Intelligent directional and quantitative smell generation method in virtual reality olfaction interaction

Abstract

The invention discloses an intelligent directional and quantitative smell generation method in virtual reality olfaction interaction, and relates to the technical field of virtual reality. The invention provides an intelligent, directional and quantitative smell generation method in order to solve the technical problems that in smell generation work in virtual reality in the prior art, the variety and generation precision of smell are limited, an accurate smell concentration control mechanism is lacked, and interactive linkage with user actions and a virtual environment is lacked. In combination with an odor molecule dynamic combination technology, a deep learning prediction model, closed-loop concentration control and a spatial orientation transmission system, richer, personalized and immersive olfactory interaction experience is provided. Comprising a smell synthesis module, a digital smell control unit, an intelligent smell prediction model, a directional smell transmission system and a virtual wind field simulation module. The method is suitable for being applied to the work of intelligent directional and quantitative smell generation in virtual reality olfaction interaction.

Description

Technical Field

[0001] The present invention relates to the field of virtual reality technology, and in particular to intelligent directional and quantitative odor generation in virtual reality olfactory interaction. Background Art

[0002] With the rapid development of virtual reality (VR) technology, human sensory experience is gradually expanding from two-dimensional to three-dimensional, from vision and hearing to multi-dimensional senses including touch, smell, etc. In recent years, the application of visual, auditory and tactile technologies in virtual reality has made significant progress, and related equipment such as virtual displays, spatial audio systems and tactile feedback devices have become relatively mature. However, as an important but relatively complex part of the five senses of virtual reality, olfactory interaction is still lagging behind in development and has failed to provide users with the same level of immersive experience as vision and hearing.

[0003] Most existing virtual reality olfactory generation systems use a simple release method based on pre-stored odors, usually storing a limited number of basic odors in the odor library and releasing the odors through valve control. This technical approach has significant limitations, mainly in the following aspects:

[0004] First, the limited variety of smells is the main problem faced by traditional olfactory generation systems. Due to the constraints of storage capacity and system design, most odor generation devices can only store a few fixed smells, such as basic types such as floral and fruity scents, and cannot flexibly generate diverse and complex smells. This results in the user's olfactory choices being limited during the experience, and it is impossible to truly simulate the changes in smells in complex scenes.

[0005] Secondly, the existing system has low accuracy in odor generation. Traditional odor control systems mostly rely on mechanical control valves, which make it difficult to accurately control the release concentration and mixing ratio of different odors, and can only achieve simple quantitative control. This method makes the odor generation effect unstable, especially during scene switching or dynamic changes, it is difficult to achieve delicate adjustment of odor intensity.

[0006] Third, the lack of spatial perception and directional control capabilities is also a major flaw in existing olfactory generation technology. In a virtual reality environment, the user's behavior and position have a crucial impact on sensory feedback, and the existing olfactory system cannot form an accurate linkage with the user's movements, resulting in a mismatch between odor release and user position, affecting the user's immersion and experience. For example, when a user approaches an object in a virtual environment, it is difficult for existing technologies to synchronously increase the concentration of related odors, which makes the experience lack realism and interactivity.

[0007] In recent years, some researchers have tried to improve the performance of olfactory generation systems through the combination of chemical odor molecules and deep learning technology, such as realizing a certain dynamic combination of odors based on basic odor storage, or predicting user preferences through machine learning algorithms. However, these attempts still face many challenges, such as insufficient control accuracy of odors, limited personalized adjustment capabilities, and poor implementation of the linkage between odors and virtual scenes.

[0008] In summary, the existing technologies in the field of olfactory generation in virtual reality mainly face the following technical problems: limitation of odor types and generation accuracy, lack of precise odor concentration control mechanism, and lack of interactive linkage with user actions and virtual environment. Summary of the invention

[0009] In order to solve the technical problems of the existing technology in the olfactory generation work in virtual reality, such as the limitation of odor types and generation accuracy, the lack of accurate odor concentration control mechanism, and the lack of interactive linkage with user actions and virtual environment, the present invention proposes an intelligent, directional and quantitative odor generation method, which combines odor molecule dynamic combination technology, deep learning prediction model, closed-loop concentration control and spatial directional transmission system to provide a richer, personalized and immersive olfactory interactive experience, thereby overcoming the shortcomings of the existing technology, specifically:

[0010] The intelligent directional and quantitative odor generation method in virtual reality olfactory interaction is implemented based on the following system:

[0011] The odor synthesis module generates odors based on the chemical odor molecule theory through the dynamic combination of multiple basic odor molecular units. The odor synthesis module includes a basic odor storage unit, an odor mixing chamber, a micro pump and a control chip to achieve the generation of multiple complex odors;

[0012] The digital odor control unit, based on high-precision electromagnetic control technology, can achieve precise adjustment of the odor release concentration. It includes a solenoid valve and a concentration sensor. It controls the opening and closing time of the solenoid valve through a pulse width modulation signal and controls the release concentration of the odor through closed-loop feedback.

