VR environment airflow feedback method based on user behaviors and related device

Abstract

The invention discloses a VR environment airflow feedback method based on user behaviors and a related device, and the method comprises the steps: obtaining behavior data and environment event data of a user in a VR environment, and carrying out the event processing according to the priority attribute of the environment event data; analyzing the scene change according to the event processing result and the user behavior data, generating scene prediction data, and determining an initial airflow parameter; controlling a distributed fan array to generate a basic airflow effect according to the initial airflow parameter; generating airflow perception characteristics according to perception feedback of a user on the basic airflow effect and optimizing airflow parameters; and controlling the fan array to generate an airflow feedback signal according to the optimized airflow parameter. According to the technical scheme, accurate space-time synchronization of airflow feedback and user behaviors is achieved through multi-dimensional behavior data collection and scene prediction analysis, the user experience is improved through personalized perception feature optimization, and the problems that behavior perception lags behind and the user adaptability is poor in an existing VR airflow feedback system are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual reality human-computer interaction, and particularly to a VR environment airflow feedback method based on user behavior and related devices. Background Art

[0002] Virtual reality (VR) technology is an interactive technology that generates a three-dimensional space through computer simulation. It provides users with immersive visual and auditory experiences through head-mounted display devices. By real-time tracking of users' head movements and hand operations, and cooperating with a high-precision display system and a spatial audio system, VR technology enables users to obtain a highly realistic interaction experience in a virtual environment. With the continuous development of technology, VR has been widely applied in many fields such as game entertainment, education and training, medical rehabilitation, and industrial design. Its core value lies in creating a real and credible immersive experience for users.

[0003] The existing VR airflow feedback technologies are mainly divided into two schemes: an in-headset fan system and an external wind force simulation device. The in-headset fan system is mainly used for device heat dissipation, directly discharging the airflow outside the device, and failing to effectively utilize the airflow to enhance the user's immersive experience. Although the external wind force simulation device can provide wind force feedback, it is bulky, complex to install, and has a low integration degree with the VR system, and cannot accurately adjust the airflow parameters according to the user's real-time behavior. These technical solutions cannot accurately perceive and analyze the complex behaviors of users in a virtual environment, and the generated airflow feedback often lags behind the user's actions, resulting in asynchronous visual and wind feeling experiences, and even causing sensory conflicts. At the same time, there are obvious differences in the sensitivity of different users to airflow perception. The existing systems lack personalized adaptation capabilities and are difficult to meet the diverse user needs. Therefore, how to construct a multi-dimensional airflow feedback system that can real-time perceive, accurately analyze complex user behavior patterns, and generate airflow feedback that conforms to human physiological expectations has become an urgent technical problem to be solved. Summary of the Invention

[0004] The main purpose of the present invention is to construct a multi-dimensional airflow feedback system that can real-time perceive, accurately analyze complex user behavior patterns, and generate airflow feedback that conforms to human physiological expectations.

[0005] The first aspect of the present invention provides a VR environment airflow feedback method based on user behavior. The VR environment airflow feedback method based on user behavior includes: Obtain user behavior data and environmental event data of a user in a VR environment, perform event processing according to the priority attribute of the environmental event data, and obtain an event processing result; Analyze the scene changes in the VR environment according to the event processing result and the user behavior data, generate scene prediction data, and determine initial airflow parameters according to the scene prediction data; Control the distributed fan array according to the initial airflow parameters, adjust the airflow characteristics, and generate a basic airflow effect; Generate user airflow perception characteristics based on the user's perception feedback of the basic airflow effect, and optimize the initial airflow parameters based on the user airflow perception characteristics to obtain optimized airflow parameters; Control the distributed fan array according to the optimized airflow parameters to generate an airflow feedback signal.

[0006] Preferably, obtaining the user behavior data and environmental event data of the user in the VR environment, and performing event processing according to the priority attributes of the environmental event data to obtain an event processing result, including: Obtain the user's head movement parameters, hand movement data, and historical behavior records, and combine the head movement parameters, hand movement data, and historical behavior records through time series correlation operations to form user behavior data; Obtain natural environmental event data, interaction event data, and game logic event data in the virtual environment, combine the natural environmental event data, interaction event data, and game logic event data through data merging operations to form environmental event data, and perform spatio-temporal attribute extraction on the environmental event data through spatial coordinate extraction and timestamp extraction to generate event spatio-temporal data; Perform quantization calculations on the importance and urgency of each event in the environmental event data, and generate event priority attributes according to the results of the quantization calculations of the importance and urgency; Perform a priority sorting operation on the environmental event data according to the quantization calculation results in the event priority attributes to form an event priority sequence; Perform priority weighting processing and spatio-temporal information fusion processing on the environmental event data based on the event priority sequence and the event spatio-temporal data to generate an event processing result containing event priority information and event spatio-temporal information.

[0007] Preferably, analyzing the scene changes in the VR environment according to the event processing result and the user behavior data, generating scene prediction data, and determining the initial airflow parameters according to the scene prediction data, including: According to the event priority information and event spatio-temporal information in the event processing result, extract physical event parameters, interaction event parameters, and game event parameters through spatial situation analysis and time series correlation analysis, and combine the physical event parameters, interaction event parameters, and game event parameters through parameter vectorization operations to generate scene prediction source data; Perform temporal pattern recognition operations and behavioral law extraction operations on the historical behavior records in the user behavior data to generate a scenario response pattern and a behavioral change law, and perform data reconstruction on the scenario prediction source data through pattern matching operations and law mapping operations to generate user scenario prediction data; Analyze the scenario change time series based on the user scenario prediction data through temporal probability distribution calculation and state prediction operations to generate scenario prediction data, where the scenario prediction data includes an event prediction time, an event occurrence location, and an event influence range; Generate initial airflow parameters including an airflow direction parameter, an airflow intensity parameter, and an airflow duration parameter by performing spatio-temporal parameter mapping matrix operations on the event prediction time, the event occurrence location, and the event influence range in the scenario prediction data.

[0008] Preferably, the analyzing the scenario change time series based on the user scenario prediction data through temporal probability distribution calculation and state prediction operations to generate scenario prediction data includes: Perform in-depth probability analysis on the scenario response pattern in the user scenario prediction data, and generate a scenario type prediction probability matrix according to the response probability distribution of the user in different VR scenario types; Perform type recognition operations on the behavioral characteristics of the user in high-altitude scenarios, underwater scenarios, and land scenarios according to the scenario type prediction probability matrix to generate scenario transition prediction data; Based on the scenario transition prediction data, calculate the scenario transition time and transition area through a scenario boundary recognition algorithm to generate scenario transition prediction parameters; According to the scenario transition prediction parameters, perform a fitting operation on the behavioral change trend of the user before and after the scenario transition to generate a user behavior prediction curve, and perform spatio-temporal mapping operations on the user behavior prediction curve and the scenario transition prediction parameters to generate scenario prediction data.

[0009] Preferably, the controlling the distributed fan array according to the initial airflow parameters to adjust the airflow characteristics to generate a basic airflow effect includes: According to the airflow direction parameter and the airflow intensity parameter in the initial airflow parameters, calculate the power distribution coefficients of the fans in the distributed fan array through a power distribution algorithm, and dynamically allocate the rotational speeds of the fans through a rotational speed dynamic optimization algorithm to generate a fan control sequence; According to the airflow duration parameter in the initial airflow parameters and the fan control sequence, generate a fan start-stop time difference and a power change curve through a temporal optimization algorithm, and dynamically adjust the start timing and power change of each fan through a power curve fitting operation to generate a pulse airflow control signal; According to the pulse air flow control signal, the eddy current intensity, diffusion angle and pulse frequency of the air flow are adjusted by eddy current intensity calculation, diffusion angle optimization and pulse frequency modulation operations to generate air flow adjustment parameters. The air flow adjustment parameters are matched with the real-world air flow parameter model in the preset air flow characteristic library through a feature matching algorithm to generate air flow characteristic matching data; According to the air flow characteristic matching data, multiple fans of the distributed fan array are spatially grouped through an annular distribution optimization algorithm, and the power and timing of each group of fans are jointly controlled through a group collaborative control algorithm to generate a basic air flow effect including an air flow direction value, an air flow intensity value and an air flow duration value.

[0010] Preferably, generating a user air flow perception feature according to the user's perception feedback on the basic air flow effect, and optimizing the initial air flow parameters based on the user air flow perception feature to obtain optimized air flow parameters, including: Obtain the wind direction recognition data and wind force perception data of the user for the basic air flow effect, calculate the deviation value between the wind direction recognition data and the standard wind direction and the ratio between the wind force perception data and the standard wind force, and generate a wind direction perception compensation coefficient and a wind force scaling factor; Obtain the user's head turning reaction data and action adjustment data, calculate the temporal change value of the user's head turning reaction data and the amplitude change value of the action adjustment data, and generate a user air flow adaptation curve; Perform a weighted fusion operation on the wind direction perception compensation coefficient, the wind force scaling factor and the user air flow adaptation curve to generate a user air flow perception feature; Perform a vector operation on the wind direction perception compensation coefficient and the air flow direction parameter in the initial air flow parameters, and perform a multiplication operation on the wind force scaling factor and the air flow intensity parameter in the initial air flow parameters to generate air flow reference parameters; Perform parameter mapping operations on the temporal change value and the amplitude change value in the user air flow adaptation curve with the air flow reference parameters respectively to generate optimized air flow parameters including a corrected air flow direction parameter, an adjusted air flow intensity parameter and an optimized air flow duration parameter.

