Dynamic air supply adjusting method, device and equipment based on human body induction and medium

Abstract

The invention relates to a dynamic air supply adjusting method, device and equipment based on human body induction and a medium. The method comprises the following steps: acquiring multi-source sensing data of a target human body; according to the multi-source sensing data, performing state classification on the target human body to obtain a state classification result of the target human body; the state classification result comprises static and moving; according to the three-dimensional position and the state classification result of the target human body, predicting a future motion track of the target human body to obtain a predicted track of the target human body; according to the prediction track, the optimal air supply angle and air speed are calculated, and air supply parameters are obtained; the air supply parameters comprise air supply angles and air speeds; according to the air supply parameters, an air supply control instruction is generated and sent to an air supply device; the air supply control instruction is used for indicating adjustment of air supply parameters of the air supply device and executing dynamic tracking air supply. By means of the method, the human body comfort degree of air supply can be improved, and real-time accurate dynamic tracking air supply is achieved.

Description

Technical Field

[0001] The present invention belongs to the field of air conditioning, and in particular relates to a dynamic air supply adjustment method, device, equipment and medium based on human body induction. Background Art

[0002] With the development of air supply technology, air supply technology uses infrared, visual, or other sensing methods to perceive the position and activity of the human body in real time. The goal is to accurately and proactively guide airflow to the areas where people are active, improving local thermal comfort and optimizing energy efficiency. Currently, dynamic air supply has become an important way to achieve personalized thermal environment regulation.

[0003] Traditional dynamic air supply control methods primarily employ a strategy combining fixed zone division with simple feedback control. The system typically divides the space into several pre-set zones. When a person is detected in a zone, the corresponding air supply device (such as adjustable guide vanes or zone air vents) is activated to deliver air to the static zone. Alternatively, based on the human position reported by sensors, the air supply terminal (such as a swinging air vent) is controlled to steer toward that point for directional air supply. These two approaches form the basis of current mainstream implementations.

[0004] However, existing methods have significant defects when dealing with moving targets. The human body is not static in space, and its position changes continuously. When a person moves from one area to another, or from one point to another, there is an inevitable lag in the system's response. When people move quickly or frequently, it is difficult for the air supply to track the target in real time and accurately, resulting in the airflow acting on unmanned areas for a long time or lagging behind in covering new demand points. The target person feels that the air supply is interrupted or delayed, and the thermal comfort (such as the wind sensation index or PMV) decreases. At the same time, the ineffective air supply area increases, and the air supply lag is disconnected from the moving target. Summary of the Invention

[0005] Based on this, it is necessary to provide a dynamic air supply adjustment method, device, equipment and medium based on human body sensing that can achieve dynamic chasing blowing in response to the above technical problems.

[0006] In a first aspect, the present application provides a dynamic air supply adjustment method based on human body sensing, comprising:

[0007] Acquire multi-source perception data of the target human body;

[0008] Based on the multi-source perception data, the state of the target human body is classified to obtain the state classification result of the target human body; the state classification result includes stillness and movement;

[0009] According to the classification results of the three-dimensional position and state of the target body, the future motion trajectory of the target body is predicted to obtain the predicted trajectory of the target body;

[0010] According to the predicted trajectory, the optimal air supply angle and wind speed are calculated to obtain the air supply parameters; the air supply parameters include the air supply angle and wind speed;

[0011] Based on the air supply parameters, an air supply control instruction is generated and sent to the air supply device; the air supply control instruction is used to instruct the adjustment of the air supply parameters of the air supply device and perform dynamic tracking of air supply.

[0012] In one embodiment, based on multi-source perception data, the target human body is classified into static and motion states, and the state classification result of the target human body is obtained, including:

[0013] Performing fusion processing on the multi-source perception data to obtain fused multi-source perception data;

[0014] Extract the visual and inertial features of the target human body from the fused multi-source perception data. The visual features include the human body outline and motion trajectory; the inertial features include acceleration and angular velocity.

[0015] Perform bimodal fusion of visual features and inertial features to obtain fused behavioral features;

[0016] The fused behavior features are classified and processed to obtain the state classification results.

[0017] In one embodiment, predicting the future motion trajectory of the target body based on the three-dimensional position and state classification results of the target body to obtain the predicted trajectory of the target body includes:

[0018] Calculate the three-dimensional position of the target body based on the coordinate system of the target body in space;

[0019] Use Kalman filtering to perform short-term trajectory prediction on the three-dimensional position of the target body to obtain the basic predicted trajectory;

[0020] Real-time monitoring of the state classification results and the three-dimensional position of the target human body to obtain monitoring results;

[0021] The basic predicted trajectory is adjusted according to the monitoring results, the deviation of the basic predicted trajectory is corrected, and the predicted trajectory of the target human body is obtained.

