Situation awareness method and system based on intelligent fan

Abstract

The invention relates to the technical field of sensing data analysis, in particular to a situation sensing method and system based on an intelligent fan. The method comprises the following steps: obtaining human shape position change data of an intelligent fan user and human shape motion posture change data of the intelligent fan user, carrying out human shape motion interframe change identification analysis and Kalman filtering human shape tracking, and carrying out human shape motion range analysis at the same time; the dynamic change range of the human shape motion of the intelligent fan user is obtained; the intelligent sensing range of the intelligent fan is obtained, situation awareness cross judgment is carried out on the human shape motion dynamic change range of an intelligent fan user, and when it is determined that the intelligent fan user is in a close or away sensing state, a wind speed acceleration or deceleration control instruction is generated through response of the intelligent fan and acts on a built-in control system of the intelligent fan. The corresponding intelligent fan air speed automatic acceleration or deceleration work is executed; the method can focus on recognition and tracking of the whole human shape, and aims at optimizing the sensing state of the intelligent fan.

Description

Technical Field

[0001] The present invention relates to the technical field of perception data analysis, and in particular to a situation perception method and system based on intelligent fans. Background Art

[0002] As an important household appliance, smart fans not only have basic wind speed adjustment functions, but also connect with other smart devices through the Internet to provide a more convenient and comfortable user experience. By equipping smart fans with a variety of sensors, such as temperature sensors, humidity sensors, light sensors, and motion sensors, these sensors can monitor environmental conditions and user activity status in real time, collect relevant data, and provide basic information for the fan's intelligent decision-making. Through the embedded computing platform, the data collected by the sensors can be analyzed in real time to identify the user's usage habits and environmental change patterns, so as to make intelligent predictions. For example, the user's posture changes in different time periods can be learned, and the wind speed and wind direction can be automatically adjusted to ensure the best comfort and energy efficiency under different situational awareness conditions. However, traditional smart fans still rely on basic remote controls or mobile phone applications for operation, lacking the ability to perceive the changes in the user's human motion posture in real time, and thus cannot automatically adjust the wind speed of the smart fan according to the user's human activity status (such as approaching, moving away), resulting in a waste of fan energy and a decline in user experience. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide a method and system based on intelligent fan situation awareness to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a situation awareness method based on intelligent fans includes the following steps:

[0005] Step S1: obtaining the humanoid position change data and the humanoid motion posture change data of the smart fan user, and performing humanoid motion frame change recognition analysis on the humanoid position change data and the humanoid motion posture change data of the smart fan user to obtain the humanoid motion change data of the smart fan user;

[0006] Step S2: performing Kalman filtering humanoid tracking on the humanoid motion change data of the smart fan user to obtain the humanoid motion trajectory of the smart fan user; performing humanoid motion range analysis on the humanoid motion trajectory of the smart fan user to obtain the dynamic change range of the humanoid motion of the smart fan user; obtaining the smart fan intelligent perception range, and performing situational perception cross-judgment on the dynamic change range of the humanoid motion of the smart fan user based on the smart fan intelligent perception range to obtain the smart fan user approaching perception state and the smart fan user far away perception state;

[0007] Step S3: when it is determined that the smart fan is in a user approaching sensing state, the smart fan response is used to generate a wind speed acceleration control instruction and acts on the built-in control system of the smart fan to perform the corresponding smart fan wind speed automatic acceleration work;

[0008] Step S4: When it is determined that the smart fan user is away from the sensing state, the smart fan response is used to generate a wind speed deceleration control instruction and act on the built-in control system of the smart fan to execute the corresponding smart fan wind speed automatic deceleration work.

[0009] Further, step S1 includes the following steps:

[0010] Step S11: Acquire the position change data of the human figure of the smart fan user and the movement posture change data of the human figure of the smart fan user;

[0011] Step S12: performing time series synchronization processing on the smart fan user humanoid position change data and the smart fan user humanoid motion posture change data to obtain the user humanoid position change time series data and the user humanoid motion posture change time series data in the same time sequence dimension;

[0012] Step S13: performing a user humanoid three-dimensional space conversion on the user humanoid position change time series data and the user humanoid motion posture change time series data in the same time sequence dimension to obtain a user humanoid action three-dimensional time series change space;

[0013] Step S14: dividing the time frame position and posture distribution of the three-dimensional temporal change space of the user humanoid action to obtain the user humanoid position distribution sub-data and the user humanoid posture distribution sub-data corresponding to each time frame;

[0014] Step S15: performing humanoid action frame-to-frame change recognition analysis on the user humanoid position distribution sub-data and the user humanoid posture distribution sub-data corresponding to each time frame to obtain the intelligent fan user humanoid action motion change data.

[0015] Further, step S13 includes the following steps:

[0016] Step S131: extracting the position and posture change coordinate corner points of the user human figure position change time series data and the user human figure motion posture change time series data in the same time sequence dimension, and obtaining the user human figure position change coordinate corner point set and the user human figure posture change coordinate corner point set in the same time sequence dimension;

[0017] Step S132: performing a three-dimensional space coordinate system geometric transformation on the user humanoid position change coordinate corner point set and the user humanoid posture change coordinate corner point set in the same time sequence dimension to obtain the user humanoid position change coordinate point set and the user humanoid posture change coordinate point set in the three-dimensional space coordinate system;

[0018] Step S133: performing three-dimensional space mapping modeling of humanoid motion according to the coordinate point set of the user humanoid position change and the coordinate point set of the user humanoid posture change in the three-dimensional space coordinate system to obtain an initial three-dimensional space model of the user humanoid motion;

[0019] Step S134: performing humanoid motion smoothing filtering processing on the initial three-dimensional spatial model of the user's humanoid motion to obtain a three-dimensional smooth spatial model of the user's humanoid motion;

[0020] Step S135: Simulate the dynamic time series change of the three-dimensional smooth space model of the user's humanoid action to obtain the three-dimensional time series change space of the user's humanoid action.

[0021] Further, step S15 includes the following steps:

[0022] Step S151: performing position distribution vector analysis on the user humanoid position distribution sub-data corresponding to each time frame to obtain the user humanoid position distribution space sub-vector corresponding to each time frame;

[0023] Step S152: performing inter-frame distance variation statistics calculation on the user figure position distribution space sub-vectors corresponding to each time frame, and obtaining user figure position distance variation values ​​between each adjacent continuous time frame;

[0024] Step S153: performing quantitative calculation of the posture angle change rate between consecutive frames for the user humanoid posture distribution sub-data corresponding to each time frame, so as to obtain the user humanoid posture angle change rate between each adjacent consecutive time frame;

[0025] Step S154: performing inter-frame change recognition and analysis of humanoid motion according to the user humanoid position distance change value and the user humanoid posture angle change rate between each adjacent continuous time frame, and obtaining the intelligent fan user humanoid motion change data.

[0026] Further, step S153 includes the following steps:

[0027] Performing posture Euler angle standardization processing on the user humanoid posture distribution sub-data corresponding to each time frame to obtain the user humanoid posture standard Euler angle corresponding to each time frame;

[0028] The standard Euler angles of the user humanoid posture corresponding to each time frame are analyzed for the posture change amplitude between adjacent continuous frames to obtain the user humanoid posture Euler angle change amplitude between each adjacent continuous time frame;

[0029] The rate of change of the posture angle between consecutive frames is quantitatively calculated according to the change amplitude of the Euler angle of the user humanoid posture between each adjacent continuous time frame, so as to obtain the rate of change of the user humanoid posture angle between each adjacent continuous time frame.