[0013] Intelligent odor prediction model, which uses deep learning algorithms to predict users' preferences for different odors based on historical user interaction data, and dynamically adjusts odor composition and intensity to meet users' personalized olfactory preferences;

[0014] Directional scent delivery system, including infrared and visual sensors, combined with a micro-injector array controlled by a stepper motor, which adjusts the scent injection direction in real time according to the user's head position and orientation, achieving directional scent delivery based on the user's spatial position;

[0015] The virtual wind field simulation module simulates the wind field in the target virtual environment through the computational fluid dynamics model and controls the odor injectors in the corresponding direction to synchronize the dynamic changes of the odor with the virtual wind direction.

[0016] Methods include:

[0017] The step of selecting corresponding scene odor information according to the target virtual scene, and generating the scene odor through an odor synthesis module according to the scene odor information;

[0018] The step of releasing the scene odor and adjusting the release concentration in real time;

[0019] The step of adjusting the composition of the scene smell in real time according to the user's preference for different smells;

[0020] The step of adjusting the direction of the scent spray in real time according to the user's head position and orientation;

[0021] The step of adjusting the injection parameters in the response direction in real time according to the wind field parameters in the target virtual scene.

[0022] Furthermore, the basic odor molecule units include floral matrix, fruity matrix, woody matrix, herbal matrix, smoky matrix and spice matrix.

[0023] Furthermore, the concentration sensor in the digital odor control unit is a semiconductor gas sensor.

[0024] Furthermore, the control chip includes a microcontroller-based closed-loop control algorithm for dynamically adjusting the flow rate of the micropump according to feedback from the odor concentration sensor during the odor generation process.

[0025] Furthermore, the intelligent odor prediction model is implemented based on a long short-term memory network algorithm.

[0026] Furthermore, the stepper motor in the directional odor delivery system is a servo stepper motor.

[0027] Furthermore, the data from the infrared sensor and the visual sensor are processed by a fusion algorithm.

[0028] The present invention also provides a computer storage medium, wherein the computer storage medium stores a computer program. When the computer program is read by a computer, the computer executes the following steps:

[0029] The step of selecting corresponding scene odor information according to the target virtual scene, and generating the scene odor through an odor synthesis module according to the scene odor information;

[0030] The step of releasing the scene odor and adjusting the release concentration in real time;

[0031] The step of adjusting the composition of the scene smell in real time according to the user's preference for different smells;

[0032] The step of adjusting the direction of the scent spray in real time according to the user's head position and orientation;

[0033] The step of adjusting the injection parameters in the response direction in real time according to the wind field parameters in the target virtual scene.

[0034] The present invention also provides an intelligent directional and quantitative odor generation device in virtual reality olfactory interaction, the device comprising:

[0035] According to the target virtual scene, corresponding scene odor information is selected, and according to the scene odor information, a module for generating scene odor through an odor synthesis module is generated;

[0036] A module for releasing the scene odor and adjusting the release concentration in real time;

[0037] A module for adjusting the composition of the scene smell in real time according to the user's preference for different smells;

[0038] A module that adjusts the direction of scent spray in real time based on the user's head position and orientation;

[0039] A module that adjusts the injection parameters in the response direction in real time according to the wind field parameters in the target virtual scene.

[0040] Compared with the prior art, the technical solution provided by the present invention is beneficial in that:

[0041] This solution generates complex odors by dynamically combining multiple basic odor molecules, which greatly improves the diversity of odors compared to the traditional method of storing only a limited number of odors. The proportion of each basic odor in the odor synthesis module is adjusted in real time by the control chip, achieving a highly personalized odor generation experience, allowing users to experience a more realistic and rich olfactory feedback effect.

[0042] The digital odor control unit uses high-precision electromagnetic control technology and closed-loop feedback control to ensure that the concentration of odor release is precisely controllable within the millisecond range. This control method is more accurate than the existing simple valve control method, avoiding the problem of unstable odor release concentration or concentration not meeting expectations, and effectively improving the immersion and comfort of the user experience.

[0043] The intelligent odor prediction model uses a deep learning algorithm to predict preferences based on the user's historical interaction data and dynamically adjusts the composition and intensity of odors. Compared with the existing simple preset odor release system, this model provides more accurate odor adjustments by learning the user's personalized preferences, effectively improving the personalized level of odor interaction, and making the user's experience in the virtual reality scene more in line with personal needs.