[0022] In one embodiment, the optimal air supply angle and wind speed are calculated based on the predicted trajectory of the target human body to obtain air supply parameters, including:

[0023] Based on the predicted trajectory, calculate the basic air supply angle; the basic air supply angle is calculated using the following formula:

[0024]

[0025] Among them, θ t is the air supply angle, xfan 、y fan are the horizontal and vertical coordinates of the air supply device, is the horizontal and vertical coordinates of the predicted position on the predicted trajectory;

[0026] According to the preset influence function of air supply angle and wind speed on comfort, the preset upper limit of wind speed and the constraint conditions of air supply angle, the basic air supply angle and wind speed are adjusted to obtain the air supply parameters.

[0027] In one embodiment, the method further comprises:

[0028] Monitor the air supply status of the air supply device in real time to obtain the real-time air supply angle and real-time wind speed;

[0029] The real-time air supply angle and real-time wind speed are calculated and deviated from the air supply angle and wind speed in the air supply parameters to obtain the deviation result. The deviation result is obtained using the following formula:

[0030] δ result =w1·δ θ +w2·δ v

[0031] Among them, δ result is the deviation result, w1 and w2 are the weight values ​​of air supply angle deviation and wind speed deviation respectively, δ θ is the air supply angle deviation, δ v is the wind speed deviation;

[0032] The air supply angle deviation is obtained using the following formula:

[0033] δ θ =|θ real -θ set |·cos(θ real -θ set )

[0034] Among them, δ θ is the air supply angle deviation, θ real is the real-time air supply angle, θ set is the air supply angle in the air supply parameters;

[0035] The wind speed deviation is obtained using the following formula:

[0036]

[0037] Among them, δ v is the wind speed deviation, v real is the real-time wind speed, v set is the wind speed in the air supply parameters, v min is the preset minimum wind speed, max(·) is the maximum value function, and α is the attenuation coefficient;

[0038] Adjust the air supply parameters of the air supply device according to the deviation results.

[0039] In one embodiment, the method further comprises:

[0040] When multiple target bodies are detected, trajectory prediction is performed for each target body to obtain the corresponding predicted trajectory position of each target body;

[0041] Calculate the optimal air supply angle and wind speed based on the predicted trajectory position corresponding to each target human body;

[0042] Adjust the air supply parameters of the air supply device according to the optimal air supply angle and wind speed.

[0043] In one embodiment, the air supply device includes at least two air outlets.

[0044] In a second aspect, the present application further provides a dynamic air supply adjustment device based on human body sensing, comprising:

[0045] A data acquisition module is used to obtain multi-source perception data of the target human body;

[0046] The state classification module is used to classify the state of the target human body based on multi-source perception data and obtain the state classification results of the target human body; the state classification results include stillness and movement;

[0047] The trajectory prediction module is used to predict the future motion trajectory of the target body based on the three-dimensional position and state classification results of the target body, and obtain the predicted trajectory of the target body;

[0048] The air supply parameter module is used to calculate the optimal air supply angle and wind speed based on the predicted trajectory to obtain the air supply parameters; the air supply parameters include air supply angle and wind speed;

[0049] The air supply control instruction generation module is used to generate air supply control instructions based on air supply parameters and send air supply control instructions to the air supply device; the air supply control instructions are used to instruct the adjustment of the air supply parameters of the air supply device and perform dynamic tracking of air supply.

[0050] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in the first aspect when executing the computer program.

[0051] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method described in the first aspect when the computer program is executed by a processor.

[0052] The above-mentioned dynamic air supply adjustment method, device, computer equipment and storage medium based on human body sensing can comprehensively and accurately capture human body-related information by obtaining multi-source perception data of the target human body; classify the target human body according to the multi-source perception data, and clearly distinguish between static and moving states, which helps to improve the recognition of the target human body's behavior pattern; combine the three-dimensional position and state classification results of the target human body to predict the future motion trajectory, and grasp the dynamic direction of the human body in advance; calculate the optimal air supply angle and wind speed based on the predicted trajectory to ensure that the air supply parameters meet the actual needs of the human body; finally generate and send air supply control instructions to enable the air supply device to perform dynamic tracking of air supply, and realize intelligent and personalized adjustment of air supply; adjust the air supply parameters in real time and accurately according to the different states and position changes of the human body, improve human comfort, reduce energy consumption, and realize dynamic tracking of air supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 1 is a flow chart of a dynamic air supply adjustment method based on human body sensing in one embodiment;

[0055] Figure 2 Schematic diagram of the structure of a dynamic air supply adjustment device based on human body sensing in one embodiment;

[0056] Figure 3 FIG. 1 is a schematic diagram of a computer device in one embodiment. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0058] In one embodiment, Figure 1 As shown, a dynamic air supply adjustment method based on human body sensing is provided. This embodiment uses the method applied to a human body sensing air supply terminal as an example. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0059] S101, obtaining multi-source perception data of a target human body.