[0030] Further, step S2 includes the following steps:

[0031] Step S21: performing humanoid motion state vector description analysis on the humanoid motion change data of the smart fan user to obtain a humanoid motion state description vector set of the smart fan user, wherein the humanoid motion state description vector set of the smart fan user includes a humanoid motion position state vector of the user, a humanoid motion speed state vector of the user, a humanoid motion acceleration state vector of the user, and a humanoid motion angular velocity state vector of the user;

[0032] Step S22: designing a state transfer equation according to the humanoid motion state description vector set of the smart fan user to generate a user humanoid motion state transfer equation;

[0033] Step S23: performing Kalman filtering humanoid tracking on the motion change data of the humanoid user of the smart fan based on the humanoid motion state transfer equation of the user to obtain the humanoid motion trajectory of the smart fan user;

[0034] Step S24: performing a human motion range analysis on the human motion trajectory of the smart fan user to obtain a dynamic range of human motion change of the smart fan user;

[0035] Step S25: Obtain the intelligent sensing range of the smart fan, and perform situational awareness cross-judgment on the dynamic change range of the human motion of the smart fan user based on the intelligent sensing range of the smart fan, to obtain the smart fan user approaching sensing state and the smart fan user far away sensing state.

[0036] Further, step S23 includes the following steps:

[0037] Step S231: extracting humanoid motion observation parameters from the humanoid motion change data of the smart fan user to obtain a humanoid motion change observation parameter set, wherein the humanoid motion change observation parameter set includes a humanoid motion observation position, a humanoid motion observation speed, a humanoid motion observation acceleration, and a humanoid motion observation angular velocity;

[0038] Step S232: performing humanoid motion state observation estimation on the user humanoid action motion change observation parameter set based on the user humanoid motion state transfer equation to obtain the user humanoid motion state observation estimation;

[0039] Step S233: performing Kalman filter gain update processing on the user humanoid motion state observation estimate based on the user humanoid action motion change observation parameter set to obtain the user humanoid motion state gain optimization estimate;

[0040] Step S234: based on the user humanoid motion state gain optimization estimation, the humanoid motion trajectory of the smart fan user is tracked and reconstructed to obtain the humanoid motion trajectory of the smart fan user.

[0041] Further, step S233 includes the following steps:

[0042] Performing Kalman filtering on the user humanoid motion change observation parameter set to obtain the user humanoid motion observation noise data and the user humanoid motion observation noise filtering parameter set;

[0043] Based on the user humanoid motion observation noise data, the observation noise covariance matrix of the user humanoid motion change observation parameter set is calculated to obtain the user humanoid motion observation noise covariance matrix;

[0044] By introducing system noise data and calculating the Kalman gain of the user humanoid motion observation noise filter parameter set based on the user humanoid motion observation noise covariance matrix, the user humanoid motion observation noise Kalman gain is obtained;

[0045] Based on the Kalman gain of the user's humanoid action motion observation noise, the user's humanoid motion state observation estimation is updated with a state gain to obtain an optimized estimate of the user's humanoid motion state gain.

[0046] Further, step S25 includes the following steps:

[0047] Step S251: calibrate and divide the dynamic range of the human-shaped motion of the user of the smart fan into time-series point ranges to obtain the corresponding relative range of the human-shaped motion of the user at each time-series point;

[0048] Step S252: Obtain the intelligent sensing range of the intelligent fan, and estimate the point situation sensing intersection area for the corresponding relative range of the user humanoid movement at each time point based on the intelligent sensing range of the intelligent fan, so as to obtain the situation sensing intersection area between the user humanoid movement range and the intelligent fan sensing range at each time point;

[0049] Step S253: The time-series area change trend of the situation awareness intersection area between the user's human motion range and the smart fan perception range at each time-series point is judged. If the change trend of the situation awareness intersection area size at each time-series point is gradually increasing, the situation awareness state between the user and the smart fan is judged to be the smart fan user approaching perception state; if the change trend of the situation awareness intersection area size at each time-series point is gradually decreasing, the situation awareness state between the user and the smart fan is judged to be the smart fan user far away perception state.

[0050] Furthermore, the present invention also provides a situation awareness system based on an intelligent fan, which is used to execute the situation awareness method based on an intelligent fan as described above. The situation awareness system based on an intelligent fan includes:

[0051] A user humanoid frame difference recognition module is used to obtain the humanoid position change data of the smart fan user and the humanoid motion posture change data of the smart fan user, and perform humanoid motion frame change recognition analysis on the humanoid position change data of the smart fan user and the humanoid motion posture change data of the smart fan user, so as to obtain the humanoid motion change data of the smart fan user;

[0052] The user human figure situation perception judgment module is used to perform Kalman filtering human figure tracking on the human figure motion change data of the smart fan user to obtain the human figure motion trajectory of the smart fan user; perform human figure motion range analysis on the human figure motion trajectory of the smart fan user to obtain the dynamic change range of the human figure motion of the smart fan user; obtain the smart fan intelligent perception range, and perform situation perception cross judgment on the dynamic change range of the human figure motion of the smart fan user based on the smart fan intelligent perception range, so as to obtain the smart fan user approaching perception state and the smart fan user far away perception state;

[0053] The user proximity sensing fan acceleration control module is used to generate a wind speed acceleration control instruction using the smart fan response and act on the built-in control system of the smart fan to perform the corresponding smart fan wind speed automatic acceleration work when the user proximity sensing state is determined.

[0054] The user-away-from-sensing fan deceleration control module is used to generate a wind speed deceleration control instruction using the smart fan response when determining that the smart fan is in the user-away-from-sensing state, and act on the built-in control system of the smart fan to execute the corresponding smart fan wind speed automatic deceleration work.

[0055] Beneficial effects of the present invention:

