A human body aging virtual-real combination experience system based on VR and AI technology

Abstract

The application discloses a kind of human aging virtual-real combination experience systems based on VR and AI technology, it is related to human aging virtual-real combination experience field, including aging experience wearing device, aging experience VR visual device, AI man-machine interaction module and central control module.The system of the present application creates realistic virtual environment using VR technology, combined with a variety of aging effect simulation, so that users immerse in aging experience wholeheartedly, can obtain more real, profound feeling.Real-time perception of user behavior and emotion, real-time adjustment of virtual environment and aging effect, provide more intelligent, flexible interactive experience, enhance the sense of reality and interesting of experience.Through the analysis of user personal information and real-time behavior data by AI technology, customized aging experience is provided for each user, and the aging situation that different individuals may face is more accurately reflected.

Description

Technical Field

[0001] This invention relates to the field of human aging experience technology, and in particular to a virtual-real combined human aging experience system based on VR and AI technologies. Background Technology

[0002] As society ages, the demand for improved living experiences, health care, and related products and services for the elderly is growing. Understanding the difficulties and needs faced by the elderly in their daily lives is crucial for developing more suitable environments, products, and services. However, due to a lack of real-life experience with aging, young people often struggle to fully and deeply understand the inconveniences faced by the elderly.

[0003] Existing aging experience methods are mostly simple simulations, failing to provide an immersive, personalized, and accurate aging experience environment. For example, some traditional simulations merely mimic single symptoms of aging, such as blurred vision or hearing loss, by wearing simple visual or auditory assistive devices. This fails to comprehensively cover the multifaceted changes in physical function and cognitive abilities associated with aging, and makes it difficult to customize the experience for different individuals. Furthermore, the lack of intelligent interaction and real-time feedback mechanisms prevents dynamic adjustments to the experience to better reflect real-world aging conditions. Therefore, developing a virtual-real hybrid system that can achieve a more realistic, comprehensive, and personalized aging experience is of significant practical importance. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a virtual and real human aging experience system based on VR and AI technology to achieve a more realistic, comprehensive and personalized aging experience.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] A virtual-real hybrid experience system for human aging based on VR and AI technologies includes:

[0007] Aging experience wearable device: includes various physical resistance devices and a multi-axis sensor array. The physical resistance devices are used to simulate the aging of bodily functions, specifically including adjustable dampers set at joints and weight modules set at limbs, which are used to increase the resistance of joint movement and the weight of the limbs, respectively. The multi-axis sensor array is used to collect motion data of various parts of the user's body in real time and transmit the collected motion data to the central control module.

[0008] Aging Experience VR Visual Device: Employs a headset for visual and auditory impairments. The headset includes a visual impairment engine, an auditory impairment engine, and a head posture fusion module. The visual impairment engine is used to construct various realistic virtual visual impairment scenes, the auditory impairment engine is used to simulate hearing impairment, and the head posture fusion module is used to collect and fuse the user's head movement data in real time to achieve synchronous switching of perspectives.

[0009] AI Human-Computer Interaction Module: Includes an AI learning companion and a knowledge base. The knowledge base contains knowledge related to the elderly in the fields of medicine, psychology, and sociology. The AI ​​learning companion is used to generate accurate and easy-to-understand answers based on the user's questions or needs, through rapid retrieval and analysis of the knowledge base, and based on the current experience scenario and the user's action data, and provide feedback to the user in the form of voice or text.

[0010] The central control module includes a user database and a central processing unit. The user database stores personal information pre-entered by users. The central processing unit receives data from sensors in the aging experience wearable device and the aging experience VR visual device. Based on the data and the user's personal information, it dynamically adjusts the physical resistance parameters of the physical resistance device and the aging effect parameters of the visual obstacle scene in the aging experience VR visual device to create personalized experience modes for different users. The central processing unit also manages the interaction process between the AI ​​learning companion and the user.

[0011] Furthermore, the central control module also includes a feedback and rating module, which collects rating data from users after the experience ends and optimizes the data for personalized experience modes by combining it with detailed data records during the experience.