[0011] Preferably, controlling the distributed fan array according to the optimized air flow parameters to generate an air flow feedback signal, including: According to the corrected air flow direction parameter, the adjusted air flow intensity parameter and the optimized air flow duration parameter in the optimized air flow parameters, generate multi-channel control parameters through a piecewise linear mapping operation, and perform channel allocation on the multi-channel control parameters through a channel optimization algorithm to generate fan control parameters; Calculate the triggering moment of the visual effect in the virtual environment through the state prediction algorithm, and perform a time-delay compensation operation on the fan control parameters based on the triggering moment of the visual effect to generate an air flow synchronization control parameter; According to the air flow synchronization control parameter, perform a priority assignment on the operating states of the fans in the distributed fan array through the resource dynamic allocation algorithm, and adjust the resource allocation weights of the fans based on the user's gaze direction data to generate a fan operation sequence; Verify the data of the fan operation sequence with the standard air flow sequence in the preset air flow parameter library through the cross-validation algorithm, and dynamically calibrate the fan operation sequence through the deviation correction algorithm to generate an air flow feedback signal.

[0012] The second aspect of the present invention provides a VR environment air flow feedback device based on user behavior, and the VR environment air flow feedback device based on user behavior includes: A behavior perception module, configured to obtain user behavior data and environmental event data of a user in a VR environment, perform event processing according to the priority attributes of the environmental event data, and obtain an event processing result; A scene prediction module, configured to analyze the scene changes in the VR environment according to the event processing result and the user behavior data, generate scene prediction data, and determine an initial air flow parameter according to the scene prediction data; An air flow control module, configured to control a distributed fan array according to the initial air flow parameter, adjust air flow characteristics, and generate a basic air flow effect; A parameter optimization module, configured to generate a user air flow perception feature according to the user's perception feedback on the basic air flow effect, and optimize the initial air flow parameter based on the user air flow perception feature to obtain an optimized air flow parameter; A feedback generation module, configured to control the distributed fan array according to the optimized air flow parameter to generate an air flow feedback signal.

[0013] The third aspect of the present invention provides a VR environment air flow feedback device based on user behavior, including: a memory and at least one processor, wherein instructions are stored in the memory, and the memory and the at least one processor are interconnected through a line; the at least one processor calls the instructions in the memory so that the VR environment air flow feedback device based on user behavior executes the steps of the above-mentioned VR environment air flow feedback method based on user behavior.

[0014] The fourth aspect of the present invention provides a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, and when it runs on a computer, it enables the computer to execute the steps of the above-mentioned VR environment air flow feedback method based on user behavior.

[0015] The technical solution provided by the embodiments of this application starts from the collection and processing of user behavior data and environmental event data. The system collects multi-dimensional behavior data such as the user's head movement and hand actions, and at the same time obtains information on natural environmental events, interaction events, and game logic events in the virtual environment. The system evaluates the priorities of these environmental events and sorts them according to their importance and urgency. Through this data collection and event processing mechanism, the system overcomes the limitation of insufficient information acquisition in traditional airflow feedback systems, laying a data foundation for subsequent precise airflow control.

[0016] Subsequently, based on the processed event information and user behavior data, the system analyzes the scene changes in the VR environment, generates scene prediction data, and determines the initial airflow parameters. Different from traditional systems, this step realizes predictive analysis - by identifying the temporal pattern of the user's historical behavior and extracting behavior rules, the system can predict the user's possible reactions in a specific scenario. This prediction mechanism generates scene prediction data including the event prediction time, occurrence location, and influence range through temporal probability distribution calculation and state prediction operations. The prediction data is converted into airflow direction, intensity, and duration parameters through a spatio-temporal parameter mapping matrix. This predictive analysis mechanism solves the problem of lagging airflow feedback in traditional systems, enabling the airflow feedback to occur synchronously with the user's behavior and environmental changes, or even slightly in advance, enhancing the spatio-temporal synchronization.

[0017] Next, the system controls the distributed fan array according to the initial airflow parameters, adjusts the airflow characteristics to generate a basic airflow effect. The system calculates the power coefficients of each fan through a power distribution algorithm, allocates the rotational speeds of each fan through a rotational speed dynamic optimization algorithm, and generates an accurate fan control sequence. The system performs multi-dimensional adjustment on the eddy current intensity, diffusion angle, and pulse frequency of the airflow, and matches the adjustment parameters with the real-world airflow model through a feature matching algorithm to ensure that the generated airflow effect meets the user's natural expectations. The system also realizes the generation of complex airflow effects through annular distribution optimization and group collaborative control. This multi-dimensional and refined airflow control mechanism breaks through the limitation of single wind force feedback in traditional systems and realizes a rich variety of airflow experiences.

[0018] The system then collects the user's perceptual feedback on the basic airflow effect, generates the user's airflow perception characteristics, and optimizes the airflow parameters based on these characteristics. The system calculates the deviation value between the user's wind direction recognition and the standard wind direction and the ratio of the wind force perception to the standard wind force, and generates personalized wind direction perception compensation coefficients and wind force scaling factors. The system also generates the user's airflow adaptation curve by analyzing the user's head turning reaction and action adjustment data. These personalized parameters are subjected to vector operations and product operations with the initial airflow parameters to generate optimized airflow parameters. This personalized adaptation mechanism solves the problem of different users' differences in airflow perception, ensuring that each user can obtain an airflow experience that conforms to their personal perception characteristics.

[0019] Finally, the system controls the distributed fan array according to the optimized airflow parameters to generate the final airflow feedback signal. The system generates multi-channel control parameters through piecewise linear mapping, calculates the visual effect trigger moment through the state prediction algorithm, and compensates for the time delay of the airflow control to ensure the precise synchronization of the airflow and visual perception. The system also dynamically adjusts the fan resource allocation weights according to the user's gaze direction to optimize resource utilization. Finally, cross-validation and deviation correction are performed to ensure the accuracy of the airflow feedback. This intelligent airflow output control mechanism solves the problem of the out-of-sync of airflow feedback and vision and hearing in traditional systems, providing a highly coordinated multi-sensory experience. Brief Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on the structures shown in these drawings.

[0021] Figure 1 It is a schematic diagram of an embodiment of the VR environment airflow feedback method based on user behavior in an embodiment of the present invention; Figure 2 It is a schematic diagram of an embodiment of the VR environment airflow feedback device based on user behavior in an embodiment of the present invention; Figure 3 It is a schematic diagram of an embodiment of the VR environment airflow feedback device based on user behavior in an embodiment of the present invention.

[0022] The realization, functional characteristics, and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiments

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0024] It should be noted that if there are directional indications (such as up, down, left, right, front, back...) involved in the embodiments of the present invention, the directional indications are only used to explain the relative position relationship and movement conditions between components in a specific posture. If the specific posture changes, the directional indications will also change accordingly.

[0025] In addition, the descriptions involving "first", "second", etc. in the present invention are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In addition, "and / or" throughout the text includes three scenarios. Taking A and / or B as an example, it includes the technical solution of A, the technical solution of B, and the technical solution that both A and B are satisfied at the same time. In addition, the technical solutions between various embodiments can be combined with each other, which must be based on the ability of those of ordinary skill in the art to implement. When the combination of technical solutions conflicts with each other or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present invention.

[0026] An embodiment of the present application provides a method for airflow feedback in a VR environment based on user behavior. Figure 1 It is a flowchart of a method for airflow feedback in a VR environment based on user behavior provided by an embodiment of the present application. In this embodiment, the method includes: Please refer to Figure 1 , obtain user behavior data and environmental event data of the user in the VR environment, perform event processing according to the priority attribute of the environmental event data, and obtain an event processing result; In an embodiment of the present invention, the obtaining user behavior data and environmental event data of the user in the VR environment, performing event processing according to the priority attribute of the environmental event data, and obtaining an event processing result includes: Obtain user head movement parameters, hand movement data, and historical behavior records, and combine the head movement parameters, hand movement data, and historical behavior records through time series correlation operations to form user behavior data; Obtain natural environment event data, interaction event data, and game logic event data in a virtual environment, combine the natural environment event data, interaction event data, and game logic event data through data merging operations to form environment event data, and perform spatio-temporal attribute extraction on the environment event data through spatial coordinate extraction and timestamp extraction to generate event spatio-temporal data; Perform quantization calculations on the importance and urgency of each event in the environment event data, and generate event priority attributes based on the results of the quantization calculations of the importance and urgency; Perform a priority sorting operation on the environment event data according to the quantization calculation results in the event priority attributes to form an event priority sequence; Perform priority weighting processing and spatio-temporal information fusion processing on the environment event data based on the event priority sequence and the event spatio-temporal data to generate an event processing result containing event priority information and event spatio-temporal information.