[0056] 1. Compared with the prior art, the situation awareness method based on the smart fan proposed in the present invention has the beneficial effect of capturing the user's position information and motion posture by using sensors installed on the smart fan (such as infrared sensors, depth cameras, accelerometers, etc.). The position change data includes the position distance and angle between the user and the smart fan, and the motion posture change data covers the user's body posture (such as sitting, standing, walking, turning, etc.). This step can understand the user's activity status and relative position in real time, thereby providing basic data support for subsequent intelligent interaction. By analyzing these data, the smart fan can better adapt to the needs of the user, such as increasing the wind speed when the user approaches, or automatically reducing the wind speed when the user leaves, thereby providing basic data guarantee for the subsequent processing process. At the same time, by performing humanoid motion frame change recognition and analysis on the humanoid position change data of the smart fan user and the humanoid motion posture change data of the smart fan user, the user's motion changes between different frames can be identified. This process can reveal the dynamic characteristics of user behavior, such as the frequency, amplitude and pattern changes of the motion, etc., which can not only respond to the user's current needs in a timely manner, but also predict the user's future behavior by analyzing historical data. This intelligent analysis not only improves the user's comfort, but also lays the foundation for the self-optimization and intelligence of future smart home devices, and enhances the interactivity between users and smart fans. Secondly, by performing Kalman filtering humanoid tracking on the humanoid motion change data of the smart fan user, Kalman filtering is an effective state estimation method that can process noisy observation data to accurately estimate the user's position and motion trajectory, and can track the user's position in real time in a dynamic environment. This real-time tracking capability enables the smart fan to adjust the wind direction and wind speed in time when the user approaches to provide better comfort and user experience. In addition, the application of Kalman filtering can also significantly improve the system's ability to predict user motion. By continuously updating the state estimation, the influence of sensor noise can be offset to a certain extent, thereby improving the tracking accuracy. By analyzing the human motion range of the human motion trajectory of the smart fan user, it can help understand the user's activity area and behavior pattern. This step not only provides background information for the use of the smart fan, but also provides an important basis for optimizing the fan's working strategy. Through the analysis of the motion trajectory, the user's activity range and habits can be identified. For example, if the user often moves in a certain area, the smart fan can optimize the wind blowing mode based on this information. In the area where the user often moves, the flexibility of adjusting the wind speed and wind direction can be increased to meet the user's comfort needs in the area. Conversely, if the user's activity range is small, the smart fan can reduce energy consumption and extend its service life.It also obtains the intelligent sensing range of the smart fan, and through situational awareness cross-judgment of the dynamic change range of the human motion of the user of the smart fan based on the intelligent sensing range of the smart fan, it can more accurately identify the user's approach and distance status, thereby optimizing its working mechanism, so that the smart fan can flexibly respond to user needs in different environments and usage scenarios. When the user approaches the fan, it can recognize this state and increase the wind speed accordingly to enhance the user's comfort. When the user leaves, the fan can automatically reduce the wind speed or even automatically shut down when the user leaves, thereby achieving energy saving. The implementation of this judgment mechanism also enables the fan to actively adapt to changes in user behavior, and can also improve the real-time perception capability of changes in the user's human motion situation, so that it can automatically adjust the wind speed operation state of the smart fan according to the user's human activity status (such as approaching or moving away). Then, if it is determined that the user is in a proximity sensing state, the smart fan can respond quickly to generate a wind speed acceleration control instruction and act on its built-in control system. This automated response mechanism is an important manifestation of smart homes, enabling smart fans to respond in real time according to changes in user behavior. At this time, the wind speed of the smart fan will automatically accelerate to provide stronger air circulation and adapt to changes in user needs. The advantage of this control strategy is that it improves user comfort. For example, in the hot summer, when the user approaches the fan, the fan immediately accelerates to ensure sufficient cool air and enhance user satisfaction. The smart fan only provides acceleration work when necessary, avoiding unnecessary power consumption, thereby extending the service life of the smart fan. Finally, if it is determined that the user is judged to be away from the perception state, the smart fan can also generate a wind speed deceleration control instruction. This mechanism ensures that the fan automatically adjusts the working mode when the user leaves, thereby reducing the wind speed and reducing energy consumption. The smart fan can not only maintain the user's comfortable experience, but also automatically adjust after the user leaves to keep the room environment suitable. The implementation of this deceleration control instruction has a significant energy-saving effect, so that when the user is not there, the energy consumption of the fan is significantly reduced. For families who use smart devices for a long time, this can effectively reduce electricity bills. The fan's intelligent response reduces the need for human intervention, making users more convenient and comfortable in daily life, thereby improving the overall user experience of smart homes.

[0057] 2. The intelligent fan situation awareness system proposed in the present invention is generally composed of a user human frame difference recognition module, a user human situation perception judgment module, a user approach perception fan acceleration control module and a user distance perception fan deceleration control module. It can implement any intelligent fan situation awareness method described in the present invention, and is used to combine the operations between computer programs running on each module to implement the intelligent fan situation awareness method. The internal structures of the system cooperate with each other, which can greatly reduce duplication of work and manpower investment, and can quickly and effectively provide a more accurate and efficient intelligent fan situation awareness process, thereby simplifying the operation process of the intelligent fan situation awareness system. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0059] Figure 1 It is a schematic diagram of the steps of the intelligent fan situation awareness method according to the present invention;

[0060] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0061] Figure 3 for Figure 2 Detailed step flow chart of step S13 in FIG. DETAILED DESCRIPTION

[0062] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0063] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0064] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0065] To achieve this, please refer to Figures 1 to 3 The present invention provides a situation awareness method based on intelligent fans, the method comprising the following steps:

[0066] Step S1: obtaining the humanoid position change data and the humanoid motion posture change data of the smart fan user, and performing humanoid motion frame change recognition analysis on the humanoid position change data and the humanoid motion posture change data of the smart fan user to obtain the humanoid motion change data of the smart fan user;

[0067] Step S2: performing Kalman filtering humanoid tracking on the humanoid motion change data of the smart fan user to obtain the humanoid motion trajectory of the smart fan user; performing humanoid motion range analysis on the humanoid motion trajectory of the smart fan user to obtain the dynamic change range of the humanoid motion of the smart fan user; obtaining the smart fan intelligent perception range, and performing situational perception cross-judgment on the dynamic change range of the humanoid motion of the smart fan user based on the smart fan intelligent perception range to obtain the smart fan user approaching perception state and the smart fan user far away perception state;

[0068] Step S3: when it is determined that the smart fan is in a user approaching sensing state, the smart fan response is used to generate a wind speed acceleration control instruction and acts on the built-in control system of the smart fan to perform the corresponding smart fan wind speed automatic acceleration work;

[0069] Step S4: When it is determined that the smart fan user is away from the sensing state, the smart fan response is used to generate a wind speed deceleration control instruction and act on the built-in control system of the smart fan to execute the corresponding smart fan wind speed automatic deceleration work.

[0070] In the embodiment of the present invention, please refer to Figure 1 FIG. 1 is a schematic diagram of a process flow of a method for situation awareness based on an intelligent fan according to the present invention. In this example, the method for situation awareness based on an intelligent fan includes the following steps:

[0071] Step S1: obtaining the humanoid position change data and the humanoid motion posture change data of the smart fan user, and performing humanoid motion frame change recognition analysis on the humanoid position change data and the humanoid motion posture change data of the smart fan user to obtain the humanoid motion change data of the smart fan user;

[0072] In an embodiment of the present invention, the position and posture information of the user is obtained by using an infrared sensor and a depth camera, wherein the infrared sensor measures the distance between the user and the smart fan in real time by emitting infrared rays and receiving reflected signals, thereby determining the position change of the user, and the depth camera captures the dynamic posture of the user through stereoscopic vision, and obtains the relative motion posture change positions of various parts of the user's body by analyzing the depth information in the image, thereby obtaining the humanoid position change data of the smart fan user and the humanoid motion posture change data of the smart fan user. The previously acquired humanoid position change data of the smart fan user and the humanoid motion posture change data of the smart fan user are synchronized by using a timestamp, so as to ensure the consistency of the time dimension of the data by adding a timestamp to the data collected by each sensor, and smooth the time series data by an interpolation method or a filtering algorithm to eliminate noise and irregular fluctuations in the data, and at the same time, the humanoid position change time series data and the motion posture change time series data previously obtained after time series synchronization are converted into a three-dimensional space, so as to convert the two-dimensional depth data into a three-dimensional space of the user humanoid in a three-dimensional coordinate system by using a three-dimensional reconstruction algorithm, and generate a three-dimensional motion trajectory space containing time, position and posture information, and the previously acquired three-dimensional conversion is performed by setting a certain time interval. The three-dimensional time series change space obtained is then divided into time frame distribution, so as to extract the corresponding user humanoid position distribution and posture distribution for each time frame, and then, the inter-frame changes of humanoid motion are recognized and analyzed by the humanoid position distribution and posture distribution in each time frame, so as to train the divided sub-data by adopting machine learning algorithms, such as support vector machine (SVM) or convolutional neural network (CNN), and identify the user's motion changes in different time frames by extracting motion features to form user humanoid motion motion change data, and use data statistical tools to further analyze the identified motion change data, including motion change parameters such as motion position, speed, acceleration and angular velocity, and finally obtain the smart fan user humanoid motion motion change data.