[0060] Among them, the multi-source perception data of the target human body is obtained through various sensor devices such as pyroelectric infrared sensors, millimeter-wave radars or depth cameras deployed at the human body sensing air supply terminal; for example, the pyroelectric infrared sensor can sense the presence of the human body by capturing the infrared radiation emitted by the human body; the millimeter-wave radar can use electromagnetic waves to detect the human body and obtain the three-dimensional position and position information of the human body; the depth camera collects the outline and posture of the human body; the temperature and humidity sensor collects environmental data.

[0061] S102: Classify the state of the target human body according to the multi-source perception data to obtain a state classification result of the target human body.

[0062] Among them, the state classification results include stillness and movement.

[0063] For example, the human body sensing air supply terminal analyzes and processes motion-related data from multi-source sensing data to categorize the human body as either stationary or in motion. Specifically, the human body's motion is determined by calculating the distance the human body moves in three dimensions over a certain period of time and the frequency change of the millimeter-wave radar echo. If the human body's movement is small and inconspicuous, and the frequency change of the millimeter-wave radar echo is small, the state is determined to be stationary. If the human body moves significantly and the frequency change of the millimeter-wave radar echo is large, the state is determined to be in motion, thus obtaining the state classification result.

[0064] S103 , predicting the future motion trajectory of the target body according to the three-dimensional position and state classification results, and obtaining the predicted trajectory of the target body.

[0065] For example, the human body sensing air supply terminal inputs the three-dimensional position and state classification results of the target person into a pre-trained trajectory prediction model to obtain the target person's predicted motion trajectory. Specifically, for a target person in a stationary state, the trajectory prediction model predicts that the target person's position will remain unchanged for a short period of time. For a target person in motion, the trajectory prediction model comprehensively analyzes the target person's historical movement direction, speed, and other data, and uses a Kalman filter algorithm or an LSTM time series prediction algorithm to infer the target person's position changes over a period of time, thereby forming a predicted trajectory of the target person's movement.

[0066] S104: Calculate the optimal air supply angle and wind speed based on the predicted trajectory to obtain air supply parameters.

[0067] Among them, the air supply parameters include air supply angle and wind speed.

[0068] Specifically, the human body sensing air supply terminal determines the key areas of the target human body as the target areas for air supply based on the motion trajectory; through geometric calculations, it confirms the air supply angle that allows the air to accurately reach the target area; at the same time, it monitors the temperature data of the target human body and adjusts the wind speed to meet the comfort level of the target human body and avoid discomfort, and finally obtains the air supply parameters.

[0069] S105 , generating an air supply control instruction based on the air supply parameters and sending the air supply control instruction to the air supply device; the air supply control instruction is used to instruct the adjustment of the air supply parameters of the air supply device to perform dynamic tracking of air supply.

[0070] For example, the human body sensing air supply terminal converts the obtained air supply parameters into air supply control instructions that can be executed by the air supply device and sends them to the air supply device. The air supply device adjusts the air supply according to the air supply control instructions, achieving dynamic tracking of air supply to the target person. The air supply device includes at least two air outlets, and the multiple air outlets are used to meet the air supply needs of the target person.

[0071] In the above-mentioned dynamic air supply adjustment method based on human body sensing, multi-source sensors are used to collect data of the target human body, and the data is analyzed and processed to realize the state judgment of the target human body, the prediction of the motion trajectory and the optimization of the air supply parameters. By sending control instructions in a timely manner, the air supply device is driven to adjust in real time, which can make the air supply accurately fit the real-time state changes and real-time position changes of the human body, improve the comfort of the human body when receiving air supply, and realize dynamic tracking blowing.

[0072] Optionally, classifying the target human body into static and moving states based on the multi-source perception data to obtain a state classification result of the target human body includes the following steps:

[0073] S1001: Perform fusion processing on each multi-source perception data to obtain fused multi-source perception data.

[0074] For example, the human body sensing air supply terminal performs spatiotemporal alignment and fusion processing on the multi-source sensor data it collects. Timestamp synchronization technology ensures temporal consistency of different sensor data. A coordinate system conversion algorithm is then used to unify the time-synchronized sensor data into the human body sensing air supply terminal's global coordinate system. An adaptive weighted fusion method is used to combine the dominant information from each sensor data source to ultimately generate fused multi-source sensor data containing multi-dimensional information such as the target person's position, posture, and motion speed. This dominant information includes the camera's visual outline, the radar's three-dimensional position, and the inertial sensor's motion acceleration.

[0075] S1002, extracting visual features and inertial features of the target human body from the fused multi-source perception data.

[0076] Among them, visual features include human body contours and motion trajectories, which are used to characterize the spatial position changes and overall movement trends of the human body; inertial features include acceleration and angular velocity, which are used to characterize the dynamic behavior details of the human body.