[0027] The following specifically describes the steps involved in the above embodiments: Obtaining the user's head movement parameters, hand movement data, and historical behavior records is achieved through the sensor system built into the VR headset. The head movement parameters are collected by the six-axis gyroscope and accelerometer integrated in the headset, including pitch angle, yaw angle, roll angle, and their corresponding angular velocity and angular acceleration values, with a sampling frequency of over 90Hz to ensure the accuracy of motion capture. The hand movement data is collected by the motion sensors and touchpad built into the VR controller, recording information such as hand position, rotation angle, and grasping force. The historical behavior records are stored in the system cache, containing the user's behavior data from the past 30 seconds to 5 minutes. The time-series correlation operation correlates the head movement parameters, hand movement data, and historical behavior records within the same time period through the time window sliding algorithm. Specifically, when implemented, the system normalizes the multi-dimensional data within a 200ms data window, uses the timestamp alignment technique to ensure data time-series consistency, and finally forms comprehensive user behavior data containing time-series characteristics. For example, when the user turns their head to look at an enemy behind in a VR game and waves their arm for a defensive action, the system will synchronously record the angular velocity of the rapid head rotation, the position change trajectory of the arm, and correlate it with the historical action sequence to identify this as a "defensive reaction" behavior pattern. This multi-dimensional behavior data capture and time-series correlation processing ensure that the system can accurately understand the user's complex behavior patterns and provide accurate behavior basis for subsequent airflow feedback.

[0028] Obtaining event data in the virtual environment is achieved through the communication interface with the VR application. Natural environment event data includes physical phenomena such as storms, explosions, and water flows in the virtual world, which is obtained by reading the status data of the game physics engine; interaction event data includes operations such as collisions, grabs, and throws between the user and virtual objects, which is obtained by listening to the interaction event system of the game engine; game logic event data such as enemy attacks, flight acceleration, teleportation, etc. for specific scenario events is obtained through the game logic interface. The data merging operation uses an event standardization framework to convert event data from different sources into a unified data structure, including attributes such as event type, triggering object, and impact parameters. Spatial coordinate extraction extracts event occurrence location, impact range, and direction information from the event data through a three-dimensional coordinate mapping algorithm; timestamp extraction records the start time, duration, and end time of the event. For example, when an explosion event occurs in the game, the system records the center coordinates of the explosion (x = 45.6, y = 2.1, z = 30.2), the impact radius of 8 meters, the occurrence time point at T + 2.3 seconds, and the duration of 0.8 seconds. This comprehensive collection of environmental event data and extraction of spatio-temporal attributes enables the system to accurately grasp the characteristics of various events in the virtual environment, providing an environmental basis for the generation of precise airflow effects.

[0029] The priority processing of environmental event data adopts a multi-dimensional evaluation method. The quantification calculation of importance is based on the impact degree of the event on the user experience. Using the impact factor scoring method, according to factors such as event type (physical event, interaction event, game event), impact range (local impact, medium range, global impact), and impact intensity (slight, medium, strong), an importance score of 0 - 100 is calculated. The quantification calculation of urgency is based on the time characteristics of the event, considering the time interval from the event occurrence to its impact on the user, duration, and change rate, and also generates an urgency score of 0 - 100. For example, the importance score of the sudden enemy attack event is 85 (high impact intensity, medium impact range), and the urgency score is 90 (extremely short time interval); while the importance score of the gentle breeze blowing event in the environment is 30 (slight impact intensity, global impact), and the urgency score is 20 (long duration). This step enables the system to distinguish the priorities of different events through quantitative evaluation of the impact characteristics of the events, ensuring that critical events are processed in a timely manner, thus achieving an orderly airflow feedback control in the complex and changeable virtual environment.

[0030] The priority sorting operation adopts a weighted sorting algorithm, which calculates the weighted sum of the importance score and the urgency score according to the ratio of 7:3 to generate a comprehensive priority score. The system sorts the environmental event data in descending order according to the comprehensive priority score to form an event priority sequence. For example, when the system simultaneously detects three events: "Enemy attack" (comprehensive priority 86.5), "Strong wind area" (comprehensive priority 65.2), and "Object floating" (comprehensive priority 42.1), the system will process these events in the order of enemy attack, strong wind area, and object floating. This priority sorting based on quantitative calculation ensures that the system can reasonably allocate processing resources when facing multiple concurrent events and prioritize responding to the events that have the greatest impact on the user experience.

[0031] Priority weighting processing and spatio-temporal information fusion processing are processes that combine the event priority sequence with the event spatio-temporal data. Priority weighting processing assigns processing weights to events with different priorities through an exponential decay function. High-priority events obtain higher weights to ensure the dominant airflow effect. Spatio-temporal information fusion processing uses spatial clustering and time correlation algorithms to perform correlation analysis on events that are close in space and time, and calculates the comprehensive influence effect through vector synthesis methods. For example, when two events, "Explosion shock" and "Airflow disturbance", are close in spatial position (distance < 5 meters) and overlap in time (overlap > 0.5 seconds), the system regards these two events as associated events and comprehensively calculates their direction, intensity, and duration parameters. The final generated event processing result includes a sorted event list, the priority weights of each event, spatio-temporal characteristic parameters, and a description of the association relationship. This fusion processing of priority and spatio-temporal information ensures that the system can comprehensively consider the correlation and impact differences between events, provides a structured event processing result for the subsequent generation of airflow parameters, and thus can generate airflow effects that conform to physical laws and user cognitive expectations.

[0032] Please continue to refer to Figure 1 , analyze the scene changes in the VR environment according to the event processing result and the user behavior data, generate scene prediction data, and determine the initial airflow parameters according to the scene prediction data; In one embodiment of the present invention, the step of analyzing the scene changes in the VR environment according to the event processing result and the user behavior data, generating scene prediction data, and determining the initial airflow parameters according to the scene prediction data includes: According to the event priority information and event spatio-temporal information in the event processing result, extract physical event parameters, interaction event parameters, and game event parameters through spatial situation analysis and time series correlation analysis, and combine the physical event parameters, interaction event parameters, and game event parameters through parameter vectorization operations to generate scene prediction source data; Perform temporal pattern recognition operations and behavior pattern extraction operations on the historical behavior records in the user behavior data to generate a scenario response pattern and a behavior change pattern, and perform data reconstruction on the scenario prediction source data through pattern matching operations and rule mapping operations to generate user scenario prediction data; Analyze the scenario change time series based on the user scenario prediction data through temporal probability distribution calculation and state prediction operations to generate scenario prediction data, where the scenario prediction data includes an event prediction time, an event occurrence location, and an event influence range; Generate initial airflow parameters including an airflow direction parameter, an airflow intensity parameter, and an airflow duration parameter by performing spatio-temporal parameter mapping matrix operations on the event prediction time, the event occurrence location, and the event influence range in the scenario prediction data.

[0033] The following specifically describes the steps involved in the above embodiments: Extracting parameters based on the event processing results is achieved through a refined analysis method. Spatial situation analysis uses a three-dimensional spatial clustering algorithm to process the location, scope, and directionality of the event occurrence, construct a spatial impact map of the virtual environment, and identify key areas and impact paths. Temporal correlation analysis, on the other hand, analyzes the time sequence, interval, and persistence of the event occurrence through time series analysis techniques to identify causal relationships and patterns in time. The system extracts three types of event parameters: physical event parameters (such as wind speed magnitude, wind direction angle, turbulence intensity), interaction event parameters (such as collision force, operation speed, interaction duration), and game event parameters (such as damage value, event trigger condition, game state change). Parameter vectorization operations convert these parameters into standardized numerical vectors, with each dimension representing a specific physical or logical attribute. For example, in a VR flight simulation scenario, the physical event "airflow disturbance" extracted by the system includes a wind speed magnitude of 12.5 m / s, an azimuth angle of 45° (the angle relative to the user's direct front on the horizontal plane), an elevation angle of -10° (the angle relative to the horizontal plane, with a negative value indicating downward), a duration of 2.3 seconds, and a disturbance frequency of 1.2 Hz; the interaction event "pushing the control lever" includes an operation speed of 0.8 m / s, a deflection angle of 25°, and a continuous force of 75%; the game event "airplane acceleration" includes an acceleration rate of 6.5 m / s² and a state change from uniform speed to acceleration. After these parameters are vectorized, they form a high-dimensional data structure describing the complete scenario state, that is, the scenario prediction source data. This multi-dimensional parameter extraction and vectorization process can comprehensively capture the physical characteristics and event features in the virtual environment, providing accurate data support for accurately simulating various airflow effects.

[0034] The analysis of historical behavior records adopts a method that combines deep learning and statistical analysis. The time-series pattern recognition operation uses the sliding window technique and recurrent neural network (RNN) to process the time-series data of user behavior, and identifies recurring behavior patterns, such as typical reaction sequences like suddenly stopping - turning the head - stepping back. The behavior pattern extraction operation extracts the common reaction patterns of users to specific scenario stimuli through association rule mining and frequency analysis, such as "when facing an object approaching at high speed, users have a 75% probability of choosing to dodge sideways". The results of these two operations respectively form a scenario reaction pattern (a behavior sequence template under specific circumstances) and a behavior change pattern (the reaction probability distribution under different stimuli). Subsequently, the system calculates the similarity between the user's current behavior and the identified reaction patterns through pattern matching operations, and finds the most matching behavior pattern; through rule mapping operations, it conducts correlation analysis between the current scenario conditions and historical behavior patterns to predict possible behavior reactions. These analysis results are integrated with the scenario prediction source data to generate user scenario prediction data containing personalized reaction predictions. For example, the system analyzes that when an "explosion event" occurs, a certain user has an 83% probability of performing a behavior sequence of "turning around backward - stepping back 3 steps - crouching down with the head down", and the reaction time is on average 0.65 seconds. This personalized behavior analysis and prediction achieve personalized customization of airflow feedback, greatly improving the system's adaptability to different user habits and solving the problem of lack of personalized adaptation in existing technologies.