[0044] The directional odor delivery system uses multiple infrared and visual sensors to capture the user's head position in real time, and combines it with a micro-injector array controlled by a stepper motor to achieve directional odor delivery. Compared with the lack of spatial perception in existing odor systems, this method significantly enhances the linkage between odor and user behavior, allowing odor release to accurately match the user's position and movement, enhancing the immersive and interactive experience of olfactory feedback.

[0045] The virtual wind field simulation module uses a computational fluid dynamics model to simulate the wind field in the virtual scene and dynamically adjusts the flow direction and diffusion speed of the smell. This method effectively overcomes the lack of directionality in the existing technology of smell transmission, allowing users to feel the reality of the smell flowing with the wind in the virtual scene, significantly enhancing the spatial sense and realism of the smell experience.

[0046] The multi-dimensional parameter adjustment mechanism adjusts the odor generation in real time according to environmental factors such as temperature and humidity in the virtual scene. This method makes the odor release more consistent with the actual conditions of the virtual scene through more refined environmental parameter control, enhancing the authenticity and delicacy of the scene experience. Compared with the traditional odor generation system that does not consider environmental factors, this mechanism makes the odor interaction more dynamic and diversified.

[0047] It is suitable for use in intelligent directional and quantitative odor generation in virtual reality olfactory interaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of intelligent odor generation and mixing system;

[0049] Figure 2 This is the flow chart of odor release and closed-loop feedback control. DETAILED DESCRIPTION

[0050] Implementation method 1: Combination Figure 1 and 2 , the technology of the present invention is as follows:

[0051] include:

[0052] Basic odor storage and selection: The system first stores a variety of basic odors in the odor synthesis module, and selects appropriate odors for combination according to the needs of the virtual scene.

[0053] Odor synthesis and mixing: Selected base odors are dynamically combined to create complex odors and precisely mixed in a mixing chamber.

[0054] Closed-loop control of odor release: The mixed odor is subjected to concentration adjustment and closed-loop feedback control through a digital odor control unit to ensure that the odor concentration meets expectations.

[0055] Personalized odor prediction: Based on historical user interaction data, the intelligent odor prediction model predicts the user's olfactory preferences and dynamically adjusts the composition and intensity of the odor.

[0056] Directional odor delivery: The sensor obtains the user's head position and direction information, and combines with the injector controlled by the stepper motor to achieve directional delivery of odor.

[0057] Virtual wind field simulation: The virtual wind field simulation module controls the diffusion of odor in real time to make it conform to the wind direction and flow rate in the virtual environment.

[0058] Multi-dimensional parameter adjustment: Through real-time monitoring of parameters such as temperature and humidity, the odor release strategy is adjusted to make the odor changes in the virtual scene more realistic.

[0059] Detailed description:

[0060] Step 1: Basic scent storage and selection

[0061] The system first stores a variety of basic scents through the scent synthesis module. These basic scents include floral matrix, fruity matrix, woody matrix, herbal matrix, smoky matrix, etc. Each matrix is ​​stored in an independent basic scent storage unit. The control chip selects and controls the release amount of the scent in the corresponding storage unit according to the requirements of the virtual scene. The selection process is based on user input or scene preset, so that the scent combination can meet the specific scene requirements.

[0062] Step 2: Odor synthesis and mixing

[0063] After the selected basic scent, the micro pump delivers different basic scents to the scent mixing chamber in proportion. In the mixing chamber, various scents are fully mixed through vortex mixing technology to generate complex scent combinations. The control chip accurately controls the proportion of different basic scents by receiving user input or system instructions to ensure that the generated scent matches the scene. For example, in a virtual forest scene, more herbal matrices (such as menthol) can be selected to combine with a small amount of woody scent matrices (such as santalin) to simulate the real forest scent.

[0064] Specific:

[0065] The present invention generates complex odors by dynamically controlling the ratio of basic odors, and the formula is as follows:

[0066]

[0067] W i : The final weight ratio of the i-th basic smell.

[0068] w i : Initial scene weight, used to determine the initial importance of the base odor i in a specific virtual scene.

[0069] Δw i :The environmental factor adjustment weight is dynamically adjusted by environmental parameters (such as temperature\(T\) and humidity\(H\)). The specific calculation is as follows:

[0070] Δw i =f(T,H)

[0071] Among them, f(T, H) is the environmental adjustment function based on historical data fitting, which reflects the importance of temperature and humidity on odor.

[0072] P i : User preference weight, calculated by the deep learning model based on the user's historical data, is used to indicate the user's preference for a certain smell.

[0073] Step 3: Closed-loop control of odor release

[0074] The mixed smell will be released through the digital smell control unit (DCOU). The DCOU contains multiple solenoid valves and concentration sensors, which control the opening and closing time of the solenoid valves through pulse width modulation (PWM) signals to accurately adjust the release concentration of the smell. The concentration sensor monitors the concentration of the smell in real time and feeds the monitoring data back to the control system to achieve closed-loop regulation, ensuring that the concentration of the smell remains within the set range, thereby preventing the smell from being too strong or too weak.