[0073] Step S2: performing Kalman filtering humanoid tracking on the humanoid motion change data of the smart fan user to obtain the humanoid motion trajectory of the smart fan user; performing humanoid motion range analysis on the humanoid motion trajectory of the smart fan user to obtain the dynamic change range of the humanoid motion of the smart fan user; obtaining the smart fan intelligent perception range, and performing situational perception cross-judgment on the dynamic change range of the humanoid motion of the smart fan user based on the smart fan intelligent perception range to obtain the smart fan user approaching perception state and the smart fan user far away perception state;

[0074] In an embodiment of the present invention, a statistical analysis of humanoid motion state parameters is performed by combining the motion change data of the humanoid action of the smart fan user obtained by the previous motion change analysis, and the user's position, speed, acceleration and angular velocity are obtained by statistical analysis, and these humanoid motion state parameters and timestamps are converted into description vectors, thereby converting the humanoid motion state description vector set constituting the user. After obtaining the user humanoid motion state description vector set, the corresponding description vector and the disturbance term are used to construct a state transfer equation. The state transfer equation uses the kinematic principle in physics to predict the state of the next time step based on the current motion state description, and selects a suitable mathematical model, such as a linear model or a nonlinear model, according to the characteristics of the system and the application scenario. For example, the state transfer equation is specifically:

[0075]

[0076] Where X(t) is the user position state vector, expressed as X(t) = [x(t), y(t), z(t)] T , where T is the transposed symbol, usually representing the three-dimensional coordinate position of the user at time t, and v(t) is the user velocity state vector, which can be expressed as v(t) = [v x (t),v y (t),v z (t)] T , usually represents the user's three-dimensional coordinate direction velocity at time t, a(t) is the user's acceleration state vector, and is expressed as a(t) = [a x (t),a y (t),a z (t)] T , usually represents the acceleration of the user in the three-dimensional coordinate direction at time t, ω(ω) is the user's angular velocity state vector, which can be expressed as ω(t)=[ω x (t),ω y (t),ω z (t)] T, usually represents the angular velocity of the user's three-dimensional coordinate direction at time t, Δt is the time step, τ is the integral time variable parameter, J(τ) is the motion position state disturbance term, W(τ) is the velocity state disturbance term, F(t) is the external force state vector affecting acceleration at time t, M(τ) is the acceleration additional disturbance term, I ―1 is the inverse matrix of the user's moment of inertia, T(t) is the external torque applied to the user at time t, R(τ) is the random disturbance term affecting the angular acceleration, and the previously analyzed and collected intelligent fan user humanoid motion change data is applied to the Kalman filter algorithm by combining the previously designed user humanoid motion state transfer equation, so as to continuously update the state estimate through a series of measurement values, and use the Kalman gain to balance the difference between the predicted value and the actual measurement value. The implementation process of the algorithm includes a prediction stage and an update stage. The prediction stage uses the previous state vector and transfer equation to predict the current state. The update stage adjusts the predicted state vector and random disturbance terms by combining the new measurement value to obtain the corresponding motion state gain update estimate, and tracks and reconstructs the humanoid motion trajectory. Starting from the gain update estimate obtained in the previous step, a trajectory reconstruction algorithm (such as the least squares method or the spline interpolation method) is used to continuously track and model the user's motion state, thereby obtaining the intelligent fan user humanoid motion trajectory. Secondly, the motion range of the smart fan user humanoid motion trajectory obtained after previous tracking and reconstruction is statistically analyzed by time-series segmentation to discretize the user's humanoid motion trajectory and divide the continuous motion trajectory into several segments for easy analysis. In each motion segment, the user's motion range is calculated, which can be achieved through the maximum and minimum coordinate values. At the same time, combined with the user's motion mode (such as still, walking, running, etc.), the envelope area of ​​the user's motion can be drawn. The shape and area of ​​the area are analyzed through geometric methods, and then the user's motion dynamic change range is identified, thereby obtaining the dynamic change range of the smart fan user's humanoid motion. Then, by determining the intelligent perception range of the smart fan, which can be achieved through sensors installed on the smart fan (such as infrared sensors, ultrasonic sensors, etc.), at the same time, by combining the previously determined intelligent perception range of the smart fan, the user's dynamic motion change range is subjected to situational awareness judgment analysis, so as to define the effective perception area of ​​the smart fan according to the intelligent perception range. The working range of the sensor (such as infrared, ultrasonic or other sensors) is usually used for modeling, and the user's dynamic motion change range is cross-compared with the perception area of ​​the smart fan for judgment. By setting a threshold, it is judged whether the user is in a state of approaching or moving away. If the intersection area of ​​the user's motion position in the perception area of ​​the smart fan becomes larger and larger, the smart fan is judged as "the smart fan user is in a state of approaching perception"; conversely, if the intersection area of ​​the user's motion position in the perception area of ​​the smart fan becomes smaller and smaller, it is judged as "the smart fan user is in a state of moving away".

[0077] Step S3: when it is determined that the smart fan is in a user approaching sensing state, the smart fan response is used to generate a wind speed acceleration control instruction and acts on the built-in control system of the smart fan to perform the corresponding smart fan wind speed automatic acceleration work;

[0078] In an embodiment of the present invention, if it is determined that the user is judged to be in a proximity sensing state, the intelligent fan responds by generating a wind speed acceleration control instruction. The instruction is directly connected to the fan motor through the built-in control system to quickly adjust the wind speed. The control system uses PWM (pulse width modulation) technology to accurately control the speed of the fan motor, thereby achieving the required wind speed acceleration. In specific implementation, the control system dynamically adjusts the input voltage of the motor by setting the ratio of wind speed increase to achieve smooth acceleration of the wind speed. During this process, the intelligent fan can provide real-time feedback on the user's position changes to ensure that the wind speed can respond in time when the user approaches, thereby enhancing the user's comfort experience and ultimately executing the corresponding intelligent fan automatic wind speed acceleration work.

[0079] Step S4: When it is determined that the smart fan user is away from the sensing state, the smart fan response is used to generate a wind speed deceleration control instruction and act on the built-in control system of the smart fan to execute the corresponding smart fan wind speed automatic deceleration work.

[0080] In an embodiment of the present invention, if it is determined that the user is judged to be in a away perception state, the smart fan responds by generating a wind speed deceleration control instruction. This instruction also acts on the fan motor through the built-in control system to achieve automatic deceleration of the wind speed. The control system uses a similar PWM technology to set the deceleration amplitude to ensure that the motor gradually reduces the speed. In specific operations, by real-time monitoring of the relative position changes between the user and the fan, the deceleration time delay is adjusted to ensure a natural reduction in wind speed. At this time, the smart fan can effectively reduce energy consumption and provide a more suitable environment, ensuring that the user no longer feels the interference of strong wind when away, and finally executing the corresponding smart fan automatic wind speed deceleration work.

[0081] Further, step S1 includes the following steps:

[0082] Step S11: Acquire the position change data of the human figure of the smart fan user and the movement posture change data of the human figure of the smart fan user;

[0083] Step S12: performing time series synchronization processing on the smart fan user humanoid position change data and the smart fan user humanoid motion posture change data to obtain the user humanoid position change time series data and the user humanoid motion posture change time series data in the same time sequence dimension;

[0084] Step S13: performing a user humanoid three-dimensional space conversion on the user humanoid position change time series data and the user humanoid motion posture change time series data in the same time sequence dimension to obtain a user humanoid action three-dimensional time series change space;

[0085] Step S14: dividing the time frame position and posture distribution of the three-dimensional temporal change space of the user humanoid action to obtain the user humanoid position distribution sub-data and the user humanoid posture distribution sub-data corresponding to each time frame;

[0086] Step S15: performing humanoid action frame-to-frame change recognition analysis on the user humanoid position distribution sub-data and the user humanoid posture distribution sub-data corresponding to each time frame to obtain the intelligent fan user humanoid action motion change data.