[0035] The scene change time-series analysis is based on probability statistical models and state prediction techniques. The time-series probability distribution calculation uses a Bayesian network model to calculate the probability distribution of specific events occurring at different future time points according to the current scene state and historical change patterns. The state prediction operation predicts the movement trajectories and state changes of objects, characters, and environmental elements in the scene through Markov decision processes and forward dynamics models. The system synthesizes the results of these two calculations to generate scene prediction data containing three key elements: the event prediction moment (the time point of event occurrence accurate to the millisecond level), the event occurrence location (three-dimensional coordinate values), and the event influence range (the size and direction of a spherical or ellipsoidal influence area). For example, in a VR shooting game, the system predicts that the "enemy rocket explosion" event will occur 1.35 seconds after the current time point at the coordinates (30.2, 4.5, 15.8), and the influence range is a spherical area with a radius of 8.5 meters. This precise scene prediction technology solves the time lag problem in traditional airflow feedback systems, enabling airflow feedback to be precisely synchronized with visual and auditory perceptions, and even triggering at the microsecond level in advance, thus eliminating the temporal conflicts between multiple senses and significantly enhancing the user's sense of immersion.

[0036] The method of converting scenario prediction data into airflow parameters combines physical simulation and perceptual psychology. The spatio-temporal parameter mapping matrix is a multi-dimensional transformation matrix designed based on the principles of fluid mechanics and human wind perception characteristics, which maps the spatio-temporal characteristics of events onto the physical parameters of airflow. The specific operations are as follows: Convert the event prediction time into the airflow trigger time point and the gradient curve; convert the azimuth and elevation angles of the event occurrence position relative to the user into the airflow direction parameters (horizontal angle and vertical angle); convert the event influence range and intensity into the airflow intensity parameter (wind force percentage) and the airflow duration parameter (millisecond timing value). For example, the predicted "explosion" event is mapped to: the airflow direction is 35° horizontal angle and 10° vertical angle relative to the user's face, the intensity is 75% of the maximum wind force, the duration is 850 milliseconds, and the gradient curve is "rapid rise - plateau - slow decline". This parameter mapping based on the physical model makes the airflow feedback highly consistent with the virtual event in physical characteristics. Users can accurately understand the nature, direction, and intensity of events in the virtual environment through airflow perception, greatly improving the spatial positioning ability and immersive feeling of the VR experience, and at the same time avoiding the sensory conflict problem caused by the asynchronous airflow feedback and visual perception in traditional systems.

[0037] In an embodiment of the present invention, the analysis of the scenario change time sequence by calculating the time sequence probability distribution and performing the state prediction operation according to the user scenario prediction data to generate the scenario prediction data includes: Perform in-depth probability analysis on the scenario response patterns in the user scenario prediction data, and generate a scenario type prediction probability matrix according to the response probability distribution of the user in different VR scenario types; According to the scenario type prediction probability matrix, perform type recognition operations on the behavioral characteristics of the user in high-altitude scenarios, underwater scenarios, and land scenarios to generate scenario transition prediction data; Based on the scenario transition prediction data, calculate the scenario transition moment and transition area through the scenario boundary recognition algorithm to generate scenario transition prediction parameters; According to the scenario transition prediction parameters, perform a fitting operation on the trend of the user's behavioral changes before and after the scenario transition to generate a user behavior prediction curve, and perform a spatio-temporal mapping operation on the user behavior prediction curve and the scenario transition prediction parameters to generate scenario prediction data.

[0038] The following specifically describes the steps involved in the above embodiments: The in-depth probability analysis of user scenario prediction data is achieved through hierarchical Bayesian calculation and conditional probability statistics. First, the system classifies the user's scenario response patterns according to different VR scenario types, mainly including three categories: high-altitude scenarios, underwater scenarios, and land scenarios. Each category is further divided into multiple sub-categories (such as cliff edges, building tops, flight states, etc. in high-altitude scenarios). Then, the system statistically analyzes the occurrence frequency and conditional probability of the response patterns of users in each scenario type, and establishes a two-way association probability table of user behavior - scenario type. Using the maximum a posteriori probability estimation method, the system calculates the probability value that a given behavior sequence belongs to a specific scenario type, forming an M×N scenario type prediction probability matrix, where M represents the number of recognized behavior patterns and N represents the number of scenario types. For example, in the VR roller coaster experience, the system finds that the association probability between the user behavior pattern "grip the armrest - lean forward - take a deep breath" and the "about to fall" scenario is 0.92, while the association probability with the "smooth movement" scenario is only 0.15. This in-depth probability analysis enables the system to accurately infer the current and upcoming scenario types from user behavior, provides scenario-related context information for airflow feedback, and significantly enhances the system's understanding ability of the user's current immersion state.

[0039] The scenario type recognition operation is implemented based on multi-feature classification and feature weight adaptation methods. The system extracts the high-confidence scenario type prediction results from the scenario type prediction probability matrix, and sets feature recognition criteria for high-altitude scenarios, underwater scenarios, and land scenarios respectively. For high-altitude scenarios, the system focuses on analyzing features such as the frequency of the user's head moving up and down, the change in the line-of-sight pitch angle, and the degree of hand clenching; for underwater scenarios, the system pays attention to features such as the change in breathing rhythm, the slowdown of movement speed, and the adjustment of body balance; for land scenarios, the system focuses on features such as gait patterns, turning behaviors, and interaction frequencies. Through the weight calculation of feature vectors and multi-data fusion, the system determines the current and upcoming scenario types to be converted, and generates scenario conversion prediction data including the scenario type, conversion probability, and estimated conversion time. For example, in a VR adventure game, when the user switches from a land scenario to an underwater scenario, the system detects behavioral features such as a slowdown in gait change, looking up, and taking a deep breath, and calculates that the scenario conversion probability is 87%, and it is expected to complete the conversion after 3.2 seconds. This scenario type recognition technology solves the problem of large differences in airflow feedback requirements in different scenarios, enabling the system to dynamically adjust the airflow parameter standards according to the scenario characteristics and provide users with a more physically intuitive airflow experience.

[0040] The scene boundary recognition algorithm is achieved through spatial topology analysis and scene transition feature extraction. First, the system identifies the boundary regions of different scene types in the three-dimensional space of the virtual environment, such as water surfaces, cliff edges, cave entrances, etc., and establishes a mapping table of boundary transition regions. Then, based on the user's current position, moving speed, and direction, combined with the scene transition prediction data, the system calculates the estimated time for the user to reach the scene boundary (scene transition moment) and the specific coordinates of the transition point position (transition region). The system uses a dynamic time window algorithm to process boundary determination. The window size is adaptively adjusted according to the user's moving speed. A small window of 0.5 - 1 second is used when the moving speed is fast, and a large window of 1 - 2 seconds is used when the moving speed is slow, ensuring the accuracy and real-time nature of boundary determination. The generated scene transition prediction parameters include the transition start time point, transition duration, pre-transition scene type, post-transition scene type, and transition trajectory coordinate sequence. For example, the system predicts that the user will reach the water surface boundary 2.5 seconds after the current time point, the transition duration is 1.2 seconds, transitioning from the land scene to the underwater scene, and the transition trajectory is a vertical descent from coordinates (25, 3, 40) to coordinates (25, 0, 44). This scene boundary recognition technology solves the problem of poor coherence of airflow feedback during the scene transition, enabling the system to accurately predict and prepare the corresponding airflow change sequence before the scene transition, ensuring a smooth change in airflow feedback during the transition process.

[0041] The generation and spatio-temporal mapping of the user behavior prediction curve are achieved through trend fitting and parameter correlation. The system uses polynomial regression or spline interpolation methods to perform curve fitting on the historical behavior data of the user before and after the scene transition, obtaining a continuous function representing the variation of the user behavior parameters over time, that is, the user behavior prediction curve. These behavior parameters include key indicators such as head position coordinates, steering angle, movement speed, and interaction status. Then, the system performs spatio-temporal mapping operations on the user behavior prediction curve and the scene transition prediction parameters, calculates the expected position, orientation, and status of the user at each key time point during the scene transition, and correlates the corresponding environmental events and physical effects. The finally generated scene prediction data includes a complete sequence of event prediction moments, three-dimensional coordinates of the event occurrence locations, and dynamic change range parameters of the influence. For example, in the scene transition from land to water, the system predicts that the user will have a downward acceleration behavior 0.5 seconds before contacting the water surface, the head will tilt forward by 25° at the moment of contacting the water surface, and deceleration and buoyancy perception behaviors will start to appear 0.3 seconds after entering the water. Based on this, the system generates a complete prediction sequence of water splash events (t = 0s, coordinates (25, 0, 44)), water entry resistance events (from t = 0.1s to 0.5s, the influence range gradually expands from the head to the whole body), and underwater buoyancy events (starting from t = 0.5s, the whole body range). This comprehensive prediction technology enables the system to plan a complete sequence of airflow effects in advance, realizing a highly coherent and physically reasonable airflow feedback experience, and solving the problem of airflow feedback faults or incoordination during scene transitions in the prior art.