[0075] In order to ensure that the concentration of odor release meets expectations, the present invention uses a closed-loop feedback control system. The mathematical description of odor concentration regulation is as follows:

[0076]

[0077] Where, C(t): odor concentration at time t.

[0078] C d : Target concentration is the desired odor concentration.

[0079] e(t)=C d -C m (t): Error, which represents the difference between the target concentration and the currently monitored concentration.

[0080] C m(t): monitored concentration at time t.

[0081] K p , K i , K d : They are proportional, integral and differential gains, respectively, used to adjust the response speed and accuracy of feedback control.

[0082] The integral term is used to eliminate long-term errors and ensure stable concentration.

[0083] The rate of change of error is used to predict the trend of error and respond quickly.

[0084] Step 4: Personalized odor prediction

[0085] The closed-loop control results of odor release will become the input of the personalized odor prediction model. The intelligent odor prediction model uses a deep learning algorithm to predict the user's preference for different odors by analyzing the user's historical interaction data. The model combines the user's physiological response data (such as heart rate, breathing rate, etc.) to automatically adjust the composition and concentration of the odor to meet the user's personalized needs. For example, when the user enters the virtual garden scene, if the model predicts that the user prefers floral scent, the proportion of floral matrix will be increased to enhance the user's immersive experience.

[0086] In order to achieve personalized odor generation, the present invention uses a preference prediction model based on deep learning, and the prediction formula is as follows:

[0087] P i =σ(W h ·h t-1 +W x ·x t +b)

[0088] Among them, P i : The preference for the i-th odor.

[0089] σ(·): Activation function. Here, the Sigmoid activation function is used to limit the output to the interval [0,1].

[0090] W h , W x : are the weight matrices for hidden states and inputs, respectively.

[0091] h t-1 : The hidden state of the previous moment, indicating the user’s past olfactory preference history.

[0092] x t : Current input data, including environmental parameters and user's physiological feedback data (such as heart rate, breathing rate, etc.).

[0093] b: Bias term.

[0094] Step 5: Directed Scent Delivery

[0095] After the personalized smell synthesis is completed, the system enters the directional smell delivery stage. The spatial perception and tracking system obtains the user's head position and movement direction in real time through infrared and visual sensors, and inputs the data into the control system. The control system controls the direction of the micro-injector through a stepper motor to synchronize the smell with the user's movements. For example, when the user moves in a certain direction in the virtual kitchen scene, the system increases the concentration of food odor in that direction to enhance the user's immersion.

[0096] Step 6: Virtual Wind Farm Simulation

[0097] To further enhance the realism of the smell, the virtual wind field simulation module simulates the wind direction and airflow in the virtual environment in real time through a computational fluid dynamics (CFD) model. Based on the calculation results of the wind field model, the system controls the smell injector array so that the diffusion of the smell matches the virtual wind direction. For example, when the user walks in a virtual grassland scene, the system can simulate the feeling of a breeze and naturally deliver the smell of grass and wildflowers to the user's nose, significantly enhancing the spatial sense and realism of the smell.

[0098] The diffusion of odor is simulated in a virtual environment. The present invention dynamically controls the diffusion of odor through a computational fluid dynamics model (CFD), and the formula is as follows:

[0099]

[0100] in In Location and the wind speed vector at time t.

[0101] The pressure gradient term represents the change in air pressure at different locations in the virtual scene.

[0102] The viscous diffusion term, where μ is the viscosity coefficient of air, describes the diffusion behavior of odor in the virtual environment. The external force term represents the force applied by the odor spray device, including the direction and strength of the sprayer.

[0103] Step 7: Multi-dimensional parameter adjustment

[0104] In the final multi-dimensional adjustment stage, the system uses a variety of sensors to monitor the environmental parameters in the virtual scene in real time, including temperature, humidity, etc. The control module adjusts the release concentration and diffusion rate of the odor according to these environmental parameters. For example, in the virtual baking scene, when the temperature rises, the system will gradually increase the concentration of the bread aroma, allowing users to experience a more realistic baking process. At the same time, changes in humidity will also affect the residence time of the odor. For example, in a simulated forest scene after rain, increased humidity will make the smell of soil and grass stronger.

[0105] Implementation method 2: The specific implementation method of the present invention is:

[0106] 1. Intelligent odor generation and mixing system

[0107] 1.1 Odor synthesis module

[0108] This technical solution introduces an odor synthesis module, which is designed based on the theory of chemical odor molecules and can combine and synthesize a variety of basic odor molecules. Different from the existing system that only relies on pre-stored fixed odors, this solution realizes a more flexible odor generation process through the dynamic combination of odor molecule units. For example, by precisely controlling the proportion of different basic odor molecules, it is possible to simulate the production of a variety of complex odors such as floral, fruity, and woody, greatly expanding the types of odors that users experience.