[0087] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in FIG. 1 , in this embodiment, step S1 includes the following steps:

[0088] Step S11: Acquire the position change data of the human figure of the smart fan user and the movement posture change data of the human figure of the smart fan user;

[0089] In an embodiment of the present invention, the position and posture information of the user is obtained by utilizing an infrared sensor and a depth camera. The infrared sensor measures the distance between the user and the smart fan in real time by emitting infrared rays and receiving reflected signals, thereby determining the position change of the user. The depth camera captures the dynamic posture of the user through stereoscopic vision, and obtains the relative motion posture change position of various parts of the user's body by analyzing the depth information in the image, thereby finally obtaining the human position change data of the smart fan user and the human motion posture change data of the smart fan user.

[0090] Step S12: performing time series synchronization processing on the smart fan user humanoid position change data and the smart fan user humanoid motion posture change data to obtain the user humanoid position change time series data and the user humanoid motion posture change time series data in the same time sequence dimension;

[0091] In an embodiment of the present invention, the previously acquired smart fan user human shape position change data and the smart fan user human shape motion posture change data are synchronously processed in time series by using a timestamp, so as to add a timestamp to the data collected by each sensor to ensure the consistency of the time dimension of the data, and smooth the time series data by interpolation or filtering algorithm to eliminate noise and irregular fluctuations in the data, so as to integrate the synchronized data into a unified data set to form a human shape position change time series data and motion posture change time series data arranged in time sequence, and finally obtain the user human shape position change time series data and user human shape motion posture change time series data under the same time sequence dimension.

[0092] Step S13: performing a user humanoid three-dimensional space conversion on the user humanoid position change time series data and the user humanoid motion posture change time series data in the same time sequence dimension to obtain a user humanoid action three-dimensional time series change space;

[0093] In an embodiment of the present invention, the user humanoid position change timing data and motion posture change timing data obtained after time synchronization in the same time sequence dimension are converted into three-dimensional space, so that the two-dimensional depth data is converted into the user humanoid three-dimensional space in a three-dimensional coordinate system by utilizing a three-dimensional reconstruction algorithm, and based on computer vision technology, combined with the user's motion posture obtained by a motion capture system, a three-dimensional motion trajectory space containing time, position and posture information is generated. The trajectory shows the dynamic changes of the user in the three-dimensional space, can more intuitively reflect the user's action performance, and finally obtains the three-dimensional time series change space of the user humanoid action.

[0094] Step S14: dividing the time frame position and posture distribution of the three-dimensional temporal change space of the user humanoid action to obtain the user humanoid position distribution sub-data and the user humanoid posture distribution sub-data corresponding to each time frame;

[0095] In an embodiment of the present invention, a certain time interval is set to divide the distribution of time frames for the three-dimensional temporal change space of the user humanoid action obtained after the three-dimensional conversion, so as to extract the corresponding user humanoid position distribution and posture distribution for each time frame, thereby ensuring that the user behavior characteristics in each time frame are clearly expressed, and finally obtaining the corresponding user humanoid position distribution sub-data and user humanoid posture distribution sub-data in each time frame.

[0096] Step S15: performing humanoid action frame-to-frame change recognition analysis on the user humanoid position distribution sub-data and the user humanoid posture distribution sub-data corresponding to each time frame to obtain the intelligent fan user humanoid action motion change data.

[0097] In an embodiment of the present invention, by performing recognition and analysis of inter-frame changes in humanoid motion for user humanoid position distribution sub-data and posture distribution sub-data in each time frame, the divided sub-data are trained by adopting a machine learning algorithm, such as a support vector machine (SVM) or a convolutional neural network (CNN), and by extracting motion features, the user's motion changes in different time frames are identified to form user humanoid motion motion change data, and using data statistical tools, the identified motion change data is further analyzed, including motion change parameters such as motion position, speed, acceleration, and angular velocity, to finally obtain the smart fan user humanoid motion motion change data.

[0098] Further, step S13 includes the following steps:

[0099] Step S131: extracting the position and posture change coordinate corner points of the user human figure position change time series data and the user human figure motion posture change time series data in the same time sequence dimension, and obtaining the user human figure position change coordinate corner point set and the user human figure posture change coordinate corner point set in the same time sequence dimension;

[0100] Step S132: performing a three-dimensional space coordinate system geometric transformation on the user humanoid position change coordinate corner point set and the user humanoid posture change coordinate corner point set in the same time sequence dimension to obtain the user humanoid position change coordinate point set and the user humanoid posture change coordinate point set in the three-dimensional space coordinate system;

[0101] Step S133: performing three-dimensional space mapping modeling of humanoid motion according to the coordinate point set of the user humanoid position change and the coordinate point set of the user humanoid posture change in the three-dimensional space coordinate system to obtain an initial three-dimensional space model of the user humanoid motion;

[0102] Step S134: performing humanoid motion smoothing filtering processing on the initial three-dimensional spatial model of the user's humanoid motion to obtain a three-dimensional smooth spatial model of the user's humanoid motion;

[0103] Step S135: Simulate the dynamic time series change of the three-dimensional smooth space model of the user's humanoid action to obtain the three-dimensional time series change space of the user's humanoid action.

[0104] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 2 FIG. 1 is a schematic diagram of a detailed step flow chart of step S13 in FIG. 1 . In this embodiment, step S13 includes the following steps:

[0105] Step S131: extracting the position and posture change coordinate corner points of the user human figure position change time series data and the user human figure motion posture change time series data in the same time sequence dimension, and obtaining the user human figure position change coordinate corner point set and the user human figure posture change coordinate corner point set in the same time sequence dimension;

[0106] In an embodiment of the present invention, a corner point detection algorithm is used to sample the coordinate corner points of the user humanoid position change time series data and the user humanoid motion posture change time series data obtained after time synchronization in the same time sequence dimension, so as to accurately capture the user's spatial position coordinates and motion posture coordinates at different time points, and extract the user's position coordinates (such as x, y, z coordinates) and posture changes (such as joint angles, body posture, etc.) corresponding to each time point through a data processing algorithm, and finally sample to obtain the user humanoid position change coordinate corner point set and the user humanoid posture change coordinate corner point set in the same time sequence dimension.

[0107] Step S132: performing a three-dimensional space coordinate system geometric transformation on the user humanoid position change coordinate corner point set and the user humanoid posture change coordinate corner point set in the same time sequence dimension to obtain the user humanoid position change coordinate point set and the user humanoid posture change coordinate point set in the three-dimensional space coordinate system;

[0108] In an embodiment of the present invention, a geometric transformation of a three-dimensional coordinate system is performed on a set of coordinate corner points of a user's humanoid position change and a set of coordinate corner points of a user's humanoid posture change in the same time sequence dimension obtained after coordinate corner point extraction, so as to establish a standard three-dimensional space coordinate system, and the original two-dimensional or local coordinates are converted into global three-dimensional coordinates by means of homogeneous coordinates or transformation matrices. The tools used include mathematical software (such as MATLAB or Python's NumPy library), which can realize matrix operations and coordinate transformations. In specific operations, the origin and coordinate axis directions in the space are first defined, and then the user's coordinate points are transformed using transformation matrices such as translation, rotation and scaling, and finally the user's humanoid position change coordinate point set and the user's humanoid posture change coordinate point set in the three-dimensional space coordinate system are obtained.

[0109] Step S133: performing three-dimensional space mapping modeling of humanoid motion according to the coordinate point set of the user humanoid position change and the coordinate point set of the user humanoid posture change in the three-dimensional space coordinate system to obtain an initial three-dimensional space model of the user humanoid motion;

[0110] In an embodiment of the present invention, three-dimensional space mapping modeling is performed by combining the coordinate point set of the user's humanoid position change and the coordinate point set of the user's humanoid posture change in the three-dimensional space coordinate system after previous conversion, so as to construct the user's initial three-dimensional space model by connecting and interpolating the converted coordinate point sets using a modeling tool that may include three-dimensional modeling software (such as Blender or Autodesk Maya) or a dedicated computer graphics library (such as OpenGL). This process includes defining the model's geometric shape, bone structure and motion trajectory, so that the user's movements can be realistically mapped to the three-dimensional model. At the same time, during the modeling process, attention is paid to maintaining the details of the model to ensure the continuity between various parts, and finally the user's humanoid movement initial three-dimensional space model is constructed.