[0042] Please continue to refer to Figure 1 , according to the initial airflow parameters, control the distributed fan array, adjust the airflow characteristics, and generate a basic airflow effect; In an embodiment of the present invention, the controlling the distributed fan array according to the initial airflow parameters, adjusting the airflow characteristics, and generating a basic airflow effect includes: According to the airflow direction parameter and the airflow intensity parameter in the initial airflow parameters, calculate the power distribution coefficients of the fans in the distributed fan array through a power distribution algorithm, and dynamically allocate the rotation speeds of the fans through a rotation speed dynamic optimization algorithm to generate a fan control sequence; According to the airflow duration parameter in the initial airflow parameters and the fan control sequence, generate the fan start-stop timing difference and the power change curve through a timing optimization algorithm, and dynamically adjust the start timing and power change of each fan through power curve fitting operations to generate a pulse airflow control signal; According to the pulsed air flow control signal, the eddy current intensity, diffusion angle, and pulse frequency of the air flow are adjusted by eddy current intensity calculation, diffusion angle optimization, and pulse frequency modulation operations to generate air flow adjustment parameters. The air flow adjustment parameters are matched with the real-world air flow parameter model in the preset air flow characteristic library through a feature matching algorithm to generate air flow characteristic matching data; According to the air flow characteristic matching data, the multiple fans of the distributed fan array are spatially grouped by an annular distribution optimization algorithm, and the power and timing of each group of fans are linked and controlled by a group collaborative control algorithm to generate a basic air flow effect including an air flow direction value, an air flow intensity value, and an air flow duration value.

[0043] The following is a specific description of the steps involved in the above embodiments: The process of the power distribution algorithm for processing the initial air flow parameters adopts a vector decomposition and weight mapping mechanism. The system first reads the air flow direction parameters (horizontal angle and vertical angle) and air flow intensity parameters (wind force percentage value) in the initial air flow parameters, and converts the air flow direction parameters into three-dimensional unit vectors. Then, the system calculates the angle deviation of each fan relative to the target air flow direction according to the spatial layout of the fans in the VR headset, and calculates the direction contribution coefficient of each fan through cosine similarity. The direction contribution coefficient of a fan is 1.0 when its orientation is completely consistent with the target air flow direction, decreases as the angle deviation increases, and is 0 when the angle deviation exceeds 60 degrees. The air flow intensity parameter is linearly mapped to the fan power reference value (in the range of 0-100%). The system multiplies the direction contribution coefficient by the power reference value to obtain the initial power distribution coefficient of each fan. The rotational speed dynamic optimization algorithm further fine-tunes the initial power distribution coefficient by considering the air flow interference and energy consumption balance between the fans. This algorithm calculates the spatial relationship between the fans. Fans that are close to each other and have similar angles will enhance the air flow effect on each other, while fans with opposite directions will cancel each other out. Based on these relationships, the system dynamically adjusts the power coefficients of each fan, and finally generates a fan control sequence including each fan ID, the target rotational speed value (usually discrete values in the range of 0-5000 RPM), and the rotational speed change rate. For example, when the system needs to simulate a strong wind from a 45-degree angle in front of the user's left, the fans in the left front obtain 90% power distribution, the fans in the front obtain 65% power distribution, while the fans on the right only obtain 10% low power distribution. This precise power distribution and rotational speed optimization mechanism solves the problem of inaccurate air flow direction perception in traditional systems, enabling users to accurately perceive the source direction of the air flow, and significantly enhancing the spatial positioning ability and immersion.

[0044] The process of the timing optimization algorithm for processing the air flow duration parameter adopts a multi-phase collaborative control method. The system decomposes the air flow duration parameter (usually in milliseconds) into a start-up phase, a stable phase, and a decay phase, which account for 20%, 60%, and 20% of the total duration respectively. This proportion allocation is determined based on the perception characteristics of human beings for air flow changes, and can create a natural and smooth air flow perception experience. The timing optimization algorithm calculates the sequence and time difference of fan start and stop according to the power distribution coefficient and spatial layout of each fan in the previous step, and generates the timing difference of fan start and stop. For simulating the near-field air flow effect, the system starts the fans successively in the air flow propagation direction, and the time difference ranges from 30 to 100 milliseconds; for the surrounding air flow effect, the system starts the fans in the order of the circular layout, and the time difference is a fixed 50 milliseconds. The power curve fitting operation describes the change of the power of each fan over time as a continuous curve, including a rising section (acceleration curve), a plateau section (stable power), and a falling section (deceleration curve). The system uses the cubic spline interpolation method to smooth the power change process, ensuring the continuous change of the air flow intensity and avoiding discomfort caused by sudden changes. The finally generated pulse air flow control signal includes a list of fan IDs, the power-time curve data of each fan, and the precise start and stop time points. For example, when simulating an explosion air flow, the control signal generated by the system makes the central fan start first and quickly reach 100% power, and then the surrounding fans start successively and reach powers ranging from 70% to 90%. The whole process lasts for 850 milliseconds. This timing optimization and power curve fitting technology enables the system to create a rich variety of timing effects, such as pulsed, gradual, and wavy air flow patterns, greatly enriching the types of air flow feedback in the VR experience.

[0045] The air flow characteristic adjustment process is achieved through precise physical parameter control and pattern matching. The eddy current intensity calculation is based on the principles of fluid dynamics, and the formation of eddy currents is controlled by adjusting the power difference and start timing difference of adjacent fans. The range of eddy current intensity values is 0 - 1.0, where 0 represents laminar flow (smooth air flow), and 1.0 represents strong turbulence (highly irregular air flow). The diffusion angle optimizes and adjusts the diffusion range of the air flow, which is achieved by controlling the fan combination mode and power distribution. The range of the diffusion angle is 10 - 120 degrees. Small angles are used to simulate concentrated air flows (such as directional winds), and large angles are used to simulate ambient air flows (such as ambient winds). The pulse frequency modulation operation controls the periodic changes of the air flow, and the frequency range is 0 - 5 Hz, where 0 represents a stable air flow, and higher frequencies are used to simulate oscillating air flows. These parameters are calculated to form a set of air flow adjustment parameters. The feature matching algorithm compares these parameters with the templates in the preset air flow characteristic library, which contains the parameter configurations of typical air flow patterns such as "natural breeze", "strong wind", "explosion shock", and "underwater air flow". The system uses the Euclidean distance to calculate the similarity between the parameter set and each template, selects the template with the highest similarity as the basis, and makes adjustments in combination with the current specific requirements to generate air flow characteristic matching data. For example, when the system needs to simulate the effect of "valley breeze", the feature matching algorithm identifies that the closest template is "natural breeze" (eddy current intensity 0.3, diffusion angle 60 degrees, pulse frequency 0.2 Hz), and then makes fine adjustments to match the current scenario requirements. This air flow characteristic adjustment and matching technology enables the system to create air flow effects that meet the user's physical expectations, avoids complex air flow characteristics that cannot be expressed simply by wind speed and direction, and greatly improves the realism and recognition of air flow feedback.

[0046] In the final stage of airflow effect generation, a grouping control and collaborative optimization method is adopted. The annular distribution optimization algorithm matches the eddy current intensity, diffusion angle, and pulse frequency parameters in the data according to the airflow characteristics, and functionally groups the distributed fan arrays. The system divides the fans into three groups: the core group (mainly responsible for directionality and intensity), the auxiliary group (responsible for diffusion and filling), and the modulation group (responsible for timing changes and special effects). The grouping uses a spatial clustering method and is divided according to the physical positions of the fans and the types of airflow contributions. The group collaborative control algorithm implements differentiated but coordinated control strategies for each group of fans according to the grouping results. The fans in the core group use precise power control and synchronous start and stop; the fans in the auxiliary group use gradient power distribution and staggered start and stop; the fans in the modulation group use periodic power changes and specific timing control. Through this hierarchical control structure, the system realizes the precise generation of complex airflow effects, while optimizing energy use and heat generation. The finally generated basic airflow effect includes precise airflow direction values (horizontal and vertical angles, with an accuracy of ±5 degrees), airflow intensity values (percentage values, with an accuracy of ±3%), and airflow duration values (millisecond-level accuracy). For example, in a simulated storm scenario, the basic airflow effect generated by the system includes strong wind with a main direction of (-30°, 15°) (intensity 85%) and a duration of 2300 milliseconds, along with direction fluctuations within ±15 degrees and intensity fluctuations of 5% - 20%, creating a realistic storm effect. This grouping collaborative control technology solves the problem that a single control strategy cannot meet complex airflow requirements, enabling the system to create a highly simulated and diverse airflow experience with limited fan resources, greatly enhancing the expressiveness and immersion of airflow feedback in the VR environment.