[0109] Specific:

[0110] Hardware structure: The odor synthesis module consists of multiple basic odor storage units, an odor mixing chamber, a micro pump, and a control chip. Each basic odor storage unit contains a basic odor compound, and the micro pump is used to deliver different basic odors to the mixing chamber in proportion.

[0111] Basic scent types: Common basic scents include:

[0112] Floral base (such as rose alcohol, jasmine lactone): used to generate various floral scents.

[0113] Fruity aroma matrix (such as ethyl caproate, ethyl butyrate): used to generate different fruity aromas, such as apple, strawberry, lemon, etc.

[0114] Woody base (such as cedar oil, sandalwood alcohol): used to simulate woody smells.

[0115] Herbal matrix (such as menthol, thyme oil): used to generate herbal and fresh scents.

[0116] Smoky base (such as guaiac wood oil, vanillin): used to produce smoky and roasted aromas.

[0117] Odor mixing and release: The control chip receives user input or system instructions to control the operation of the micro pump and adjust the ratio of different basic odors. In the mixing chamber, the odor is mixed by vortex and finally released by the micro nozzle.

[0118] Dynamic Combinatorial Logic:

[0119] Ratio control: The mixing ratio of different odor matrices is adjusted in real time through algorithms. For example, in a virtual garden scene, the system may increase the proportion of floral matrices (such as rose alcohol) and moderately mix in herbal matrices (such as menthol) to simulate the odor levels in a real environment.

[0120] Scene matching: The smell combination logic will be dynamically adjusted according to the virtual scene the user is in. For example, in a virtual kitchen, the aroma of roasted fruit can be simulated by combining a fruity base (such as ethyl caproate) with a smoky base (such as vanillin).

[0121] User preference learning: Through user feedback data, the system can gradually optimize the mixing ratio of basic scents to match the user's olfactory preferences, thereby improving the immersiveness of the user experience.

[0122] Mixing ratio algorithm:

[0123] Input parameters: scene type, user location, environmental parameters (such as temperature, humidity), and user preference weight.

[0124] Mixing ratio calculation:

[0125] Scene weight assignment: First, assign initial weights to each basic odor matrix according to the scene type. For example, in a garden scene, the initial weight of the floral matrix is ​​0.5, the herbal matrix is ​​0.3, and the woody matrix is ​​0.2.

[0126] Logic: The scene weight assignment is based on predefined scene odor matching data in the database, which stores the odor combinations and their weights corresponding to different scenes. For example, the database entry corresponding to the garden scene may define the default weight ratios of floral, herbal, and woody scents. These ratios are derived based on experimental research and user experience analysis.

[0127] Environmental parameter adjustment: Adjustments are made based on current environmental parameters. For example, when the temperature rises, the weight of the fruity aroma matrix can be appropriately increased to simulate the aroma of tropical fruits.

[0128] Logic: Environmental parameter odor adjustment is based on a set of rule mapping functions, for example:

[0129] Among them, represents the adjusted weight, represents the initial weight, is the temperature, is the humidity, and is the function adjusted according to the temperature and humidity. The function is obtained by fitting historical data. For example, when the temperature rises above 25°C, the weight of fruity odor increases by 0.1.

[0130] User preference weighting: The weight of each basic odor matrix is ​​adjusted according to the user's historical preference. For example, if the user prefers floral scent, the floral scent matrix weight is increased by 0.1 based on the initial weight.

[0131] Logic: User preferences are predicted by a deep learning model that uses historical data about the user (such as reactions to specific smells) to adjust the weights. The weighting formula is:

[0132] Wherein, is the user's preference for smell (ranging from 0 to 1).

[0133] Normalization: Normalize the weights of all matrices to ensure that the sum is 1.

[0134] logic:

[0135] This formula ensures that the sum of the weights of all odors is 1, preventing one odor from being overly prominent.

[0136] Odor ratio output: According to the weighted weight ratio, the final mixing ratio of each basic odor is calculated, and the micro pump is controlled to perform precise odor synthesis.

[0137] 1.2 Odor prediction model

[0138] In order to enhance the user's sense of immersion, this solution establishes a user's olfactory preference model based on a deep learning algorithm. By analyzing the historical data of user interactions, the system can learn the user's preferences and sensitivity to specific smells. In a virtual reality scene, the odor generation module can dynamically adjust the composition and intensity of the odor according to the user's personalized preferences. For example, when the user is in a virtual forest scene, the system can prioritize the floral odors that the user prefers and suppress certain plant odors that the user may not like, thereby achieving intelligent olfactory regulation. This personalized odor generation process makes the olfactory experience not only rich and diverse, but also optimized for individual needs.