[0012] Furthermore, the hearing and visual impairment helmet is a VR helmet, which includes a high-resolution display screen, a speaker, and multiple sensors. The high-resolution display screen serves as a visual impairment engine to construct standard scenes of daily life for the elderly, including home environments, hospitals, and shopping malls, and to depict and render details of light, sound, and object layout within these standard scenes. It is also used to blur scenes, change colors, and reduce field of vision to simulate the visual effects of presbyopia and cataracts. Multiple sensors serve as a head posture fusion module to collect data and track the user's head movements in real time, enabling synchronized switching of perspectives. The speaker serves as a hearing impairment engine to change the volume, timbre, and spatial positioning of sound, simulating the auditory effects of hearing loss and tinnitus.

[0013] Furthermore, the visual impairment engine includes a standard scene rendering module, a pathological feature calculation module, a hybrid weight buffer module, a central control module, a post-processing compositing module, a lens distortion and dispersion processing module, and a high-resolution display module. Constructing various realistic virtual visual impairment scenes includes the following steps:

[0014] Standard scene rendering data and head motion data are acquired and fed into the pathological feature calculation module for pathological feature calculation. Independent textures for four common age-related visual impairments are generated in parallel: presbyopia, cataracts, macular degeneration, and glaucoma. The standard scene rendering data includes RGBA data containing color information and Depth data containing depth information. The independent texture generation rules for each visual impairment are as follows: macular degeneration corresponds to a texture with central masking and contrast adjustment effects; presbyopia corresponds to a texture with focal length shift and radial blur effects; cataracts correspond to a texture with reduced light transmittance and scattering noise effects; and glaucoma corresponds to a texture with peripheral visual field narrowing effects.

[0015] The four independent textures generated by the pathological feature calculation module are written into the four channels of the hybrid weight buffer module, and the value of each channel corresponds to the intensity parameter of a visual impairment. The central control module adjusts the intensity parameter of each channel of the hybrid weight buffer module in real time, with the adjustment range being 0-100% to achieve a stepless transition between different visual impairment effects.

[0016] The post-processing compositing module uses a single-pass blending shader to composite the visual obstacle textures corresponding to each channel in the blending weight buffer module.

[0017] The lens distortion and chromatic aberration processing module receives the synthesized image output by the post-processing synthesis module, and superimposes the lens distortion effect and chromatic aberration effect simulating the distortion of the VR lens itself onto it to generate the final distorted frame.

[0018] The high-resolution display module receives the final distorted frame output by the lens distortion and dispersion processing module and displays the image; at the same time, externally input head motion data is synchronously fed back to the pathological feature calculation module with a delay of <8ms, ensuring that the VR scene perspective and visual impairment effect are refreshed synchronously.

[0019] Furthermore, the AI ​​learning companion specifically includes:

[0020] The data acquisition module is used to collect motion data, physiological data and environmental data from the aging experience wearable device and the aging experience VR vision device in real time at a frequency of 1kHz, and transmit them to the context detection engine module.

[0021] The context detection engine module employs a BERT+GNN fusion network to perform real-time inference at 50Hz on the received action-environment-physiological trimodal features. If the inference confidence is >0.8, the trigger condition is determined to be met and the subsequent process is initiated; if the confidence is ≤0.8, data monitoring continues. The trigger conditions include at least the following three categories: bending over to pick up an object with knee joint damping >70%, cataract shader activated with blinking frequency >20 times / min, and walking speed <0.4m / s with weight >80%.

[0022] The knowledge graph calling module is used to receive the current context vector output by the context detection engine module, match it with the built-in knowledge graph containing 12,000 nodes through vector retrieval, and return the Top-1 best knowledge point;

[0023] The contextualized content generation module is used to feed the Top-1 knowledge points returned by the knowledge graph calling module into the fine-tuning model, combine them with the user's personal information to generate explanations, suggestions or Q&A texts with a length of ≤15 seconds, and simultaneously synthesize corresponding graphic cards;

[0024] The voice emotion compensation module is used to perform TTS processing on the text output by the contextualized content generation module, dynamically reducing the speech rate by 10-20% and lowering the fundamental frequency by 15%, so that the speech output spectrum falls within the auditory sensitive area of ​​the elderly in the range of 300-3000Hz.

[0025] The multimodal output module is used to play the voice processed by the voice emotion compensation module through the speaker, and at the same time push the graphic card to the upper right corner of the VR headset display. The graphic card disappears automatically after 5 seconds.