[0047] Please continue to refer to Figure 1 , generate user airflow perception characteristics according to the user's perception feedback on the basic airflow effect, and optimize the initial airflow parameters based on the user airflow perception characteristics to obtain optimized airflow parameters; In an embodiment of the present invention, the generating user airflow perception characteristics according to the user's perception feedback on the basic airflow effect, and optimizing the initial airflow parameters based on the user airflow perception characteristics to obtain optimized airflow parameters includes: Obtain the wind direction recognition data and wind force perception data of the user on the basic airflow effect, calculate the deviation value between the wind direction recognition data and the standard wind direction and the ratio between the wind force perception data and the standard wind force, and generate a wind direction perception compensation coefficient and a wind force scaling factor; Obtain the user's head turning reaction data and action adjustment data, calculate the timing change value of the user's head turning reaction data and the amplitude change value of the action adjustment data, and generate a user airflow adaptation curve; Perform a weighted fusion operation on the wind direction perception compensation coefficient, the wind force scaling factor, and the user air flow adaptation curve to generate a user air flow perception feature; Perform a vector operation on the wind direction perception compensation coefficient and the air flow direction parameter in the initial air flow parameters, and perform a multiplication operation on the wind force scaling factor and the air flow intensity parameter in the initial air flow parameters to generate air flow reference parameters; Perform a parameter mapping operation on the time series change value and the amplitude change value in the user air flow adaptation curve with the air flow reference parameters respectively to generate optimized air flow parameters including a corrected air flow direction parameter, an adjusted air flow intensity parameter, and an optimized air flow duration parameter.

[0048] The following specifically describes the steps involved in the above embodiments: Obtaining the user's perception data of the basic air flow effect is achieved through a method combining implicit feedback and explicit testing. The system collects two types of key data: wind direction recognition data and wind force perception data. The wind direction recognition data is obtained by analyzing the user's head turning behavior after receiving air flow stimulation. The system uses the gyroscope sensor built into the head-mounted display to record the change in the user's head orientation and compares it with the expected turning direction. For example, when the system generates an air flow from a 30-degree angle in the front, ideally the user should perceive this direction; but in actual measurement, a certain user may perceive the air flow direction as a 40-degree angle in the front, resulting in a deviation of 10 degrees. The system calculates the deviation value between the wind direction recognition data and the standard wind direction, including the horizontal angle deviation and the vertical angle deviation, with a typical range between ±5 degrees and ±25 degrees. The wind force perception data is obtained by analyzing the reaction amplitude of the user to air flows of different intensities, such as physiological reaction indicators such as the body backward tilt angle and the head elevation height. The system calculates the ratio between the actually perceived wind force value of the user and the output standard wind force value to generate a wind force scaling factor. For example, when the system outputs an air flow with 75% intensity, a certain user may only perceive an effect equivalent to 60% intensity, resulting in a wind force scaling factor of 0.8. The extraction of such personalized perception parameters solves the problem of large perception differences among different users and provides accurate individual characteristic data for subsequent air flow parameter optimization.

[0049] The acquisition and processing of user behavior response data adopt real-time monitoring and time series analysis methods. The head turning response data is collected through the gyroscope and accelerometer built into the headset, recording the head movement trajectory of the user after receiving the airflow feedback, including the turning speed, turning angle, and turning duration. The system uses the sliding window technique (window size: 500 milliseconds) to segment the head turning data, calculates the average turning speed and cumulative angle change within each window, and obtains the time series change values of head turning. The action adjustment data is obtained by analyzing the position change of the user's hand controller and the body posture change, recording the body adjustment actions made by the user to adapt to the airflow stimulation, such as leaning forward, backward, or sideways. The system calculates the amplitude change values of these actions, including displacement distance, angle change, and duration. After time synchronization processing of the two types of data, a user airflow adaptation curve representing the change of the user's response to the airflow over time is formed. For example, under strong airflow stimulation, the user may first have a rapid head turn (peak speed: 40 degrees per second), followed by a slight backward tilt of the body (angle: 8 degrees), and finally a process of gradually adapting and returning to a stable state. The collection and processing of this dynamic response data enable the system to understand the user's adaptation process to the airflow and provide a time series basis for providing a more natural and comfortable airflow experience.

[0050] The generation of user airflow perception characteristics adopts multi-feature fusion technology. The system performs weighted fusion operations on the wind direction perception compensation coefficient, wind force scaling factor, and user airflow adaptation curve, with a weight ratio of 3:3:4. This ratio setting is carefully adjusted based on the different sensitivities of humans to the direction, intensity, and time series changes of the airflow. During the fusion process, the system extracts features from the user airflow adaptation curve to obtain feature parameters including adaptation speed, peak response, and stabilization time. After all features are standardized, a user airflow perception feature vector in a unified format is formed. This feature vector contains the user's perception compensation requirements for the airflow direction, intensity scaling requirements, and time series adaptation characteristics, comprehensively describing the airflow perception characteristics of a specific user. For example, the airflow perception characteristics of a certain user may indicate that their judgment of the horizontal airflow is 15% to the left, their perception of the airflow intensity is 20% lower than the standard, and they have a slower adaptation process to sudden airflow (it takes 1.2 seconds to reach a stable state). The generation of this fused feature solves the problem that traditional systems cannot comprehensively consider the multi-dimensional perception characteristics of users and provides a complete user model for the subsequent precise personalized optimization of airflow parameters.

[0051] The generation of airflow reference parameters adopts vector correction and intensity adjustment methods. The system performs vector operations on the wind direction perception compensation coefficient and the airflow direction parameter in the initial airflow parameters, specifically rotating and adjusting the direction vector according to the compensation coefficient. For example, for a user with a perception bias to the left, the system will shift the airflow direction to the right by an appropriate angle to compensate, ensuring that the wind direction perceived by the user is consistent with the design intention. Figure 1The wind force scaling factor is multiplied by the air flow intensity parameter to linearly adjust the wind force intensity. For example, for users with weak wind force perception (scaling factor of 1.25), the system will increase the air flow with an original intensity of 70% to 87.5%, ensuring that users perceive the expected wind force effect. This parameter adjustment process takes into account the non-linear characteristics of human perception, especially using non-linear mapping functions in the extremely low intensity (<20%) and extremely high intensity (>90%) intervals to avoid over-adjustment or under-adjustment. The generated air flow reference parameters include the direction angle and intensity value after personalized adjustment, laying the foundation for the next step of timing optimization. This vector correction and intensity adjustment mechanism solves the problem of inconsistent air flow experience caused by differences in perception among users, enabling each user to obtain an air flow perception experience that conforms to the design intention.

[0052] The generation of optimized air flow parameters adopts a timing mapping and comprehensive optimization method. The system performs parameter mapping operations on the timing change value and amplitude change value in the user's air flow adaptation curve respectively with the air flow reference parameters. The timing change value is used to adjust the time parameters of the air flow effect, such as the start rate, duration, and decay rate. For users with a slower adaptation speed, the system will extend the gradual change time of the air flow (increase by 20 - 50%), making the air flow change more smoothly; for sensitive users, the system will shorten the start time (reduce by 10 - 30%) to provide faster feedback. The amplitude change value is used to refine and adjust the intensity distribution curve of the air flow, and optimize the time distribution of the air flow intensity according to the peak response and steady-state response of the user. The finally generated optimized air flow parameters include three core components: the corrected air flow direction parameter (considering the user's direction perception deviation), the adjusted air flow intensity parameter (considering the user's intensity perception sensitivity), and the optimized air flow duration parameter (considering the user's adaptation speed and response characteristics). For example, in a VR flight simulation, for a user with a direction perception deviation of +10 degrees, an intensity perception coefficient of 0.85, and a moderate adaptation speed, the system optimizes the original air flow feedback designed as "straight ahead, intensity 75%, duration 1.2 seconds" to "10 degrees to the left, intensity 88%, duration 1.5 seconds, and add a 200-millisecond gradual start section". This comprehensive parameter optimization mechanism ensures that the air flow feedback can adapt to the unique perception characteristics of each user, significantly improving the accuracy and comfort of the air flow feedback, and solving the core problem that standardized air flow parameters cannot meet the needs of different users.

[0053] Please continue to refer to Figure 1 , and control the distributed fan array according to the optimized air flow parameters to generate an air flow feedback signal.

[0054] In an embodiment of the present invention, the controlling the distributed fan array according to the optimized air flow parameters to generate an air flow feedback signal includes: According to the corrected airflow direction parameter, adjusted airflow intensity parameter, and optimized airflow duration parameter in the optimized airflow parameters, multi-channel control parameters are generated through piecewise linear mapping operations, and the multi-channel control parameters are channel-allocated through a channel optimization algorithm to generate fan control parameters; The triggering moment of the visual effect in the virtual environment is calculated through a state prediction algorithm, and a time delay compensation operation is performed on the fan control parameters based on the triggering moment of the visual effect to generate airflow synchronization control parameters; According to the airflow synchronization control parameters, the operating states of the fans in the distributed fan array are prioritized through a resource dynamic allocation algorithm, and the resource allocation weights of the fans are adjusted based on the user's gaze direction data to generate a fan operation sequence; The fan operation sequence is data-verified with the standard airflow sequence in the preset airflow parameter library through a cross-validation algorithm, and the fan operation sequence is dynamically calibrated through a deviation correction algorithm to generate an airflow feedback signal.