[0111] Step S134: performing humanoid motion smoothing filtering processing on the initial three-dimensional spatial model of the user's humanoid motion to obtain a three-dimensional smooth spatial model of the user's humanoid motion;

[0112] In an embodiment of the present invention, a smoothing filtering process is performed on the initial three-dimensional spatial model of the user's humanoid motion obtained by the previous mapping construction, aiming to remove jitter and discontinuity in the model and enhance the naturalness of the motion performance. A low-pass filter or a smoothing algorithm (such as Bezier curve, Kalman filter, etc.) can be used for processing. This process requires the introduction of time series analysis technology to smooth the motion trajectory of each coordinate point. In a specific implementation, each point of the model is first iteratively calculated, and a weighted average is performed based on the coordinate values ​​of the previous and subsequent time points to reduce the impact of mutations and outliers. Through such smoothing processing, the obtained three-dimensional smooth spatial model can better reflect the fluency and naturalness of the actual motion, and finally a three-dimensional smooth spatial model of the user's humanoid motion is obtained.

[0113] Step S135: Simulate the dynamic time series change of the three-dimensional smooth space model of the user's humanoid action to obtain the three-dimensional time series change space of the user's humanoid action.

[0114] In an embodiment of the present invention, a dynamic time-series change simulation analysis is performed on the three-dimensional smooth space model of the user's humanoid action obtained after smoothing, aiming to generate a motion change space that conforms to the law of time evolution. This process can use animation production software (such as Unity or Unreal Engine), through key frame animation technology and interpolation methods, based on the previously obtained smooth model, to create a series of dynamic effects. During the simulation process, the timeline is associated with each key posture of the model to ensure that the transition between actions is smooth and natural. In addition, the physical engine is used to perform real-time motion simulation of the model to take into account the influence of factors such as gravity and collision, thereby generating a more realistic motion performance, and finally obtaining a three-dimensional time-series change space of the user's humanoid action.

[0115] Further, step S15 includes the following steps:

[0116] Step S151: performing position distribution vector analysis on the user humanoid position distribution sub-data corresponding to each time frame to obtain the user humanoid position distribution space sub-vector corresponding to each time frame;

[0117] In an embodiment of the present invention, the user human figure position distribution sub-data corresponding to each time frame obtained by previously dividing the time frame is subjected to vector conversion of the position distribution, so as to convert the user position coordinates (such as two-dimensional or three-dimensional coordinates) in each time frame into a vector form. This process can be achieved by a vectorization algorithm, such as using the principal component analysis (PCA) method to perform dimensionality reduction processing on the position distribution, extract the main eigenvectors of the user position distribution, and finally obtain the user human figure position distribution space sub-vector corresponding to each time frame.

[0118] Step S152: performing inter-frame distance variation statistics calculation on the user figure position distribution space sub-vectors corresponding to each time frame, and obtaining user figure position distance variation values ​​between each adjacent continuous time frame;

[0119] In the embodiment of the present invention, after completing the extraction of the user humanoid position distribution space sub-vectors corresponding to each time frame, it is necessary to perform statistical calculation of the vector distance of the user humanoid position between adjacent continuous time frames, so as to obtain the displacement of the user in different time periods by calculating the Euclidean distance between the space sub-vectors of adjacent time frames. In the specific implementation, the space sub-vector of each time frame is first represented as a mathematical vector P t and P t―1 , and then calculate the distance d = ‖P t ―P t―1 ‖, this operation can be implemented using a vector operation library such as NumPy, and finally the change value of the user humanoid position distance between each adjacent continuous time frame is obtained.

[0120] Step S153: performing quantitative calculation of the posture angle change rate between consecutive frames for the user humanoid posture distribution sub-data corresponding to each time frame, so as to obtain the user humanoid posture angle change rate between each adjacent consecutive time frame;

[0121] In the embodiment of the present invention, the three-dimensional posture coordinates of each joint are collected from the user humanoid posture distribution sub-data corresponding to each time frame obtained by the previous division, and the user's posture is calculated to be expressed as Euler angles (i.e., roll angle, pitch angle and yaw angle), and is standardized by the following calculation formula: The standard Euler angles of the user humanoid posture corresponding to each time frame that has been standardized before are statistically calculated for the posture change amplitude between adjacent continuous time frames, so as to calculate the Euler angle change amplitude between each time frame by the following adjacent continuous frame change amplitude calculation method, for example, Δθ = |θ t ―θ t―1 |, where Δθ represents the magnitude of the posture change between adjacent time frames, θ t and θ t―1 are the attitude Euler angle values ​​of the current and previous time frames respectively. Then, the attitude angle change rate between adjacent continuous time frames is quantitatively calculated by combining the Euler angle change amplitude obtained by the previous quantitative calculation and the corresponding time interval, so as to calculate the rate through the corresponding rate calculation formula ΔT is the time interval between adjacent time frames, which is usually determined by the sensor sampling frequency. On this basis, the change amplitude Δθ is divided by the time interval to finally obtain the user humanoid posture angle change rate between each adjacent continuous time frame.

[0122] Step S154: performing inter-frame change recognition and analysis of humanoid motion according to the user humanoid position distance change value and the user humanoid posture angle change rate between each adjacent continuous time frame, and obtaining the intelligent fan user humanoid motion change data.

[0123] In an embodiment of the present invention, by combining the user figure position distance change value and the user figure posture angle change rate between each adjacent continuous time frame obtained by previous quantitative calculation, the change recognition analysis of the user human figure motion of the smart fan user is performed, so as to define a threshold to determine whether the user motion has changed significantly. For example, when the position change exceeds a set distance and the posture angle change rate exceeds a specific range, it can be identified that the user has a new motion movement change trend. At this time, mathematical statistical tools can be used to further analyze the identified motion change data, including motion change parameters such as motion position, speed, acceleration and angular velocity, and finally obtain the smart fan user human figure motion change data.

[0124] Further, step S153 includes the following steps:

[0125] Performing posture Euler angle standardization processing on the user humanoid posture distribution sub-data corresponding to each time frame to obtain the user humanoid posture standard Euler angle corresponding to each time frame;

[0126] In the embodiment of the present invention, the three-dimensional posture coordinates of each joint are collected from the user humanoid posture distribution sub-data corresponding to each time frame obtained by the previous division, and the user's posture is calculated to be expressed as Euler angles (i.e., roll angle, pitch angle and yaw angle), and is standardized by the following calculation formula: In this formula, the original angle is standardized after adjusting the mean and standard deviation to ensure that the posture angle distribution in each time frame is concentrated in the same range, thereby eliminating the deviation caused by individual differences, and finally obtaining the standard Euler angle of the user's humanoid posture corresponding to each time frame.

[0127] Preferably, the standard Euler angles of the user humanoid posture corresponding to each time frame are analyzed for the posture change amplitude between adjacent continuous frames to obtain the user humanoid posture Euler angle change amplitude between each adjacent continuous time frame;

[0128] In the embodiment of the present invention, the standard Euler angles of the humanoid posture of the user corresponding to each time frame that has been previously standardized are statistically calculated for the posture change amplitude between adjacent continuous time frames, so as to calculate the Euler angle change amplitude between each time frame by the following adjacent continuous frame change amplitude calculation method, for example, Δθ=|θ t ―θ t―1 |, where Δθ represents the magnitude of the posture change between adjacent time frames, θ t and θ t―1 are the posture Euler angle values ​​of the current and previous time frames respectively. By cyclically calculating all time frames, the change amplitude of the user humanoid posture Euler angle between each adjacent continuous time frame is finally obtained.