[0026] Furthermore, the BERT+GNN fusion network includes a BERT sub-network and a GNN sub-network. The BERT sub-network is used to extract textual features, and the GNN sub-network is used to extract structured features. The two sub-networks achieve deep fusion of action-environment-physiological features through a feature concatenation layer.

[0027] Furthermore, the process of receiving data from the sensors of the aging experience wearable device and the aging experience VR visual device, and dynamically adjusting the physical resistance parameters of the physical resistance device and the aging effect parameters of the visual obstacle scene in the aging experience VR visual device based on the data and the user's personal information, allows for the creation of personalized experience modes for different users. Specifically, this includes:

[0028] The user's age, BMI, exercise habits and past injury information are obtained from the user database. The initial personalized model is trained using the XGBoost+Bayesian regression algorithm to generate a four-dimensional parameter vector of damping-weighting-visual-auditory.

[0029] The generated four-dimensional parameter vectors of damping, counterweight, vision, and hearing are transmitted back to the damper, weight module, visual impairment engine, and hearing impairment engine at a frequency of 50Hz to perform closed-loop control. The damper, weight module, visual impairment engine, and hearing impairment engine adjust the experience parameters in real time based on the transmitted vectors.

[0030] Furthermore, the data collected after the user's experience ends, combined with detailed data records during the experience, is used to optimize the data for the personalized experience mode. Specifically, this includes: collecting the user's action, physiological data, and subjective rating data during the experience; incorporating new data into the training set through an incremental learning algorithm; continuously optimizing the parameters of the personalized model; and reapplying the updated model to the next online usage phase.

[0031] Furthermore, the head motion data is six-degree-of-freedom data, including X, Y, and Z axis position information and roll, pitch, and yaw attitude information. After receiving the head motion data, the pathological feature calculation module adjusts the viewpoint parameters of the scene rendering in real time so that the visual impairment effect changes synchronously with the head movement.

[0032] The beneficial effects of this invention are:

[0033] 1) Immersive experience: Using VR technology to create a realistic virtual environment, combined with various aging effect simulations, users can be fully immersed in the aging experience and gain a more realistic and profound feeling.

[0034] 2) Real-time interaction and dynamic adjustment: The system can perceive the user's behavior and emotions in real time, and adjust the virtual environment and aging effect in real time to provide a more intelligent and flexible interactive experience, enhancing the realism and fun of the experience.

[0035] 3) Personalized experience: By analyzing users' personal information and real-time behavioral data using AI technology, we provide each user with a customized aging experience, which more accurately reflects the aging situation that different individuals may face. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the visual impairment engine structure;

[0037] Figure 2 This is a schematic diagram of the AI ​​learning companion structure. Detailed Implementation

[0038] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] See Figures 1-2 The present invention provides a technical solution:

[0040] A virtual-real hybrid experience system for human aging based on VR and AI technologies includes:

[0041] The aging simulation wearable device includes various physical resistance devices and a multi-axis sensor array. The physical resistance devices are used to simulate the aging of bodily functions, specifically including adjustable dampers at the joints and weight modules at the limbs, which are used to increase the resistance to joint movement and the weight of the limbs, respectively. The multi-axis sensor array is used to collect motion data of various parts of the user's body in real time and transmit the collected motion data to the central control module. The physical resistance devices can simulate the aging of bodily functions such as joint stiffness and weakened muscle strength. The dampers increase the resistance to joint movement to restore the discomfort felt by the elderly when moving their joints. The weight modules simulate the heaviness of limb movement after muscle weakness.

[0042] Aging Experience VR Visual Device: Employs a headset for visual and auditory impairments. The headset includes a visual impairment engine, an auditory impairment engine, and a head posture fusion module. The visual impairment engine is used to construct various realistic virtual visual impairment scenes, the auditory impairment engine is used to simulate hearing impairment, and the head posture fusion module is used to collect and fuse the user's head movement data in real time to achieve synchronous switching of perspectives.