[0055] The following specifically describes the steps involved in the above embodiments: The generation process of the multi-channel control parameters adopts piecewise linear mapping and channel optimization techniques. The system first inputs the corrected airflow direction parameter (horizontal angle and vertical angle), adjusted airflow intensity parameter (wind force percentage value), and optimized airflow duration parameter (millisecond-level time value) in the optimized airflow parameters into the piecewise linear mapping operation unit. The piecewise linear mapping operation converts continuous airflow parameter values into discrete control signals. Specifically, the airflow direction parameter is converted into the activation state of each fan in the fan array (activation degree value between 0 and 1), the airflow intensity parameter is converted into the power level of the fan (usually 256-level PWM value), and the airflow duration parameter is converted into the start-stop timing and operation cycle of the fan. The mapping process adopts piecewise processing to adapt to the accuracy requirements of different parameter intervals. For example, a fine mapping with a 5-degree interval is used in the front area (±30 degrees) of the airflow direction, while a rough mapping with a 10-degree interval is used in the side area. The channel optimization algorithm then integrates and allocates the generated multi-channel control parameters to determine the specific control parameters of each fan (channel). The channel allocation considers the physical location of the fan, the maximum power output capacity, and the thermal management limitations, and appropriately adjusts and reallocates the control parameters. For example, when simulating strong wind in the direct front, the system will preferentially allocate front fan resources, activate the side fans appropriately to enhance the airflow feeling, but will limit the activation degree of the rear fans to save energy. The generated fan control parameters include the PWM output value (0 - 255), start time point, operation duration, and rotational speed change curve of each fan channel. This multi-channel allocation technology solves the problem of insufficient airflow expressiveness under limited fan resources, enabling the system to create rich airflow effects with limited hardware resources.

[0056] The synchronous processing of visual effects and airflow feedback adopts state prediction and time delay compensation techniques. The state prediction algorithm calculates the time delay from event trigger to actual display of visual effects by analyzing the rendering pipeline and event trigger mechanism of VR applications. The system monitors and analyzes the rendering frame rate (usually 90Hz or 120Hz), display delay (usually 5 milliseconds - 15 milliseconds), and event processing delay (usually 10 milliseconds - 30 milliseconds) of VR applications to establish an accurate timing model. Based on this model, the system precisely calculates the trigger moment of visual effects, that is, the predicted time interval from the current time point to when the user actually sees the visual effects. Subsequently, the time delay compensation operation adjusts the fan control parameters along the time axis, mainly considering three key delay factors: fan startup delay (the time from the control signal being sent to the fan reaching an effective rotational speed, usually 30 - 100 milliseconds), airflow propagation delay (the time from the fan generating airflow to being perceived by the user, usually 5 - 20 milliseconds), and perception processing delay (the time for the user's nervous system to process the airflow perception signal, usually 50 - 200 milliseconds). The system compensates in advance for the fan startup time according to these delay values to ensure that the user synchronously feels the matching airflow effect when the visual effect appears. For example, when an explosion effect will be displayed in a VR game after 500 milliseconds, the system will calculate and compensate for the total delay (about 150 milliseconds) in advance and start the fan at 350 milliseconds to ensure that the airflow sensation and the explosion visual effect reach the user synchronously. This precise time delay compensation mechanism solves the problem of airflow feedback lagging behind visual effects in traditional systems and significantly improves the coordination and immersion of the multi-sensory experience.

[0057] The generation of the fan operation sequence employs resource dynamic allocation and gaze direction optimization techniques. The resource dynamic allocation algorithm analyzes the importance and contribution degree of each fan based on the airflow synchronization control parameters, and assigns different priorities to them. The priority allocation considers three levels: the main contributing fan to the airflow direction (priority A, the highest), the auxiliary enhancement fan (priority B, medium), and the environmental effect fan (priority C, lower). The system monitors the user's gaze direction in real time, and obtains the spatial area that the user is currently focusing on through the eye tracker or head orientation sensor built into the headset. The gaze direction data (usually represented as horizontal and vertical angles) is then used to adjust the resource allocation weights of each fan. The system enhances the weights of the fans related to the user's gaze area (increasing by 15 - 30%), and reduces the weights of the fans in non-gaze areas (decreasing by 10 - 25%). For example, when the user gazes at the right area, the system enhances the output power of the right fan, while slightly reducing the power of the left and rear fans, ensuring that the limited power resources are preferentially used to enhance the airflow experience in the area where the user's current attention lies. This gaze-based resource optimization generates a highly optimized fan operation sequence by precisely controlling the operating states (on / off), power levels (PWM values), and timing parameters (start / stop time points) of each fan. This resource dynamic allocation and gaze optimization technology not only improves the energy efficiency of the system, but more importantly, enhances the quality of the airflow effect subjectively perceived by the user, solving the problem of insufficient airflow perception under limited fan resources.

[0058] The final generation of the airflow feedback signal uses cross-validation and dynamic calibration techniques. The cross-validation algorithm compares and verifies the generated fan operation sequence with the standard airflow sequence in the preset airflow parameter library. The preset airflow parameter library contains standard parameter configurations and expected effect descriptions for a variety of typical airflow scenarios (such as "breeze", "sudden airflow", "surrounding airflow", etc.). The system calculates the similarity between the currently generated operation sequence and each standard sequence in the library, identifies the closest standard mode, and analyzes the differences. The deviation correction algorithm then fine-tunes and optimizes the key parameters of the fan operation sequence to ensure that the final effect meets physical expectations and perception laws. The correction process focuses on three aspects: airflow consistency (avoiding uncoordinated effects between adjacent fans), physical rationality (ensuring that airflow changes comply with fluid physics laws), and perception balance (avoiding excessive or weak airflow in a certain area). Through these dynamic calibration adjustments, the system generates the final airflow feedback signal, which contains a complete set of fan control instructions, precise timing arrangements, and clear effect descriptions. For example, when simulating the "side wind" effect, the system may find that the initially generated fan operation sequence has problems in airflow consistency (the side fan start-up timing is not coordinated). Through the deviation correction algorithm, the fan start interval is adjusted to a fixed 30 milliseconds, and the transition interval of the power curve is optimized to ensure that the user perceives a smooth and coherent side wind effect. This cross-validation and dynamic calibration mechanism significantly improves the quality and reliability of airflow feedback and solves the problems of parameter incoordination and physical irrationality that may occur in the process of generating complex airflow effects.

[0059] The above describes the VR environment airflow feedback method based on user behavior in the embodiment of the present invention. The following describes the VR environment airflow feedback device based on user behavior in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, a VR environment airflow feedback device based on user behavior includes: The behavior perception module 101 is used to obtain the user behavior data and environmental event data of the user in the VR environment, perform event processing according to the priority attribute of the environmental event data, and obtain an event processing result; A scene prediction module 102, configured to analyze scene changes in the VR environment according to the event processing result and the user behavior data, generate scene prediction data, and determine initial airflow parameters according to the scene prediction data; The airflow control module 103 is used to control the distributed fan array according to the initial airflow parameters, adjust the airflow characteristics, and generate a basic airflow effect; A parameter optimization module 104, configured to generate a user airflow perception feature according to the user's perception feedback of the basic airflow effect, and optimize the initial airflow parameter based on the user's airflow perception feature to obtain an optimized airflow parameter; A feedback generation module 105 is configured to control the distributed fan array according to the optimized airflow parameters to generate an airflow feedback signal.

[0060] above Figure 2 From the perspective of modular functional entities, the VR environment airflow feedback device based on user behavior in the embodiments of the present invention is described in detail. Next, from the perspective of hardware processing, the VR environment airflow feedback device based on user behavior in the embodiments of the present invention is described in detail.

[0061] Figure 3 FIG. 9 is a schematic structural diagram of a VR environment airflow feedback device based on user behavior provided by an embodiment of the present invention. The VR environment airflow feedback device 200 based on user behavior may vary greatly due to configuration or performance differences, and may include one or more processors 210 (for example, one or more processors) and a memory 220, and one or more storage media 230 for storing application programs 233 or data 232 (for example, one or more mass storage device terminals). Among them, the memory 220 and the storage media 230 may be transient storage or persistent storage. The program stored in the storage media 230 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the VR environment airflow feedback device 200 based on user behavior. Further, the processor 210 may be configured to communicate with the storage media 230 and execute a series of instruction operations in the storage media 230 on the VR environment airflow feedback device 200 to implement the steps of the above-mentioned VR environment airflow feedback method based on user behavior.

[0062] The VR environment airflow feedback device 200 based on user behavior may further include one or more power supplies 240, one or more wired or wireless network interfaces 250, one or more input / output interfaces 260, and / or one or more operating systems 231, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, and so on. Those skilled in the art can understand that Figure 3 The shown structural diagram of the VR environment airflow feedback device based on user behavior does not limit the VR environment airflow feedback device provided by the present invention, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0063] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions run on a computer, the computer is caused to execute the steps of the VR environment airflow feedback method based on user behavior.

[0064] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, device, or unit can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0065] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions to cause a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0066] The above are only the preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural transformation made by using the content of the specification and drawings of the present invention under the inventive concept of the present invention, or any direct / indirect application in other related technical fields is included in the patent protection scope of the present invention.