[0077] The human-sensing ventilation terminal uses computer vision algorithms to segment the human body from the camera video stream, capturing real-time contour changes and position offset trajectories between consecutive frames, forming human contour features and motion trajectory features, i.e., visual features. The terminal uses zero-bias correction to de-drift the raw acceleration and angular velocity data collected by the inertial sensor. Using a sliding window, it calculates statistics such as the mean, variance, and peak value over the time series to extract acceleration and angular velocity features, i.e., inertial features. Computer vision algorithms include YOLO (You Only Look Once) target detection and DeepSORT (Simple Online and Realtime Tracking with a DeepAssociation Metric, deep learning for multi-target tracking) trajectory tracking.

[0078] S1003: Perform bimodal fusion of visual features and inertial features to obtain fused behavioral features.

[0079] Exemplarily, the human body sensing air supply terminal deeply fuses visual features and inertial features, dimensionally splices the visual features and inertial features to form a multi-dimensional fusion behavior feature vector that includes spatial position, motion trajectory, and dynamic acceleration, and extracts nonlinear fusion features from the multi-dimensional fusion behavior features, so that the fusion behavior features can not only retain the spatial positioning of vision and data, but also reflect the high-frequency dynamic details of inertial data, and more effectively characterize the differentiated behavior of the human body in a static or moving state.

[0080] S1004: Classify the fused behavior features to obtain a state classification result.

[0081] Among them, the state classification results include stillness and movement.

[0082] For example, the human body sensing air supply terminal uses a trained bimodal classification model to classify the fused behavioral features; the bimodal classification model inputs the fused behavioral features and outputs a binary classification result of "static" or "moving". In the bimodal training stage, by labeling a large amount of sample data containing static (such as standing or sitting still) and moving (such as walking or waving) scenes, the cross-entropy loss function is used to optimize the parameters of the bimodal classification model, so that the bimodal classification model can learn the feature differences in different states (such as the trajectory offset approaches zero and the acceleration amplitude is low in the static state; the trajectory changes significantly and the acceleration / acceleration fluctuations are large in the moving state); during state inference, the bimodal classification model processes the input features in real time, calculates the state probability through the Softmax function, and finally outputs the state classification result with the highest confidence. Among them, the bimodal classification model includes random forest, LSTM (Long Short-Term Memory) neural network and 3D convolutional network.

[0083] Optionally, predicting the future motion trajectory of the target body according to the three-dimensional position and state classification results to obtain the predicted trajectory of the target body includes the following steps:

[0084] S2001, calculating the three-dimensional position of the target body based on the coordinate system of the target body in space.

[0085] For example, the human body sensing air supply terminal establishes a global coordinate system with itself as the origin and determines the three-dimensional position of the target person through multi-sensor data such as millimeter-wave radar and camera. Specifically, the millimeter-wave radar obtains the distance, azimuth, and elevation of the target person through electromagnetic wave reflection. By combining the conversion relationship between the radar coordinate system and the global coordinate system of the human body sensing air supply terminal, the coordinates of the person in the global coordinate system are calculated. The camera uses binocular vision ranging technology and calibrated internal and external parameters to map the two-dimensional image coordinates to three-dimensional space. The human body sensing air supply terminal takes a weighted average of the data obtained by the millimeter-wave radar and camera, and then fuses them to finally obtain the three-dimensional position of the target person in space.

[0086] S2002: Use Kalman filtering to perform short-term trajectory prediction on the three-dimensional position of the target human body to obtain a basic predicted trajectory.

[0087] For example, the human body sensing air supply terminal uses the Kalman filter algorithm to construct a state-space model based on the obtained three-dimensional position to model the motion state of the target human body. The state-space model includes a state vector, a state transition equation, and an observation equation. The state vector includes position and velocity. The state transition equation assumes that the human body's motion conforms to a uniform speed or uniform acceleration model. The observation equation corresponds to the three-dimensional position measured by the sensor. The current state is predicted based on the state at the previous moment and the predicted value is corrected using the current measurement value. The Kalman filter can effectively filter out sensor noise, smooth position data, and extrapolate future short-term position coordinates to form a continuous basic prediction trajectory.

[0088] S2003, monitoring the state classification result and the three-dimensional position of the target human body in real time to obtain a monitoring result.

[0089] Exemplarily, the human body sensing air supply terminal will collect the status classification results and the three-dimensional position of the target human body in real time to form a monitoring result. When the status classification result is "stationary", the monitored three-dimensional position fluctuates little in a short time; when the status classification result is "moving", the position change amplitude is large and directional. The monitoring process uses a timer to update the status classification result and the three-dimensional position information of the target human body every 50-100 milliseconds.

[0090] S2004: Adjust the basic predicted trajectory according to the monitoring results, correct the deviation of the basic predicted trajectory, and obtain the predicted trajectory of the target human body.