[0043] In this embodiment, the hearing and visual impairment helmet is a VR helmet, which includes a high-resolution display screen, a speaker, and multiple sensors. The high-resolution display screen serves as a visual impairment engine to construct standard scenes of daily life for the elderly, including home environments, hospitals, and shopping malls, and to depict and render details of light, sound, and object layout within the standard scenes. It is also used to blur the scenes, change color, and reduce the field of vision to simulate the visual effects of presbyopia and cataract symptoms. Multiple sensors serve as a head posture fusion module to collect data and track the user's head movements in real time, achieving synchronous switching of perspectives. The speaker serves as a hearing impairment engine to change the volume, timbre, and spatial positioning of the sound, simulating the auditory effects of hearing loss and tinnitus symptoms.

[0044] like Figure 1 As shown, the visual impairment engine includes a standard scene rendering module, a pathological feature calculation module, a hybrid weight buffer module, a central control module, a post-processing compositing module, a lens distortion and dispersion processing module, and a high-resolution display module. The steps for constructing various realistic virtual visual impairment scenes are as follows:

[0045] Standard scene rendering data and head motion data are acquired and fed into the pathological feature calculation module for pathological feature calculation. Independent textures for four common age-related visual impairments are generated in parallel: presbyopia, cataracts, macular degeneration, and glaucoma. The standard scene rendering data includes RGBA data containing color information and Depth data containing depth information. The independent texture generation rules for each visual impairment are as follows: macular degeneration corresponds to a texture with central masking and contrast adjustment effects; presbyopia corresponds to a texture with focal length shift and radial blur effects; cataracts correspond to a texture with reduced light transmittance and scattering noise effects; and glaucoma corresponds to a texture with peripheral visual field narrowing effects.

[0046] The four independent textures generated by the pathological feature calculation module are written into the four channels of the hybrid weight buffer module, and the value of each channel corresponds to the intensity parameter of a visual impairment. The central control module adjusts the intensity parameter of each channel of the hybrid weight buffer module in real time, with the adjustment range being 0-100% to achieve a stepless transition between different visual impairment effects.

[0047] The post-processing compositing module uses a single-pass blending shader to composite the visual obstacle textures corresponding to each channel in the blending weight buffer module.

[0048] The lens distortion and chromatic aberration processing module receives the synthesized image output by the post-processing synthesis module, and superimposes the lens distortion effect and chromatic aberration effect simulating the distortion of the VR lens itself onto it to generate the final distorted frame.

[0049] The high-resolution display module receives the final distorted frame output by the lens distortion and dispersion processing module and displays the image; at the same time, externally input head motion data is synchronously fed back to the pathological feature calculation module with a delay of <8ms, ensuring that the VR scene perspective and visual impairment effect are refreshed synchronously.

[0050] The head motion data collected in the head posture fusion module is six-degree-of-freedom data, including X, Y, and Z axis position information and roll, pitch, and yaw attitude information. After receiving the head motion data, the pathological feature calculation module adjusts the viewpoint parameters of the scene rendering in real time so that the visual impairment effect changes synchronously with the head movement.

[0051] The AI ​​human-computer interaction module includes an AI learning companion and a knowledge base. The knowledge base contains knowledge related to the elderly in the fields of medicine, psychology, and sociology. The AI ​​learning companion is used to generate accurate and easy-to-understand answers based on the user's questions or needs, through rapid retrieval and analysis of the knowledge base, and based on the current experience scenario and the user's action data. The answers are then fed back to the user in the form of voice or text.

[0052] Before using the system, users must register through the system client or web platform, filling in detailed personal information, including age, health status, and lifestyle habits. This information will be stored in the system database as the basis for subsequent personalized experience settings.

[0053] like Figure 2 As shown, the AI ​​learning companion specifically includes:

[0054] (1) Data acquisition module, used to collect motion data, physiological data and environmental data from the aging experience wearable device and the aging experience VR vision device in real time at a frequency of 1kHz, and transmit them to the context detection engine module.

[0055] (2) The context detection engine module adopts a BERT+GNN fusion network to perform real-time inference on the received action-environment-physiological three-modal features at 50Hz. If the inference confidence is >0.8, the trigger condition is determined to be met and the subsequent process is started. If the confidence is ≤0.8, the data is monitored. The trigger condition includes at least the following three categories: bending over to pick up an object and knee joint damping >70%, cataract shader is turned on and blinking frequency >20 times / min, walking speed <0.4m / s and counterweight >80%. The BERT+GNN fusion network includes a BERT sub-network and a GNN sub-network. The BERT sub-network is used to extract textual features, and the GNN sub-network is used to extract structured features. The two complete the deep fusion of action-environment-physiological three-modal features through the feature splicing layer.