Claims

1. A method for airflow feedback in a VR environment based on user behavior, characterized in that, Including: Obtain user behavior data and environmental event data of a user in a VR environment, perform event processing according to the priority attributes of the environmental event data to obtain an event processing result; Analyze the scene changes in the VR environment according to the event processing result and the user behavior data, generate scene prediction data, and determine initial airflow parameters according to the scene prediction data; Control a distributed fan array according to the initial airflow parameters, adjust the airflow characteristics, and generate a basic airflow effect; Generate user airflow perception characteristics according to the user's perception feedback on the basic airflow effect, optimize the initial airflow parameters based on the user airflow perception characteristics to obtain optimized airflow parameters; Control the distributed fan array according to the optimized airflow parameters to generate an airflow feedback signal.

2. The method for VR environment airflow feedback based on user behavior according to claim 1, wherein The obtaining user behavior data and environmental event data of a user in a VR environment, performing event processing according to the priority attributes of the environmental event data to obtain an event processing result includes: Obtain user head movement parameters, hand movement data, and historical behavior records, and combine the head movement parameters, hand movement data, and historical behavior records through time series correlation operation to form user behavior data; Obtain natural environment event data, interaction event data, and game logic event data in the virtual environment, combine the natural environment event data, interaction event data, and game logic event data through data merging operation to form environmental event data, and perform spatio-temporal attribute extraction on the environmental event data through spatial coordinate extraction and timestamp extraction to generate event spatio-temporal data; Perform quantification calculations on the importance and urgency of each event in the environmental event data, and generate event priority attributes according to the results of the quantification calculations of the importance and urgency; Perform a priority sorting operation on the environmental event data according to the quantification calculation results in the event priority attributes to form an event priority sequence; Perform priority weighting processing and spatio-temporal information fusion processing on the environmental event data based on the event priority sequence and the event spatio-temporal data to generate an event processing result including event priority information and event spatio-temporal information.

3. The method for VR environment airflow feedback based on user behavior according to claim 1, wherein The analyzing the scene changes in the VR environment according to the event processing result and the user behavior data, generating scene prediction data, and determining initial airflow parameters according to the scene prediction data includes: According to the event priority information and event spatio-temporal information in the event processing result, extract physical event parameters, interaction event parameters, and game event parameters through spatial situation analysis and time series correlation analysis, and combine the physical event parameters, interaction event parameters, and game event parameters through parameter vectorization operation to generate scene prediction source data; Perform time series pattern recognition operation and behavior rule extraction operation on the historical behavior records in the user behavior data to generate a scene reaction pattern and behavior change rules, and perform data reconstruction on the scene prediction source data through pattern matching operation and rule mapping operation to generate user scene prediction data; Based on the predicted data of the user scenario, analyze the timing sequence of scenario changes through time-series probability distribution calculation and state prediction operation to generate scenario prediction data, where the scenario prediction data includes the event prediction time, the event occurrence location, and the event influence range; Perform space-time parameter mapping matrix operation on the event prediction time, the event occurrence location, and the event influence range in the scenario prediction data to generate initial airflow parameters including airflow direction parameters, airflow intensity parameters, and airflow duration parameters.

4. The method for VR environment airflow feedback based on user behavior according to claim 3, wherein The analysis of the timing sequence of scenario changes through time-series probability distribution calculation and state prediction operation based on the predicted data of the user scenario to generate scenario prediction data includes: Perform in-depth probability analysis on the scenario response mode in the predicted data of the user scenario, and generate a scenario type prediction probability matrix according to the response probability distribution of the user in different VR scenario types; According to the scenario type prediction probability matrix, perform type recognition operation on the behavior characteristics of the user in high-altitude scenarios, underwater scenarios, and land scenarios to generate scenario transition prediction data; Based on the scenario transition prediction data, calculate the scenario transition time and transition area through the scenario boundary recognition algorithm to generate scenario transition prediction parameters; According to the scenario transition prediction parameters, perform fitting operation on the behavior change trend of the user before and after scenario transition to generate a user behavior prediction curve, and perform space-time mapping operation on the user behavior prediction curve and the scenario transition prediction parameters to generate scenario prediction data.

5. The method for VR environment airflow feedback based on user behavior according to claim 1, wherein The control of the distributed fan array according to the initial airflow parameters to adjust the airflow characteristics and generate a basic airflow effect includes: According to the airflow direction parameters and airflow intensity parameters in the initial airflow parameters, calculate the power distribution coefficients of each fan in the distributed fan array through the power distribution algorithm, and perform dynamic distribution of the rotation speeds of each fan through the rotation speed dynamic optimization algorithm to generate a fan control sequence; According to the airflow duration parameters in the initial airflow parameters and the fan control sequence, generate the fan start-stop timing difference and power change curve through the timing optimization algorithm, and perform dynamic adjustment on the start timing and power change of each fan through the power curve fitting operation to generate a pulsed airflow control signal; According to the pulsed airflow control signal, perform parameter adjustment on the vortex intensity, diffusion angle, and pulse frequency of the airflow through vortex intensity calculation, diffusion angle optimization, and pulse frequency modulation operation to generate airflow adjustment parameters, and perform data matching on the airflow adjustment parameters and the real-world airflow parameter model in the preset airflow characteristic library through the feature matching algorithm to generate airflow characteristic matching data; According to the airflow characteristic matching data, perform spatial grouping on multiple fans of the distributed fan array through the annular distribution optimization algorithm, and perform linkage control on the power and timing of each group of fans through the group collaborative control algorithm to generate a basic airflow effect including airflow direction value, airflow intensity value, and airflow duration value.

6. The method for VR environment airflow feedback based on user behavior according to claim 1, wherein The generation of user airflow perception characteristics based on the user's perception feedback of the basic airflow effect, and the optimization of the initial airflow parameters based on the user airflow perception characteristics to obtain optimized airflow parameters includes: Obtain the wind direction recognition data and wind force perception data of the user for the basic air flow effect, calculate the deviation value between the wind direction recognition data and the standard wind direction and the ratio between the wind force perception data and the standard wind force, and generate a wind direction perception compensation coefficient and a wind force scaling factor; Obtain the user's head turning reaction data and action adjustment data, calculate the time series change value of the user's head turning reaction data and the amplitude change value of the action adjustment data, and generate a user air flow adaptation curve; Perform a weighted fusion operation on the wind direction perception compensation coefficient, the wind force scaling factor, and the user air flow adaptation curve to generate a user air flow perception feature; Perform a vector operation on the wind direction perception compensation coefficient and the air flow direction parameter in the initial air flow parameters, and perform a multiplication operation on the wind force scaling factor and the air flow intensity parameter in the initial air flow parameters to generate air flow reference parameters; Perform a parameter mapping operation on the time series change value and the amplitude change value in the user air flow adaptation curve with the air flow reference parameters respectively to generate optimized air flow parameters including a corrected air flow direction parameter, an adjusted air flow intensity parameter, and an optimized air flow duration parameter.

7. The method for airflow feedback in a VR environment based on user behavior according to claim 1, wherein Controlling the distributed fan array according to the optimized air flow parameters to generate an air flow feedback signal includes: According to the corrected air flow direction parameter, the adjusted air flow intensity parameter, and the optimized air flow duration parameter in the optimized air flow parameters, generate multi-channel control parameters through a piecewise linear mapping operation, and perform channel allocation on the multi-channel control parameters through a channel optimization algorithm to generate fan control parameters; Calculate the visual effect trigger moment in the virtual environment through a state prediction algorithm, and perform a time delay compensation operation on the fan control parameters based on the visual effect trigger moment to generate air flow synchronization control parameters; According to the air flow synchronization control parameters, perform a priority allocation on the operating states of the fans in the distributed fan array through a resource dynamic allocation algorithm, and adjust the resource allocation weights of the fans based on the user's gaze direction data to generate a fan operation sequence; Verify the data of the fan operation sequence with the standard air flow sequence in the preset air flow parameter library through a cross-validation algorithm, and perform dynamic calibration on the fan operation sequence through a deviation correction algorithm to generate an air flow feedback signal.

8. An airflow feedback device for a VR environment based on user behavior, characterized in that, The VR environment air flow feedback device based on user behavior adopts the VR environment air flow feedback method based on user behavior as described in any one of claims 1 to 7. The VR environment air flow feedback device based on user behavior includes: A behavior perception module for obtaining user behavior data and environmental event data of the user in the VR environment, and performing event processing according to the priority attribute of the environmental event data to obtain an event processing result; A scene prediction module for analyzing the scene change in the VR environment according to the event processing result and the user behavior data, generating scene prediction data, and determining initial air flow parameters according to the scene prediction data; An air flow control module for controlling a distributed fan array according to the initial air flow parameters, adjusting the air flow characteristics, and generating a basic air flow effect; A parameter optimization module, configured to generate user airflow perception features according to the user's perception feedback on the basic airflow effect, and optimize the initial airflow parameters based on the user airflow perception features to obtain optimized airflow parameters; A feedback generation module, configured to control the distributed fan array according to the optimized airflow parameters to generate an airflow feedback signal.

9. A VR environment airflow feedback device based on user behavior, characterized in that, The VR environment airflow feedback device based on user behavior includes: a memory and at least one processor, and instructions are stored in the memory; The at least one processor invokes the instructions in the memory so that the VR environment airflow feedback device based on user behavior executes the steps of the VR environment airflow feedback method based on user behavior as described in any one of claims 1-7.

10. A computer-readable storage medium, on which instructions are stored, characterized in that, When the instructions are executed by the processor, the steps of the VR environment airflow feedback method based on user behavior as described in any one of claims 1-7 are implemented.

Images