[0091] For example, when the state classification result is "stationary", it is determined whether the future position in the basic predicted trajectory exceeds the reasonable fluctuation range in the stationary state. If it exceeds, the predicted position is corrected to the current latest position to suppress the offset caused by inertial prediction; when the state classification result is "moving", the difference between the latest three-dimensional position and the historical position is extracted, the real-time movement speed and direction are calculated, the state vector of the Kalman filter is updated, and the acceleration parameters are adjusted; through the deviation correction in the stationary state and the moving state, the target human body predicted trajectory that fits the current state of the human body is obtained, ensuring that the trajectory is stable when stationary and the trajectory dynamically follows when moving.

[0092] Optionally, according to the predicted trajectory of the target human body, the optimal air supply angle and wind speed are calculated to obtain air supply parameters, including the following steps:

[0093] S3001: Calculate the basic air supply angle based on the predicted trajectory. The basic air supply angle is calculated using the following formula:

[0094]

[0095] Among them, θ t is the air supply angle, x fan 、yfan are the horizontal and vertical coordinates of the air supply device, are the horizontal and vertical coordinates of the predicted position on the predicted trajectory.

[0096] For example, the human body sensing air supply terminal establishes a local coordinate system with the air supply device as the origin, and transforms the target position on the predicted trajectory into the local coordinate system. Assuming that the horizontal and vertical coordinates of the air supply device are (x fan ,y fan ), the basic air supply angle uses the coordinate difference between the two points to calculate the horizontal angle that the air supply transposition needs to deflect by the inverse tangent function. Similarly, according to the three-dimensional position of the target human body in the predicted trajectory and the vertical coordinate of the air supply transposition, the difference in the vertical direction is calculated. Combined with the distance in the horizontal direction, the vertical air supply angle is obtained by the inverse tangent function; according to the horizontal air supply angle and the vertical air supply angle, the air supply transposition is adjusted to the basic air supply direction required in the predicted trajectory.

[0097] S3002: According to the preset influence function of the air supply angle and wind speed on the comfort, the preset upper limit of the wind speed and the constraint condition of the air supply angle, the basic air supply angle and wind speed are adjusted to obtain the air supply parameters.

[0098] Exemplarily, the human body sensing air supply terminal optimizes the basic air supply angle and wind speed according to the preset comfort influence function, the upper limit of the wind speed (i.e., the maximum wind speed that the air supply device can adjust) and the angle constraint condition (i.e., the maximum and minimum values ​​of the air supply angle that the air supply device can achieve). Specifically, when the basic air supply angle exceeds the limit range of the air supply device, the basic air supply angle is adjusted to the maximum value of the air supply angle of the air supply device; the wind speed is set according to the target human body state classification result. When in a stationary state, the wind speed is relatively low to achieve gentle air supply. When in motion, the wind speed is increased, and the comfort function is used for adjustment to find the optimal wind speed, and the wind speed must be less than the maximum wind speed that the air supply device can adjust; if the predicted trajectory shows that the target human body is moving away from the air supply device, the wind speed is gradually increased (not exceeding the maximum wind speed of the air supply device) when the angle reaches the limit to ensure that the air supply can cover; if the human body is stationary, the wind speed can be appropriately reduced to keep the angle aligned with the target human body to ensure comfortable air supply; the air supply parameters are obtained based on the adjusted air supply angle and wind speed.

[0099] Optionally, the method further comprises the following steps:

[0100] S4001, monitor the air supply status of the air supply device in real time to obtain the real-time air supply angle and real-time wind speed.

[0101] Exemplarily, the human body sensing air supply terminal obtains the real-time air supply angle and real-time wind speed of the air supply device in real time through the built-in angle sensor and wind speed sensor.

[0102] S4002: Calculate the deviation between the real-time air supply angle and the real-time wind speed and the air supply angle and wind speed in the air supply parameters to obtain a deviation result. The deviation result is obtained using the following formula:

[0103] δ result =w1·δ θ +w2·δ v

[0104] Among them, δ result为 Deviation results, w1, w2 are the weight values ​​of air supply angle deviation and wind speed deviation respectively, δ θ is the air supply angle deviation, δ v is the wind speed deviation.

[0105] In the calculation formula of the above deviation result, the overall deviation degree between the current air supply state and the target is quantified by weighted summation of the air supply angle deviation and the wind speed deviation.

[0106] S40021, where the air supply angle deviation is obtained using the following formula:

[0107] δ θ =|θ real -θ set |·cos(θ real -θ set )

[0108] Among them, δ θ is the air supply angle deviation, θ real is the real-time air supply angle, θ set It is the air supply angle in the air supply parameters.

[0109] In the above-mentioned calculation formula for the air supply angle deviation, the first term is the absolute deviation between the real-time air supply angle and the air supply angle in the air supply parameters, which is used to reflect the degree of deviation of the air supply direction; the second term is the calculation formula for the directional consistency coefficient, which is used to reflect the consistency of the air supply direction; when the air supply angle deviation is greater than the preset air supply angle deviation threshold, the angle adjustment is triggered, and the human body sensing air supply terminal adjusts the air supply angle in the air supply parameters.

