, the model at this point is the virtual experience decision network. The virtual experience decision network can make adaptive virtual experience decisions based on user characteristics, improving the accuracy and adaptability of fire rescue stress training.
[0033] Step A140: Input the experiencer feature vector into the virtual experience decision network to obtain the stress scenario experience scheme. Specifically, the preprocessed experiencer feature vector (including dimensions such as age, length of service, job type, number of historical stress exposures, and scenario preferences) is input into the trained virtual experience decision network. The virtual experience decision network adopts a three-layer BP neural network structure: the input layer dimension is consistent with the feature vector, the hidden layer contains two layers of fully connected neurons (16 neurons per layer, using the Sigmoid activation function), and the output layer has one neuron (Sigmoid activation, outputting a probability value in the 0-1 range).
[0034] After the input user feature vector is forward-propagated through the network, the output layer generates scene recommendation probabilities (e.g., 0.85 for a fire scene, 0.15 for a geological disaster), selecting the scene type corresponding to the highest probability value as the base scene type. Simultaneously, the network's intermediate layer parameters dynamically adjust the immersion mode based on the user's historical exposure frequency and job type: a "lights-off mode" (enhancing endurance training) is recommended for high exposure frequency or combatant positions, while a "lights-on mode" (reducing psychological impact) is recommended for low exposure frequency or driver positions. Furthermore, the virtual experience decision network matches video content from the VR device's storage directory based on the base scene type (e.g., a preference for "fire scene" calls the fire scene experience content directory, a preference for "car accident scene" calls the car accident scene experience content directory, a preference for "rescue scene" calls the rescue scene experience content directory, and a preference for "other scenes" calls the other scene experience content directory), and generates corresponding sound effects (e.g., explosion sounds, alarm sounds) and vibration parameters. The final output stress scene experience scheme includes: scene type (e.g., fire scene), immersion mode (lights on / off), video content path, and interaction parameters (sound effect list, vibration duration), among other data. The virtual experience decision network generates differentiated solutions based on the characteristics of the experiencer, avoiding a "one-size-fits-all" experience, improving the pertinence of fire rescue stress training, and realizing precise and dynamic VR simulation of fire rescue stress scenarios.
[0035] Step A200: Based on the aforementioned stress scenario experience scheme, the user is guided into a virtual interactive scene via a head-mounted VR display device, simultaneously acquiring user feedback data and user monitoring data. According to the stress scenario experience scheme generated in step A140, the system automatically loads corresponding virtual scene resources (such as 3D models of fire scenes, dynamic flame effects, environmental sound effect libraries, etc.) and presents them to the user's field of vision through the head-mounted VR display device. The virtual interactive scene includes immersive visuals, immersive sound effects, force feedback handles, and control interaction. The user enters an immersive video playback scene through parallax display. The scene features a 300-inch virtual screen to recreate various scenes encountered in fire rescue operations, combined with surround sound effects to create a stunning immersive experience. Furthermore, different levels of immersion for the same video content can be adjusted through light-on and light-off modes.
[0036] "Parallelism display" refers to stereoscopic vision achieved through a hardware optical solution based on head-mounted VR displays (such as the PICO Neo 4). Specifically, the head-mounted VR display has dual built-in screens providing slightly horizontally offset images to the left and right eyes respectively. Combined with the headset's optical lens system, the images for the left and right eyes are merged on the retina, creating a stereoscopic depth of field. Simultaneously, a binocular camera in the Unity engine simulates the left and right eye perspectives. The two cameras are horizontally offset according to a preset interpupillary distance, rendering images with slightly different perspectives. The rendered images are then corrected for lens distortion before being output to the headset screen. For 360° panoramic videos or 180° 3D videos, split-screen rendering is also used during playback to ensure that the left and right eyes receive the corresponding parallax images, thus creating a sense of stereoscopic depth and immersion.
[0037] The "Lights On" and "Lights Off" modes are achieved by dynamically changing the global illumination parameters in the VR virtual environment. Specifically, in Unity, a virtual cinema lobby is created. This environment includes a large virtual screen playing video and a controllable ambient light source (such as Directional Light or Ambient Light). In Lights On mode, the ambient light is set to a relatively bright intensity (e.g., Intensity=1.0), making details such as walls and floors clearly visible in the virtual environment and reducing the user's sense of closure. In Lights Off mode, the ambient light intensity is adjusted to a very low level (e.g., Intensity=0.1), almost completely dark, with only a faint glow on the screen, allowing attention to be fully focused on the video content, enhancing immersion and tension. The control panel has an "On / Off" button. When the button is clicked, the control panel sends a mode switching command (containing mode parameters) to the currently connected VR headset. Upon receiving the command, the VR device uses a script to call Unity's Lighting API to dynamically switch the preset lighting configuration. The lighting changes take effect immediately in the next rendering frame, allowing users to instantly perceive changes in ambient brightness and darkness, thereby achieving the goal of adjusting the immersive level of the same video content.