[0056] (3) Knowledge graph calling module, which is used to receive the current context vector output by the context detection engine module, match the built-in knowledge graph with 12,000 nodes through vector retrieval and return the Top-1 best knowledge point.

[0057] (4) Contextualized content generation module, which is used to send the Top-1 knowledge points returned by the knowledge graph calling module into the fine-tuning model, and generate explanation, suggestion or question and answer text with a length of ≤15 seconds by combining user personal information, and simultaneously synthesize the corresponding graphic card.

[0058] (5) Voice emotion compensation module, used to perform TTS processing on the text output by the contextualized content generation module, dynamically reduce the speech rate by 10-20% and reduce the fundamental frequency by 15%, so that the speech output spectrum falls in the auditory sensitive area of ​​the elderly in the range of 300-3000Hz.

[0059] (6) Multimodal output module, used to play the voice processed by the voice emotion compensation module through the speaker, and push the graphic card to the upper right corner of the VR headset display screen. The graphic card disappears automatically after 5 seconds.

[0060] The central control module includes a user database and a central processing unit. The user database stores personal information pre-entered by users. The central processing unit receives data from sensors in the aging experience wearable device and the aging experience VR visual device. Based on the data and the user's personal information, it dynamically adjusts the physical resistance parameters of the physical resistance device and the aging effect parameters of the visual obstacle scene in the aging experience VR visual device to create personalized experience modes for different users. The central processing unit also manages the interaction process between the AI ​​learning companion and the user.

[0061] The system receives data from sensors in the aging experience wearable device and the aging experience VR visual device. Based on the data and the user's personal information, it dynamically adjusts the physical resistance parameters of the physical resistance device and the aging effect parameters of the visual obstacle scene in the aging experience VR visual device to create personalized experience modes for different users. Specifically, this includes:

[0062] The user's age, BMI, exercise habits and past injury information are obtained from the user database. The initial personalized model is trained using the XGBoost+Bayesian regression algorithm to generate a four-dimensional parameter vector of damping-weighting-visual-auditory.

[0063] The generated four-dimensional parameter vectors of damping, counterweight, vision, and hearing are transmitted back to the damper, weight module, visual impairment engine, and hearing impairment engine at a frequency of 50Hz to perform closed-loop control. The damper, weight module, visual impairment engine, and hearing impairment engine adjust the experience parameters in real time based on the transmitted vectors.

[0064] The central control module generates a personalized aging experience plan for each user based on their personal information, relevant data from the knowledge base, and algorithm models. This includes determining the physical resistance parameters of different parts of the aging suit, selecting the virtual scene in the VR headset, and setting the degree of aging effect.

[0065] Furthermore, the central control module also includes a feedback and scoring module, used to collect scoring data from users after the experience ends, and optimize the personalized experience mode data by combining it with detailed data records during the experience. Specifically, this optimization of the personalized experience mode data by collecting user action, physiological, and subjective scoring data during the experience, incorporating new data into the training set through an incremental learning algorithm, continuously optimizing the parameters of the personalized model, and reapplying the updated model to the next online usage phase.

[0066] After the experience, the system guides users to provide feedback, including evaluations of the realism, difficulty, and enjoyment of the experience, as well as suggestions for improvement. The system collects this feedback and, combined with data recorded during the experience, optimizes and improves the system to enhance the user experience for future users.

[0067] The system utilizes VR technology to create a realistic virtual environment, combined with various aging effect simulations, allowing users to fully immerse themselves in the aging experience and gain a more authentic and profound understanding. It senses user behavior and emotions in real time, adjusting the virtual environment and aging effects accordingly to provide a more intelligent and flexible interactive experience, enhancing the realism and enjoyment of the experience. It can be applied to various fields such as training for elderly service and management personnel, and popular science education, helping to promote the development of related industries and research, and raising public awareness and attention to aging issues. Through AI technology analysis of user personal information and real-time behavioral data, it provides each user with a customized aging experience, more accurately reflecting the aging situations that different individuals may face.