[0110] S40022, where the wind speed deviation is obtained using the following formula:

[0111]

[0112] Among them, δ v is the wind speed deviation, v real is the real-time wind speed, v set is the wind speed in the air supply parameters, v min is the preset minimum wind speed, max(·) is the maximum value function, and α is the attenuation coefficient.

[0113] In the above calculation formula for wind speed deviation, the first term is the normalized wind speed deviation, and the second term is a nonlinear correction term, which is used to enhance the sensitivity to small deviations to ensure that the wind speed deviation is suppressed when it is lower than the threshold, and only responds to obvious deviations with wind speed adjustments; when the wind speed deviation is greater than the preset wind speed deviation threshold, the human body sensing air supply terminal adjusts the wind speed in the air supply parameters.

[0114] S4003: Adjust the air supply parameters of the air supply device according to the deviation result.

[0115] For example, when the deviation result is not within the allowable range, adjustments are made according to the direction of the deviation. Specifically, when θ real <θ set , the human body sensing air supply terminal sends a clockwise rotation control instruction to the air supply device, and vice versa, it sends a counterclockwise rotation control instruction; when v real <v set , the human body sensing air supply terminal sends a control instruction to the air supply device to increase the wind speed, otherwise it sends a control instruction to reduce the wind speed.

[0116] Optionally, the method further comprises the following steps:

[0117] S5001: When multiple target human bodies are detected, trajectory prediction is performed on each target human body to obtain the predicted trajectory position corresponding to each target human body.

[0118] Exemplarily, the human body sensing air supply terminal identifies multiple target human bodies in the scene, assigns a unique ID to each target and establishes a trajectory prediction model; for each target, the human body sensing air supply terminal uses the fusion data of millimeter wave radar and camera based on the single-target trajectory prediction process to update the three-dimensional position in real time, and then uses Kalman filtering to predict the motion trajectory in the next 1-2 seconds to obtain the predicted trajectory position of each target human body.

[0119] S5002: Calculate the optimal air supply angle and wind speed based on the predicted trajectory position corresponding to each target human body.

[0120] Exemplarily, the human body sensing air supply terminal spatially clusters the predicted trajectory positions of multiple target human bodies according to the predicted trajectory position corresponding to each target human body, determines the main activity area, calculates the center of mass of the area as the core target point of air supply, and calculates the basic air supply angle of the center of mass coordinate; the human body sensing air supply terminal sets a priority weight for each target human body, and calculates the wind speed by weighted average, while ensuring that the total wind speed does not exceed the upper limit of the device; if the distribution of multiple target human bodies exceeds the air supply range of the air supply device, the air supply angle is dynamically switched according to the priority order (such as 3 seconds of air supply for high-weight targets each time, and 1 second of air supply for low-weight targets each time), to ensure that each target can obtain effective air supply, thereby obtaining the best air supply angle and wind speed.

[0121] S5003: Adjust the air supply parameters of the air supply device according to the optimal air supply angle and wind speed.

[0122] For example, the human body sensing air supply terminal converts the optimal air supply angle and wind speed into control instructions for the air supply device to adjust the air supply parameters of the air supply device. During the adjustment process, the execution status of the air supply device is monitored in real time to ensure that the air supply adapts to the dynamic position changes of multiple targets.

[0123] In the above-mentioned dynamic air supply adjustment method based on human body sensing, multiple sensors such as millimeter wave radar and camera are used to obtain the state and position information of the target human body, and the accuracy of the data is improved through multi-source data fusion; state classification and trajectory prediction, combined with algorithms such as Kalman filtering and LSTM, can effectively predict the movement trend of the target human body in a single-person scenario, and improve the ability to judge the dynamic position of the target human body; through geometric positioning and multi-constraint optimization, it is ensured that the error between the air supply angle and the target human body is less than the preset threshold, thereby improving the comfort of the target human body and the safety of the air supply device; in the face of multi-person scenarios, multi-objective optimization improves the coverage accuracy of the air supply area, improves the air supply effect, balances the comfort of the group, and realizes dynamic air supply.

[0124] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0125] Based on the same inventive concept, embodiments of the present application also provide a human body sensing-based dynamic air supply adjustment device for implementing the aforementioned human body sensing-based dynamic air supply adjustment method. The implementation solution provided by this device is similar to the implementation solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the human body sensing-based dynamic air supply adjustment device provided below can be found in the above-mentioned limitations of the human body sensing-based dynamic air supply adjustment method, and will not be repeated here.