[0038] When a user enters the virtual interactive scene, the system collects user feedback data and user monitoring data in real time through a multimodal interaction interface and wearable physiological monitoring devices, providing high-granular data support for subsequent virtual experience optimization. User feedback data includes the user's subjective evaluation data and behavioral operation data. Subjective evaluation data includes voice comments recorded through the built-in microphone of the head-mounted VR display device, as well as virtual questionnaire data. Behavioral operation data includes the user's movement trajectory in the virtual scene (e.g., movement speed, stopping area), interaction frequency (e.g., number of times a fire extinguisher is used, number of times a stretcher is carried), and operation latency (e.g., the time difference between issuing a command and executing an action). User monitoring data includes the user's physiological signal data (e.g., heart rate, respiratory rate, blood oxygen saturation), eye-tracking data, head movement data, and gesture interaction data (hand movement trajectories recorded through the controller, such as grasping, pressing pressure, etc.).
[0039] Step A300: Optimize the stress scenario experience plan based on the user feedback data and the user monitoring data to obtain a virtual experience optimization plan. In one possible implementation, step A300 further includes steps A310 to A330. Step A310: Calculate a VR experience score based on the user feedback data and the user monitoring data to obtain a current experience score. Step A320: Compare the current experience score with a predetermined experience score to obtain an experience judgment label. Step A330: Adaptively adjust the stress scenario experience plan based on the experience judgment label to generate the virtual experience optimization plan.
[0040] The system anonymizes user feedback and monitoring data before distributing it to multiple VR experience evaluation experts (accessible via pre-set expert accounts in the system backend). Each expert independently scores the user based on a standardized rating scale (e.g., immersion, stress intensity, comfort, etc.), generating multiple experience score data (e.g., Expert 1 score 85, Expert 2 score 78, etc.). The average of these multiple score data is calculated to obtain the current experience score. Then, by comparing the current score with a predetermined score, an experience assessment label is generated. The predetermined score can be adaptively set. Experience assessment labels include: Excellent Experience, Improveable Experience, and High-Risk Experience. If the current experience score ≥ Predicted Experience Score + 5 points, the label is Excellent Experience. If the Predicted Experience Score - 5 points ≤ Predicted Experience Score < Predicted Experience Score + 5 points, the label is Improveable Experience. If the Predicted Experience Score < Predicted Experience Score - 5 points, the label is High-Risk Experience.
[0041] The system dynamically adjusts the stress scenario experience based on experience assessment tags, achieving personalized dynamic optimization and intelligent safety assurance for the virtual experience of stress scenarios, thereby effectively improving the stress training effect for fire and rescue personnel. The virtual experience optimization scheme includes immersion level adjustment parameters corresponding to the experience assessment tags, dynamic sound effect insertion adjustment parameters, force feedback adjustment parameters, and video content switching parameters. Specifically, when the experience assessment tag is "high-quality experience," the immersion level of subsequent scenes is automatically increased, such as switching to "lights off mode" to enhance tension, dynamically inserting more sound effects (such as explosion sound effects and scream sound effects), and extending the force feedback vibration time. When the experience assessment tag is "expandable experience," if the current experience score is low due to insufficient immersion, the system automatically switches to "lights off mode" and increases the video volume. If the stress is excessive, "lights on mode" is activated to reduce the sense of environmental oppression and decrease the frequency of sound effect triggers. When the experience assessment tag is "high-risk experience," the system automatically performs proactive intervention, such as immediately pausing video playback (activating the "playback control - pause button") and switching to a low-intensity scene (such as switching from "fire scene" to "rescue scene"). At the same time, an alarm is sent to the control terminal to prompt the administrator to intervene.
[0042] Step A400: Guide the user to perform an immersive optimization experience according to the virtual experience optimization scheme. In one possible implementation, step A400 further includes steps A410 to A420. Step A410: Parse the virtual experience optimization scheme to obtain structured adjustment instructions; Step A420: Perform VR simulation dynamic closed-loop optimization according to the structured adjustment instructions. The system first parses the virtual experience optimization scheme into a series of structured adjustment instructions, including lighting control instructions, content switching instructions, sound effect insertion instructions, and force feedback instructions. Then, through the established TCP connection, the structured adjustment instructions are accurately sent to the head-mounted VR display device. The head-mounted VR display device receives and executes the structured adjustment instructions, dynamically adjusting the lighting, audio and video content, and interactive feedback of the virtual environment, thereby achieving real-time, automated closed-loop optimization of the user's immersive experience and enabling precise and efficient stress training for fire and rescue personnel.