[0068] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A virtual-real hybrid experience system for human aging based on VR and AI technologies, characterized in that, include: Aging experience wearable device: includes various physical resistance devices and a multi-axis sensor array. The physical resistance devices are used to simulate the aging of bodily functions, specifically including adjustable dampers set at joints and weight modules set at limbs, which are used to increase the resistance of joint movement and the weight of the limbs, respectively. The multi-axis sensor array is used to collect motion data of various parts of the user's body in real time and transmit the collected motion data to the central control module. Aging Experience VR Visual Device: Employs a headset for visual and auditory impairments. The headset includes a visual impairment engine, an auditory impairment engine, and a head posture fusion module. The visual impairment engine is used to construct various realistic virtual visual impairment scenes, the auditory impairment engine is used to simulate hearing impairment, and the head posture fusion module is used to collect and fuse the user's head movement data in real time to achieve synchronous switching of perspectives. AI Human-Computer Interaction Module: Includes an AI learning companion and a knowledge base. The knowledge base contains knowledge related to the elderly in the fields of medicine, psychology, and sociology. The AI ​​learning companion is used to generate accurate and easy-to-understand answers based on the user's questions or needs, through rapid retrieval and analysis of the knowledge base, and based on the current experience scenario and the user's action data, and provide feedback to the user in the form of voice or text. The central control module includes a user database and a central processing unit. The user database stores personal information pre-entered by users, while the central processing unit receives data from sensors in the aging experience wearable device and the aging experience VR visual device. Based on the data and the user's personal information, the central processing unit dynamically adjusts the physical resistance parameters of the physical resistance device and the aging effect parameters of the visual obstacle scene in the aging experience VR visual device to create personalized experience modes for different users. The central processing unit also manages the interaction process between the AI ​​learning companion and the user. The hearing and visual impairment helmet is a VR helmet, which includes a high-resolution display, speakers, and multiple sensors. The high-resolution display serves as a visual impairment engine to construct standard scenes of daily life for the elderly, including home environments, hospitals, and shopping malls, and to depict and render details of light, sound, and object layout within these standard scenes. It is also used to blur scenes, change colors, and reduce field of vision to simulate the visual effects of presbyopia and cataracts. Multiple sensors serve as a head posture fusion module to collect data and track the user's head movements in real time, enabling synchronized switching of perspectives. The speakers serve as a hearing impairment engine to change the volume, timbre, and spatial positioning of sound to simulate the auditory effects of hearing loss and tinnitus. The visual impairment engine includes a standard scene rendering module, a pathological feature calculation module, a hybrid weight buffer module, a central control module, a post-processing compositing module, a lens distortion and dispersion processing module, and a high-resolution display module. The process of constructing various realistic virtual visual impairment scenes includes the following steps: Standard scene rendering data and head motion data are acquired and fed into the pathological feature calculation module for pathological feature calculation. Independent textures for four common age-related visual impairments are generated in parallel: presbyopia, cataracts, macular degeneration, and glaucoma. The standard scene rendering data includes RGBA data containing color information and Depth data containing depth information. The independent texture generation rules for each visual impairment are as follows: macular degeneration corresponds to a texture with central masking and contrast adjustment effects; presbyopia corresponds to a texture with focal length shift and radial blur effects; cataracts correspond to a texture with reduced light transmittance and scattering noise effects; and glaucoma corresponds to a texture with peripheral visual field narrowing effects. The four independent textures generated by the pathological feature calculation module are written into the four channels of the hybrid weight buffer module, and the value of each channel corresponds to the intensity parameter of a visual impairment. The central control module adjusts the intensity parameter of each channel of the hybrid weight buffer module in real time, with the adjustment range being 0-100% to achieve a stepless transition between different visual impairment effects. The post-processing compositing module uses a single-pass blending shader to composite the visual obstacle textures corresponding to each channel in the blending weight buffer module. The lens distortion and chromatic aberration processing module receives the synthesized image output by the post-processing synthesis module, and superimposes the lens distortion effect and chromatic aberration effect simulating the distortion of the VR lens itself onto it to generate the final distorted frame. The high-resolution display module receives the final distorted frame output by the lens distortion and dispersion processing module and displays the image; at the same time, externally input head motion data is synchronously fed back to the pathological feature calculation module with a delay of <8ms, ensuring that the VR scene perspective and visual impairment effect are refreshed synchronously.