[0139] Specific:

[0140] Step 1: Data Collection

[0141] Use the built-in sensors of the head-mounted device (such as biosensors, facial expression capture cameras, heart rate sensors, etc.) to collect real-time data on the user's reaction to different smell experiences.

[0142] Reaction data includes changes in the user's facial expression, heart rate, breathing rate, and subjective feedback ratings for specific smells (collected through the user interaction interface).

[0143] Step 2: Data Preprocessing

[0144] The collected raw data is preprocessed, including noise removal, standardization, and feature extraction. For example, facial expression data can be extracted through a convolutional neural network (CNN) to extract feature points, and heart rate data can be filtered to remove interference signals.

[0145] Feature engineering methods are used to extract key features, such as the impact of changes in odor concentration on heart rate.

[0146] Step 3: Model training

[0147] A convolutional neural network (CNN) combined with a recurrent neural network (RNN) is used to model the user's olfactory response. CNN is used to extract spatial features from visual and physiological data, and RNN is used to capture the user's time series dependency on odor changes.

[0148] The dataset is divided into training set, validation set and test set, and the model is trained using supervised learning method to minimize the error between predicted user reactions and actual reactions.

[0149] Step 4: Preference prediction

[0150] Through the trained model, the system can predict the user's preference for different smells. For example, when a user enters a scene, the system can predict the smell combination that the user is most likely to like based on the user's historical data.

[0151] The prediction results are output in the form of preference weights for subsequent adjustment of the odor mixing ratio.

[0152] Step 5: Model Update

[0153] The model will be continuously updated as users use it. Through online learning, new user response data will be incorporated into the training set, and the model will be fine-tuned regularly to improve the accuracy and personalization of the prediction.

[0154] Personalized adjustment: When a user enters a specific virtual scene, the system adjusts the scent generation strategy in real time based on the results of the prediction model to meet the user's personalized needs. For example, if the prediction model determines that the user has a high preference for a certain flower scent, the system will increase the weight of the corresponding flower scent matrix to ensure that the scent experience meets the user's preferences.

[0155] 2. Quantitative odor release control

[0156] 2.1 Digital Odor Control Unit (DCOU)

[0157] The Digital Control Odor Unit (DCOU) proposed in this solution is based on high-precision electromagnetic control technology and can accurately adjust the concentration, time and rate of odor release. The core of DCOU is to achieve millisecond-level adjustment of the concentration of each odor molecule through digital control of the odor release process. For example, when simulating a virtual cooking scene, DCOU can accurately adjust the release concentration of different odors such as spices and oils according to the different stages of the cooking process, allowing users to experience the gradually rich odor changes.

[0158] Specific:

[0159] Solenoid valve control: The DCOU contains multiple solenoid valves, which can precisely control the opening and closing time of each valve through pulse width modulation (PWM) signals, thereby adjusting the release rate of the odor.

[0160] Concentration sensor feedback: A concentration sensor is integrated into the odor release path to monitor the concentration value of the odor in real time and feed the data back to the control system to achieve closed-loop control and ensure the accuracy of odor release.

[0161] 2.2 Closed-loop feedback control system

[0162] This solution introduces a closed-loop feedback control system, which monitors the odor concentration in the environment through real-time odor sensors and feeds the monitoring data back to the control system to form a closed-loop regulation mechanism. When the odor release concentration exceeds expectations, the system can automatically adjust the release rate to keep the odor concentration within the preset range. For example, in a garden in a virtual scene, if the user is too close to a certain flower bush, resulting in a high odor concentration, the closed-loop feedback system will reduce the release of the flower fragrance to avoid the discomfort caused by the excessively strong odor.

[0163] Specific:

[0164] Real-time monitoring: The closed-loop system consists of an odor concentration sensor, a control processor, and a regulating device. The sensor detects the ambient odor concentration in real time and sends the data to the control processor.

[0165] Automatic adjustment: The control processor analyzes sensor data and compares it with the preset target concentration to automatically control the working state of the odor release device, adjust the speed and concentration of odor release, and ensure that it is always maintained within a comfortable range.

[0166] 3. Directed odor delivery technology

[0167] 3.1 Spatial perception and tracking system

[0168] In order to enhance the linkage between the sense of smell and user behavior, this solution adopts a spatial perception and tracking system, which uses multiple infrared and visual sensors to capture the user's head position, direction and movement trajectory in real time. In conjunction with a multi-nozzle micro-odor injection device, the system can achieve directional odor delivery. When the user turns his head or approaches a specific object in a virtual environment, the odor generation module will instantly adjust the direction of the odor injection to keep the odor consistent with the user's spatial position. For example, when the user approaches a certain direction in the virtual kitchen scene, the system will intelligently adjust to increase the concentration of the odor in that direction, giving the user an immersive experience.