[0126] In an exemplary embodiment, Figure 2 As shown, a dynamic air supply adjustment device 200 based on human body sensing is provided, comprising:

[0127] The data acquisition module 201 is used to acquire multi-source perception data of the target human body;

[0128] The state classification module 202 is used to classify the state of the target human body according to the multi-source sensing data to obtain the state classification result of the target human body; the state classification result includes stillness and movement;

[0129] The trajectory prediction module 203 is used to predict the future motion trajectory of the target body according to the three-dimensional position and state classification results of the target body, and obtain the predicted trajectory of the target body;

[0130] The air supply parameter module 204 is used to calculate the optimal air supply angle and wind speed according to the predicted trajectory to obtain air supply parameters; the air supply parameters include air supply angle and wind speed;

[0131] The air supply control instruction generation module 205 is used to generate an air supply control instruction based on the air supply parameters and send the air supply control instruction to the air supply device; the air supply control instruction is used to instruct the adjustment of the air supply parameters of the air supply device and perform dynamic tracking of air supply.

[0132] Furthermore, the status classification module 202 is further configured to:

[0133] Performing fusion processing on the multi-source perception data to obtain fused multi-source perception data;

[0134] Extract the visual and inertial features of the target human body from the fused multi-source perception data. The visual features include the human body outline and motion trajectory; the inertial features include acceleration and angular velocity.

[0135] Perform bimodal fusion of visual features and inertial features to obtain fused behavioral features;

[0136] The fused behavior features are classified and processed to obtain the state classification results.

[0137] Furthermore, the trajectory prediction module 203 is further configured to:

[0138] Calculate the three-dimensional position of the target body based on the coordinate system of the target body in space;

[0139] Use Kalman filtering to perform short-term trajectory prediction on the three-dimensional position of the target body to obtain the basic predicted trajectory;

[0140] Real-time monitoring of the state classification results and the three-dimensional position of the target human body to obtain monitoring results;

[0141] The basic predicted trajectory is adjusted according to the monitoring results, the deviation of the basic predicted trajectory is corrected, and the predicted trajectory of the target human body is obtained.

[0142] Furthermore, the air supply parameter module 204 is further configured to:

[0143] Based on the predicted trajectory, calculate the basic air supply angle; the basic air supply angle is calculated using the following formula:

[0144]

[0145] Among them, θ t is the air supply angle, x fan 、y fan are the horizontal and vertical coordinates of the air supply device, is the horizontal and vertical coordinates of the predicted position on the predicted trajectory;

[0146] According to the preset influence function of air supply angle and wind speed on comfort, the preset upper limit of wind speed and the constraint conditions of air supply angle, the basic air supply angle and wind speed are adjusted to obtain the air supply parameters.

[0147] Furthermore, the device further comprises:

[0148] Real-time monitoring module, used to monitor the air supply status of the air supply device in real time and obtain the real-time air supply angle and real-time wind speed;

[0149] The deviation calculation module is used to calculate the deviation between the real-time air supply angle and real-time wind speed and the air supply angle and wind speed in the air supply parameters to obtain a deviation result; the deviation result is obtained using the following formula:

[0150] δ result =w1·δ θ +w2·δ v

[0151] Among them, δ result is the deviation result, w1 and w2 are the weight values ​​of air supply angle deviation and wind speed deviation respectively, δ θ is the air supply angle deviation, δ v is the wind speed deviation;

[0152] The air supply angle deviation is obtained using the following formula:

[0153] δ θ =|θ real -θ set |·cos(θ real -θ set )

[0154] Among them, δ θ is the air supply angle deviation, θ real is the real-time air supply angle, θ set is the air supply angle in the air supply parameters;

[0155] The wind speed deviation is obtained using the following formula:

[0156]

[0157] Among them, δ v is the wind speed deviation, v real is the real-time wind speed, v set is the wind speed in the air supply parameters, v min is the preset minimum wind speed, max(·) is the maximum value function, and α is the attenuation coefficient;

[0158] The air supply parameter adjustment module is used to adjust the air supply parameters of the air supply device according to the deviation result.

[0159] Furthermore, the device further comprises:

[0160] The multi-target trajectory prediction module is used to predict the trajectory of each target body when multiple target bodies are detected, and obtain the predicted trajectory position corresponding to each target body;

[0161] The multi-target air supply parameter calculation module is used to calculate the optimal air supply angle and wind speed according to the predicted trajectory position corresponding to each target human body;

[0162] The multi-objective air supply parameter adjustment module is used to adjust the air supply parameters of the air supply device according to the optimal air supply angle and wind speed.

[0163] Furthermore, the air supply device includes at least two air outlets.

[0164] In one embodiment, Figure 3 A computer device 300 is provided, comprising:

[0165] At least one processor 301 and a memory 302 communicatively connected to at least one of the processors 301; the memory storing application code executable by the at least one processor, the application code being executed by the at least one processor so as to enable the at least one processor to perform the steps of the aforementioned method for dynamic air supply adjustment based on human body sensing;

[0166] The computer device may further include: a sensor 303;

[0167] The processor 301, the memory 301 and the sensor 303 may be connected via a bus 304 or other means. In the figure, the bus 304 is used as an example. Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0168] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0169] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0170] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. A dynamic air supply adjustment method based on human body induction, characterized in that: The method comprises: Acquire multi-source perception data of the target human body; Classifying the state of the target human body according to the multi-source perception data to obtain a state classification result of the target human body; the state classification result includes stillness and movement; Predicting a future motion trajectory of the target body according to the three-dimensional position of the target body and the state classification result to obtain a predicted trajectory of the target body; Calculating the optimal air supply angle and wind speed according to the predicted trajectory to obtain air supply parameters; the air supply parameters include the air supply angle and wind speed; According to the air supply parameters, an air supply control instruction is generated and sent to the air supply device; the air supply control instruction is used to instruct adjustment of the air supply parameters of the air supply device to perform dynamic tracking of air supply.