[0129] Preferably, the rate of change of the posture angle between consecutive frames is quantitatively calculated according to the change amplitude of the Euler angle of the user humanoid posture between each adjacent consecutive time frame, so as to obtain the rate of change of the user humanoid posture angle between each adjacent consecutive time frame.

[0130] In the embodiment of the present invention, the change amplitude of the Euler angle of the user's humanoid posture between each adjacent continuous time frame obtained by the previous quantitative calculation is combined with the corresponding time interval to perform a quantitative calculation of the change rate of the posture angle between adjacent continuous time frames, so as to calculate the corresponding rate through the corresponding rate calculation formula ΔT is the time interval between adjacent time frames, which is usually determined by the sensor sampling frequency. On this basis, the change amplitude Δθ is divided by the time interval to obtain the posture change rate between each time frame, and finally the user humanoid posture angle change rate between each adjacent continuous time frame is obtained.

[0131] Further, step S2 includes the following steps:

[0132] Step S21: performing humanoid motion state vector description analysis on the humanoid motion change data of the smart fan user to obtain a humanoid motion state description vector set of the smart fan user, wherein the humanoid motion state description vector set of the smart fan user includes a humanoid motion position state vector of the user, a humanoid motion speed state vector of the user, a humanoid motion acceleration state vector of the user, and a humanoid motion angular velocity state vector of the user;

[0133] In an embodiment of the present invention, a statistical analysis of humanoid motion state parameters is performed by combining the motion change data of a humanoid action of a smart fan user obtained by a previous motion change analysis, and the user's position, speed, acceleration and angular velocity are obtained by statistical analysis, and these humanoid motion state parameters and timestamps are converted into description vectors, thereby converting the humanoid motion state description vector set that constitutes the user, ensuring that each vector can accurately reflect the user's motion state in space, including the user's position information (user humanoid motion position state vector), motion rate (user humanoid motion speed state vector), motion acceleration change (user humanoid motion acceleration state vector) and steering change (user humanoid motion angular velocity state vector), thereby forming a complete set of motion state descriptions, and finally obtaining a smart fan user humanoid motion state description vector set.

[0134] Step S22: designing a state transfer equation according to the humanoid motion state description vector set of the smart fan user to generate a user humanoid motion state transfer equation;

[0135] In the embodiment of the present invention, after obtaining the user humanoid motion state description vector set, the corresponding description vector and the disturbance term are used to construct a state transfer equation. The state transfer equation uses the kinematic principle in physics to predict the state of the next time step based on the current motion state description, and selects a suitable mathematical model, such as a linear model or a nonlinear model, according to the characteristics of the system and the application scenario. For example, the state transfer equation is specifically:

[0136]

[0137] Where X(t) is the user position state vector, expressed as X(t) = [x(t), y(t), z(t)] T , where T is the transposed symbol, usually representing the three-dimensional coordinate position of the user at time t, and v(t) is the user velocity state vector, which can be expressed as v(t) = [v x (t),v y (t),v z (t)] T , usually represents the user's three-dimensional coordinate direction velocity at time t, a(t) is the user's acceleration state vector, and is expressed as a(t) = [a x (t),ay (t),a z (t)] T , usually represents the acceleration of the user in the three-dimensional coordinate direction at time t, ω(t) is the user's angular velocity state vector, which can be expressed as ω(t)=[ω x (t),ω y (t),ω z (t)] T , usually represents the angular velocity of the user's three-dimensional coordinate direction at time t, Δt is the time step, τ is the integral time variable parameter, J(τ) is the motion position state disturbance term, W(τ) is the velocity state disturbance term, F(t) is the external force state vector affecting acceleration at time t, M(τ) is the acceleration additional disturbance term, I ―1 is the inverse matrix of the user's moment of inertia, T(t) is the external torque applied to the user at time t, and R(τ) is the random disturbance term affecting the angular acceleration. Finally, a usable user humanoid motion state transfer equation is designed to be generated.

[0138] Step S23: performing Kalman filtering humanoid tracking on the motion change data of the humanoid user of the smart fan based on the humanoid motion state transfer equation of the user to obtain the humanoid motion trajectory of the smart fan user;

[0139] In an embodiment of the present invention, the humanoid motion change data of the user of the smart fan previously analyzed and collected is applied to the Kalman filter algorithm by combining the previously designed user humanoid motion state transfer equation, so as to continuously update the state estimate through a series of measurement values, and use the Kalman gain to balance the difference between the predicted value and the actual measurement value. The implementation process of the algorithm includes a prediction stage and an update stage. The prediction stage uses the previous state vector and transfer equation to predict the current state. The update stage adjusts the predicted state vector and the random disturbance term by combining the new measurement value, so as to obtain the corresponding motion state gain update estimate, so as to track and reconstruct the humanoid motion trajectory. Starting from the gain update estimate obtained in the previous step, a trajectory reconstruction algorithm (such as the least squares method or the spline interpolation method) is used to continuously track and model the user's motion state, and finally obtain the humanoid motion trajectory of the smart fan user.

[0140] Step S24: performing a human motion range analysis on the human motion trajectory of the smart fan user to obtain a dynamic range of human motion change of the smart fan user;

[0141] In an embodiment of the present invention, a statistical analysis of the motion range of the humanoid motion trajectory of the smart fan user obtained after previous tracking and reconstruction is performed by time-series segmentation, so as to achieve discretization processing of the humanoid motion trajectory of the user, and divide the continuous motion trajectory into several segments for easy analysis. In each motion segment, the motion range of the user is calculated, which can be specifically achieved through the maximum and minimum coordinate values. At the same time, combined with the user's motion mode (such as stillness, walking, running, etc.), the envelope area of ​​the user's motion can be drawn. Through geometric methods, the shape and area of ​​the area are analyzed, and then the dynamic change range of the user's motion is identified, thereby obtaining the user's activity space in a specific environment, and finally obtaining the dynamic change range of the humanoid motion of the smart fan user.

[0142] Step S25: Obtain the intelligent sensing range of the smart fan, and perform situational awareness cross-judgment on the dynamic change range of the human motion of the smart fan user based on the intelligent sensing range of the smart fan, to obtain the smart fan user approaching sensing state and the smart fan user far away sensing state.

[0143] In an embodiment of the present invention, by determining the intelligent sensing range of the intelligent fan, this can be achieved by using a sensor (such as an infrared sensor, an ultrasonic sensor, etc.) installed on the intelligent fan. The sensor monitors the surrounding environment in real time and draws a two-dimensional or three-dimensional sensing space range of a sensing range, thereby obtaining the intelligent sensing range of the intelligent fan. At the same time, by combining the previously determined intelligent sensing range of the intelligent fan, the dynamic range of the user's movement is subjected to situational awareness judgment and analysis, so as to define the effective sensing area of ​​the intelligent fan according to the intelligent sensing range. The working range of the sensor (such as infrared, ultrasonic or other sensors) is usually used for modeling, and the dynamic range of the user's movement is cross-compared with the sensing area of ​​the intelligent fan for judgment. By setting a threshold, it is judged whether the user is in a state of approaching or moving away. If the intersection area of ​​the user's movement position in the sensing area of ​​the intelligent fan is getting larger and larger, the intelligent fan is judged as "approaching sensing state"; on the contrary, if the intersection area of ​​the user's movement position in the sensing area of ​​the intelligent fan is getting smaller and smaller, it is judged as "moving away sensing state". According to these states, the intelligent fan can adjust the working mode in real time, for example, automatically increase the wind speed when the user approaches, and appropriately reduce the wind speed when the user moves away, to ensure the user's comfortable experience at different distances.