[0043] In summary, the VR simulation method for fire rescue emergency scenarios provided in this application has the following technical effects: 1. Based on the user's basic information and specific scenario needs, intelligent decision-making is made to generate customized stress scenario experience solutions, which effectively improves the pertinence and effectiveness of training and avoids the risks of insufficient effect or overstimulation that may be caused by "one-size-fits-all" training.
[0044] 2. Utilizing head-mounted VR displays to construct highly realistic virtual environments, combined with technologies such as parallax display, surround sound, and force feedback controllers (vibration effects), a strong sense of immersion is created. User feedback and monitoring data are recorded simultaneously to improve the comprehensiveness and targeted nature of the immersive experience.
[0045] 3. By analyzing user feedback data and user monitoring data, the system automatically evaluates the current experience (e.g., determines it as a high-quality experience / expandable experience / high-risk experience) and dynamically adjusts the virtual scene parameters accordingly. This ensures that the virtual interactive scene always adapts to the user's real-time state, forming a closed loop of "evaluation-optimization-execution," which significantly improves the safety and adaptability of fire rescue emergency training.
[0046] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.
Claims
1. A VR simulation system for fire rescue emergency scenarios, characterized in that, The system includes: The virtual experience decision-making module is used to make personalized virtual experience decisions based on the user's basic information and stress scenario experience needs, and to obtain stress scenario experience solutions. The virtual interaction module is used to guide the user into a virtual interaction scenario through a head-mounted VR display device based on the stress scenario experience scheme, and simultaneously acquire user feedback data and user monitoring data. The experience optimization module is used to optimize the stress scenario experience plan based on the user feedback data and the user monitoring data, and obtain a virtual experience optimization plan. The optimized experience module is used to guide the user to have an immersive and optimized experience based on the virtual experience optimization scheme.
2. The system as described in claim 1, characterized in that, The virtual experience decision-making module includes: The vector construction module is used to clean the data based on the user's basic information and the stress scenario experience requirements, and to construct the experiencer's feature vector. The historical data processing module is used to process data based on the historical virtual experience decision record set and obtain the historical experience decision dataset. The network construction module is used to train the virtual experience decision network based on the historical experience decision dataset; The experience scheme acquisition module is used to input the experiencer's feature vector into the virtual experience decision network to obtain the stress scenario experience scheme.
3. The system as described in claim 2, characterized in that, The network construction module includes: The data partitioning module is used to partition the historical experience decision dataset and obtain the experience decision training set and the experience decision test set. The supervised training module is used to supervise the training of the BP neural network based on the experience decision training set to obtain the experience decision basic network. The network testing module is used to test the experience decision-making foundation network according to the experience decision test set and obtain the experience decision accuracy. The reinforcement learning module is used to perform reinforcement learning on the experience decision base network based on the experience decision accuracy and according to a predetermined experience decision accuracy, to generate the virtual experience decision network.
4. The system as described in claim 1, characterized in that, The experience optimization module includes: The experience rating module is used to rate the VR experience based on the user feedback data and the user monitoring data, and to obtain the current experience rating. The rating comparison module is used to compare the current experience rating with the predetermined experience rating to obtain the experience judgment tag; The adjustment module is used to adaptively adjust the stress scenario experience scheme based on the experience judgment tags, and generate the virtual experience optimization scheme.
5. The system as described in claim 1, characterized in that, The optimized experience module includes: The scheme parsing module is used to parse the virtual experience optimization scheme and obtain structured adjustment instructions; The closed-loop optimization module is used to perform VR simulation dynamic closed-loop optimization according to the structured adjustment instructions.
6. The system as described in claim 1, characterized in that, The system includes a communication module, which is used for interactive communication between the virtual experience decision module, the virtual interaction module, the experience scheme optimization module, and the optimized experience module.
7. A VR simulation method for fire rescue emergency scenarios, characterized in that, The method is applied to the system according to any one of claims 1 to 6, and the method comprises: Personalized virtual experience decisions are made based on the user's basic information and stress scenario experience needs to obtain stress scenario experience solutions; Based on the stress scenario experience scheme, the user is guided into a virtual interactive scenario through a head-mounted VR display device, and user feedback data and user monitoring data are acquired simultaneously. Based on the user feedback data and the user monitoring data, the stress scenario experience solution is optimized to obtain a virtual experience optimization solution. The virtual experience optimization scheme guides the user to an immersive and optimized experience.