2. The method according to claim 1, characterized in that The step of classifying the target human body into a static state or a motion state based on the multi-source perception data to obtain a state classification result of the target human body includes: Performing fusion processing on the multi-source perception data to obtain fused multi-source perception data; Extracting visual features and inertial features of the target human body from the fused multi-source perception data, wherein the visual features include human body contour and motion trajectory; and the inertial features include acceleration and angular velocity; Performing bimodal fusion on the visual features and the inertial features to obtain fused behavioral features; Classification processing is performed on the fused behavior features to obtain a state classification result.

3. The method according to claim 1, characterized in that The method of predicting the future motion trajectory of the target body according to the three-dimensional position of the target body and the state classification result to obtain the predicted trajectory of the target body includes: Calculating the three-dimensional position of the target body based on the coordinate system of the target body in space; Performing short-term trajectory prediction on the three-dimensional position of the target human body using Kalman filtering to obtain a basic predicted trajectory; Real-time monitoring of the state classification result of the target human body and the three-dimensional position of the target human body to obtain a monitoring result; The basic predicted trajectory is adjusted according to the monitoring result, and the deviation of the basic predicted trajectory is corrected to obtain the predicted trajectory of the target human body.

4. The method according to claim 1, wherein The method of calculating the optimal air supply angle and wind speed according to the predicted trajectory of the target human body to obtain air supply parameters includes: According to the predicted trajectory, the basic air supply angle is calculated; the basic air supply angle is calculated using the following formula: Among them, θ t is the air supply angle, x fan 、y fan are the horizontal and vertical coordinates of the air supply device, is the horizontal and vertical coordinates of the predicted position on the predicted trajectory; According to the preset influence function of the air supply angle and wind speed on comfort, the preset upper limit of wind speed and the constraint condition of the air supply angle, the basic air supply angle and wind speed are adjusted to obtain the air supply parameters.

5. The method according to claim 4, characterized in that The method further comprises: Real-time monitoring of the air supply status of the air supply device to obtain a real-time air supply angle and a real-time wind speed; The real-time air supply angle and the real-time wind speed are subjected to a deviation calculation process with the air supply angle and wind speed in the air supply parameters to obtain a deviation result; the deviation result is obtained using the following formula: d result =w1·δ θ +w2·d v Among them, δ result is the deviation result, w1 and w2 are the weight values ​​of air supply angle deviation and wind speed deviation respectively, δ θ is the air supply angle deviation, δ v is the wind speed deviation; The air supply angle deviation is obtained using the following formula: d θ =|θ real -θ set |·cos(θ real -θ set ) Among them, δ θ is the air supply angle deviation, θ real is the real-time air supply angle, θ set is the air supply angle in the air supply parameters; The wind speed deviation is obtained using the following formula: Among them, δ v is the wind speed deviation, v real is the real-time wind speed, v set is the wind speed in the air supply parameters, v min is the preset minimum wind speed, max(·) is the maximum value function, and α is the attenuation coefficient; Adjust the air supply parameters of the air supply device according to the deviation result.

6. The method according to claim 1, characterized in that The method further comprises: When multiple target human bodies are detected, trajectory prediction is performed for each target human body to obtain the predicted trajectory position corresponding to each target human body; Calculating the optimal air supply angle and wind speed according to the predicted trajectory position corresponding to each target human body; Adjust the air supply parameters of the air supply device according to the optimal air supply angle and wind speed.

7. The method according to claim 1, characterized in that The air supply device includes at least two air outlets.

8. A dynamic air supply adjustment device based on human body induction, characterized in that: The device comprises: A data acquisition module is used to obtain multi-source perception data of the target human body; A state classification module is used to classify the state of the target human body according to the multi-source perception data to obtain a state classification result of the target human body; the state classification result includes stillness and movement; A trajectory prediction module is used to predict the future motion trajectory of the target body according to the three-dimensional position of the target body and the state classification result, so as to obtain the predicted trajectory of the target body; An air supply parameter module is used to calculate the optimal air supply angle and wind speed according to the predicted trajectory to obtain air supply parameters; the air supply parameters include air supply angle and wind speed; The air supply control instruction generation module is used to generate an air supply control instruction based on the air supply parameters and send the air supply control instruction to the air supply device; the air supply control instruction is used to instruct the adjustment of the air supply parameters of the air supply device to perform dynamic tracking of air supply.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images