[0169] Specific:

[0170] Multi-sensor fusion: The system fuses data from multiple infrared sensors and cameras to accurately obtain the user's location information.

[0171] Micro-injector array: The scent injection device consists of multiple micro-injectors, each of which is controlled by a stepper motor to ensure that the scent can adjust the injection direction in real time as the user's head moves.

[0172] 3.2 Virtual wind farm simulation

[0173] This solution innovatively introduces a virtual wind field simulation function. By modeling and simulating the wind direction and airflow in the virtual environment in real time, the system can dynamically control the flow direction and diffusion speed of the smell. For example, when the user walks on a virtual grassland, the system can simulate the feeling of a breeze blowing, and deliver the smell of grass and wildflowers to the user's nose in a natural wind-blown way. The combination of this odor delivery method and the wind field can significantly enhance the spatial sense and realism of the odor experience, which is different from the lack of directionality in odor delivery in the existing technology.

[0174] Specific:

[0175] Wind field modeling: Use computational fluid dynamics (CFD) models to simulate the wind field in the virtual scene and calculate the flow path and speed of the airflow.

[0176] Synchronous odor diffusion control: According to the calculation results of the wind field model, the system controls the odor injectors in the corresponding direction to synchronize the dynamic changes of the odor with the virtual wind direction.

[0177] 4. Multi-dimensional odor experience control

[0178] 4.1 Multidimensional parameter adjustment mechanism

[0179] In order to provide a more delicate olfactory experience, this solution introduces a multi-dimensional parameter adjustment mechanism to dynamically adjust the generation of odors according to factors such as time, temperature, and humidity in the virtual scene. For example, in a virtual baking scene, the system can automatically adjust the concentration of the bread aroma according to the temperature changes in the scene: as the oven temperature rises, the aroma of the bread gradually becomes stronger. When the temperature reaches a certain peak, the aroma reaches its strongest point and gradually weakens after baking. In addition, the system can also adjust the diffusion rate and residence time of the odor according to the virtual humidity conditions. For example, in a simulated forest scene after rain, the smell of moist soil and grass will be stronger.

[0180] Specific:

[0181] Multi-parameter sensor: collect virtual environment parameters in real time through multiple sensors such as temperature and humidity sensors.

[0182] Odor generation regulation: Based on the collected environmental parameters, the system adjusts the concentration and rate of odor release through the control module. For example, when the humidity increases, the odor residence time is increased to simulate the odor changes in the real environment.

[0183] 4.2 Linkage with scene events

[0184] The odor generation in this technical solution can also be linked with events in the virtual scene. For example, in a virtual game, when a player successfully unlocks a scene, the system can enhance the sense of reward by enhancing a certain pleasant smell (such as fruit or mint); and in some tense scenes, the system can generate some burning or pungent smells to enhance the emotional fluctuations brought about by the scene. This multi-dimensional, scene event-driven odor regulation makes the odor experience no longer static, but highly dynamic and interactive.

[0185] Specific:

[0186] Event detection module: The event detection module in the virtual reality system monitors scene changes in real time, such as player behavior, scene unlocking and changes.

[0187] Linked odor generation: When a specific event is detected, the control system automatically calls the corresponding odor generation strategy, such as increasing the concentration of pleasant odors in success scenarios to enhance the user's emotional experience.

[0188] Compared with the existing technology, the main differences and innovations of this solution include:

[0189] Intelligent odor generation and personalized adjustment: Traditional odor generation systems often rely on limited pre-stored odors. This method achieves dynamic generation and intelligent combination through the odor synthesis module, significantly improving the richness of odor types. At the same time, based on the prediction model of the user's personalized olfactory preference, the system can achieve dynamic adjustment and optimization to provide a highly personalized olfactory experience.

[0190] Precise quantitative odor control: This solution achieves millisecond-level odor release regulation through a digital odor control unit (DCOU) and maintains the precise concentration of the odor through a closed-loop feedback system, avoiding the problems of unstable release or difficult to control odor concentration in traditional systems.

[0191] Directional odor delivery and spatial interaction: Different from the traditional odor system that lacks a sense of space, this solution uses spatial perception and tracking technology and virtual wind field simulation to achieve directional odor delivery based on user location and behavior, enhancing the interactivity and immersion between odor and the virtual environment.

[0192] Multi-dimensional dynamic control: This solution can not only dynamically adjust the smell based on the multi-dimensional parameters of the virtual scene (such as temperature, humidity, and time), but also link with the events in the scene to provide a highly dynamic olfactory feedback experience. This makes the smell experience no longer single, but can change in real time with the changes in the virtual environment, thereby enhancing the user's immersion in virtual reality.