2. The virtual-real combined experience system for human aging based on VR and AI technology according to claim 1, characterized in that: The central control module also includes a feedback and rating module, which collects rating data from users after the experience ends and optimizes the data for personalized experience modes by combining it with detailed data records during the experience.

3. The virtual-real combined experience system for human aging based on VR and AI technology according to claim 1, characterized in that: The AI ​​learning companion specifically includes: The data acquisition module is used to collect motion data, physiological data and environmental data from the aging experience wearable device and the aging experience VR vision device in real time at a frequency of 1kHz, and transmit them to the context detection engine module. The context detection engine module employs a BERT+GNN fusion network to perform real-time inference at 50Hz on the received action-environment-physiological trimodal features. If the inference confidence is >0.8, the trigger condition is determined to be met and the subsequent process is initiated; if the confidence is ≤0.8, data monitoring continues. The trigger conditions include at least the following three categories: bending over to pick up an object with knee joint damping >70%, cataract shader activated with blinking frequency >20 times / min, and walking speed <0.4m / s with weight >80%. The knowledge graph calling module is used to receive the current context vector output by the context detection engine module, match it with the built-in knowledge graph containing 12,000 nodes through vector retrieval, and return the Top-1 best knowledge point; The contextualized content generation module is used to feed the Top-1 knowledge points returned by the knowledge graph calling module into the fine-tuning model, combine them with the user's personal information to generate explanations, suggestions or Q&A texts with a length of ≤15 seconds, and simultaneously synthesize corresponding graphic cards; The voice emotion compensation module is used to perform TTS processing on the text output by the contextualized content generation module, dynamically reducing the speech rate by 10-20% and lowering the fundamental frequency by 15%, so that the speech output spectrum falls within the auditory sensitive area of ​​the elderly in the range of 300-3000Hz. The multimodal output module is used to play the voice processed by the voice emotion compensation module through the speaker, and at the same time push the graphic card to the upper right corner of the VR headset display. The graphic card disappears automatically after 5 seconds.

4. The virtual-real combined experience system for human aging based on VR and AI technology according to claim 3, characterized in that: The BERT+GNN fusion network includes a BERT sub-network and a GNN sub-network. The BERT sub-network is used to extract textual features, and the GNN sub-network is used to extract structured features. The two sub-networks achieve deep fusion of action-environment-physiology features through a feature concatenation layer.

5. A virtual-real combined experience system for human aging based on VR and AI technology according to claim 1, characterized in that: The system receives data from sensors in the aging experience wearable device and the aging experience VR visual device. Based on the data and the user's personal information, it dynamically adjusts the physical resistance parameters of the physical resistance device and the aging effect parameters of the visual obstacle scene in the aging experience VR visual device to create personalized experience modes for different users. Specifically, this includes: The user's age, BMI, exercise habits and past injury information are obtained from the user database. The initial personalized model is trained using the XGBoost+Bayesian regression algorithm to generate a four-dimensional parameter vector of damping-weighting-visual-auditory. The generated four-dimensional parameter vectors of damping, counterweight, vision, and hearing are transmitted back to the damper, weight module, visual impairment engine, and hearing impairment engine at a frequency of 50Hz to perform closed-loop control. The damper, weight module, visual impairment engine, and hearing impairment engine adjust the experience parameters in real time based on the transmitted vectors.

6. The virtual-real combined experience system for human aging based on VR and AI technology according to claim 2, characterized in that: The process involves collecting rating data from participants after the experience ends, combining it with detailed data records from the experience process, and optimizing the data for the personalized experience mode. Specifically, this includes collecting participants' action, physiological, and subjective rating data during the experience process, incorporating new data into the training set through an incremental learning algorithm, continuously optimizing the parameters of the personalized model, and reapplying the updated model to the next online usage phase.

7. The virtual-real combined experience system for human aging based on VR and AI technology according to claim 1, characterized in that: The head motion data is six-degree-of-freedom data, including X, Y, and Z axis position information and roll, pitch, and yaw attitude information. After receiving the head motion data, the pathological feature calculation module adjusts the viewpoint parameters of the scene rendering in real time so that the visual impairment effect changes synchronously with the head movement.

Images