[0193] This technical solution significantly improves the olfactory interaction experience in virtual reality through intelligent odor generation and mixing systems, precise quantitative control, directional transmission technology, and multi-dimensional dynamic adjustment. Compared with existing technologies, this solution has achieved breakthroughs in odor types, generation accuracy, spatial perception, and linkage with virtual environments. Its intelligent prediction and personalized adjustment functions further enhance the immersion and comfort of user experience.

Claims

1. An intelligent directional and quantitative odor generation method in virtual reality olfactory interaction, characterized in that: Based on the following system implementation: The odor synthesis module generates odors based on the chemical odor molecule theory through the dynamic combination of multiple basic odor molecular units. The odor synthesis module includes a basic odor storage unit, an odor mixing chamber, a micro pump and a control chip to achieve the generation of multiple complex odors; The digital odor control unit, based on high-precision electromagnetic control technology, can achieve precise adjustment of the odor release concentration. It includes a solenoid valve and a concentration sensor. It controls the opening and closing time of the solenoid valve through a pulse width modulation signal and controls the release concentration of the odor through closed-loop feedback. Intelligent odor prediction model, which uses deep learning algorithms to predict users' preferences for different odors based on historical user interaction data, and dynamically adjusts odor composition and intensity to meet users' personalized olfactory preferences; Directional scent delivery system, including infrared and visual sensors, combined with a micro-injector array controlled by a stepper motor, which adjusts the scent injection direction in real time according to the user's head position and orientation, achieving directional scent delivery based on the user's spatial position; The virtual wind field simulation module simulates the wind field in the target virtual environment through the computational fluid dynamics model and controls the odor injectors in the corresponding direction to synchronize the dynamic changes of the odor with the virtual wind direction. Methods include: The step of selecting corresponding scene odor information according to the target virtual scene, and generating the scene odor through an odor synthesis module according to the scene odor information; The step of releasing the scene odor and adjusting the release concentration in real time; The step of adjusting the composition of the scene smell in real time according to the user's preference for different smells; The step of adjusting the direction of the scent spray in real time according to the user's head position and orientation; The step of adjusting the injection parameters in the response direction in real time according to the wind field parameters in the target virtual scene.

2. The intelligent directional and quantitative odor generation method in virtual reality olfactory interaction according to claim 1 is characterized in that: The basic odor molecular units include floral aroma matrix, fruit aroma matrix, wood aroma matrix, herbal matrix, smoky matrix and spice matrix.

3. The intelligent directional and quantitative odor generation method in virtual reality olfactory interaction according to claim 1 is characterized in that: The concentration sensor in the digital odor control unit is a semiconductor gas sensor.

4. The intelligent directional and quantitative odor generation method in virtual reality olfactory interaction according to claim 1, characterized in that: The control chip includes a microcontroller-based closed-loop control algorithm for dynamically adjusting the flow rate of the micropump according to feedback from the odor concentration sensor during the odor generation process.

5. The intelligent directional and quantitative odor generation method in virtual reality olfactory interaction according to claim 1 is characterized in that: The intelligent odor prediction model is implemented based on a long short-term memory network algorithm.

6. The intelligent directional and quantitative odor generation method in virtual reality olfactory interaction according to claim 1 is characterized in that: The stepper motor in the directional scent delivery system is a servo stepper motor.

7. The intelligent directional and quantitative odor generation method in virtual reality olfactory interaction according to claim 1 is characterized in that: The data of the infrared sensor and the visual sensor are processed through a fusion algorithm.

8. A computer storage medium, characterized in that The computer storage medium stores a computer program. When the computer program is read by a computer, the computer executes the following steps: The step of selecting corresponding scene odor information according to the target virtual scene, and generating the scene odor through an odor synthesis module according to the scene odor information; The step of releasing the scene odor and adjusting the release concentration in real time; The step of adjusting the composition of the scene smell in real time according to the user's preference for different smells; The step of adjusting the direction of the scent spray in real time according to the user's head position and orientation; The step of adjusting the injection parameters in the response direction in real time according to the wind field parameters in the target virtual scene.

9. An intelligent directional and quantitative odor generation device in virtual reality olfactory interaction, characterized in that: The device includes: According to the target virtual scene, corresponding scene odor information is selected, and according to the scene odor information, a module for generating scene odor through an odor synthesis module is generated; A module for releasing the scene odor and adjusting the release concentration in real time; A module for adjusting the composition of the scene smell in real time according to the user's preference for different smells; A module that adjusts the direction of scent spray in real time based on the user's head position and orientation; A module that adjusts the injection parameters in the response direction in real time according to the wind field parameters in the target virtual scene.

Images