[0144] Further, step S23 includes the following steps:

[0145] Step S231: extracting humanoid motion observation parameters from the humanoid motion change data of the smart fan user to obtain a humanoid motion change observation parameter set, wherein the humanoid motion change observation parameter set includes a humanoid motion observation position, a humanoid motion observation speed, a humanoid motion observation acceleration, and a humanoid motion observation angular velocity;

[0146] In an embodiment of the present invention, the human motion observation parameters are extracted from the motion change data of the humanoid action of the smart fan user obtained through the motion change analysis, so as to obtain the human motion observation position of the user by using a high-precision sensor (such as a lidar or an infrared sensor), and obtain the user's motion speed, acceleration and angular velocity data by using a three-dimensional accelerometer and a gyroscope, and store these data as an original observation data set. The combined use of these sensors can provide comprehensive information on the user's motion state, thereby forming a motion change observation parameter set, including the observation position, observation speed, observation acceleration and observation angular velocity of the user's motion, and finally obtaining the user's human motion motion change observation parameter set.

[0147] Step S232: performing humanoid motion state observation estimation on the user humanoid action motion change observation parameter set based on the user humanoid motion state transfer equation to obtain the user humanoid motion state observation estimation;

[0148] In an embodiment of the present invention, the motion state of the user is observed and estimated by combining the user humanoid motion state transfer equation obtained in the previous design with the corresponding motion change observation parameters in the user humanoid action motion change observation parameter set, so as to use the previous motion state transfer equation, which describes how the user's state at a certain moment is transferred to the next moment, and use the position, velocity, acceleration and angular velocity data in the observation parameter set, which are input into the motion state transfer equation, and the user's motion state is calculated through a state prediction algorithm (such as a state prediction model). Specifically, a linear or nonlinear state estimation method is applied, combined with filtering techniques such as Kalman filtering, to iteratively estimate the user's motion state, and finally obtain an observation estimate of the user's humanoid motion state.

[0149] Step S233: performing Kalman filter gain update processing on the user humanoid motion state observation estimate based on the user humanoid action motion change observation parameter set to obtain the user humanoid motion state gain optimization estimate;

[0150] In an embodiment of the present invention, the Kalman filter gain is updated for the corresponding user humanoid motion state observation estimate by combining the user humanoid action motion change observation parameter set obtained by previous observation analysis, so as to construct an observation noise covariance matrix and introduce system noise data based on the motion state observation estimate obtained in the previous step, and use the Kalman filter to filter the corresponding noise data to calculate the Kalman gain, determine the influence weight of the new observation value on the current state estimate, and in the specific calculation process, use the Kalman gain formula to combine the observation value with the estimated value, update the motion state estimate, so as to reduce the prediction error, and finally obtain the optimized estimate of the user humanoid motion state gain.

[0151] Step S234: based on the user humanoid motion state gain optimization estimation, the humanoid motion trajectory of the smart fan user is tracked and reconstructed to obtain the humanoid motion trajectory of the smart fan user.

[0152] In an embodiment of the present invention, the humanoid motion trajectory of the corresponding smart fan user action motion change is tracked and reconstructed by combining the user humanoid motion state gain optimization estimate obtained after the previous gain update. Starting from the gain optimization estimate obtained in the previous step, a trajectory reconstruction algorithm (such as the least squares method or the spline interpolation method) is used to continuously track and model the user's motion state. In specific operations, the user's historical state data and the optimized estimate are used to construct a motion trajectory model, and path interpolation is performed to form the user's complete motion trajectory. This process needs to consider the continuity and real-time nature of the motion, so the data needs to be updated in real time to ensure that the model reflects the current motion state, and finally the humanoid motion trajectory of the smart fan user is reconstructed.

[0153] Further, step S233 includes the following steps:

[0154] Performing Kalman filtering on the user humanoid motion change observation parameter set to obtain the user humanoid motion observation noise data and the user humanoid motion observation noise filtering parameter set;

[0155] In an embodiment of the present invention, a user humanoid motion change observation parameter set obtained previously through observation is input into a Kalman filter. The initialization phase of the Kalman filter needs to define an initial state vector and an initial covariance matrix. The prediction phase of the Kalman filter predicts the state according to the motion model. Then, the observation parameters are applied to the prediction results through the observation update phase. Through this method, noise data in the observation parameter set can be filtered out, and an observation noise-filtered parameter set after filtering is generated. In this process, the NumPy library in the Python programming language is used to perform matrix operations to improve computing efficiency, and the SciPy library is used to implement observation parameter processing and filtering, and finally the user humanoid motion observation noise data and the user humanoid motion observation noise-filtered parameter set are obtained.

[0156] Preferably, based on the user humanoid motion observation noise data, the observation noise covariance matrix of the user humanoid motion change observation parameter set is calculated to obtain the user humanoid motion observation noise covariance matrix;

[0157] In an embodiment of the present invention, the covariance matrix of the observation noise is calculated by combining the previously acquired user humanoid motion observation noise data for analysis. Specifically, the noise components in the user motion observation data are first extracted, and the distribution characteristics of these components are analyzed. Based on the collected user humanoid motion change observation parameter set, a statistical method (such as least squares method or maximum likelihood estimation) is used to estimate the variance and covariance of the observation noise, so as to quantitatively calculate the corresponding observation noise covariance matrix to provide a more accurate description of the noise characteristics. This process ensures that the influence of noise is effectively quantified in the motion state estimation, and finally the user humanoid motion observation noise covariance matrix is ​​obtained.

[0158] Preferably, the user humanoid motion observation noise Kalman gain is obtained by introducing system noise data and performing Kalman gain calculation on the user humanoid motion observation noise filtering parameter set based on the user humanoid motion observation noise covariance matrix;

[0159] In the embodiment of the present invention, by introducing the system noise data and combining the user humanoid motion observation noise covariance matrix obtained by the previous quantitative calculation, a suitable calculation formula is constructed to perform quantitative calculation of the Kalman gain for the corresponding user humanoid motion observation noise filter parameter set, so as to determine the characteristics of the system noise, which is usually obtained through system testing and historical data analysis. Then, combined with the user humanoid motion observation noise covariance matrix, the Kalman gain formula is used for calculation. The formula is usually K=PH T (HPH T +R) ―1, where K is the Kalman gain, P is the covariance matrix of the user humanoid motion observation noise filtering parameter set, H is the system noise data matrix, and R is the user humanoid motion observation noise covariance matrix. The Kalman gain value obtained by this formula can reflect the contribution of the measurement value to the state estimation, and finally the user humanoid motion observation noise Kalman gain is obtained.

[0160] Preferably, a state gain update process is performed on the user humanoid motion state observation estimate based on the user humanoid action motion observation noise Kalman gain to obtain the user humanoid motion state gain optimization estimate.

[0161] In the embodiment of the present invention, the state gain of the user humanoid motion state observation estimate obtained by the previous observation estimation is updated by combining the user humanoid action motion observation noise Kalman gain obtained by the previous quantization calculation. This process mainly includes adjusting the current state estimate by the Kalman gain and calculating the new state estimate value. First, the current state estimate is combined with the observation value to calculate the updated state estimate. The specific formula is: in is the state estimate after the kth update, z k This is the current motion state observation estimate. Through this step, an optimized estimate of the user's humanoid motion state is achieved, making the state estimate more accurate and consistent with the actual motion situation, and finally obtaining an optimized estimate of the user's humanoid motion state gain.

[0162] Further, step S25 includes the following steps:

[0163] Step S251: calibrate and divide the dynamic range of the human-shaped motion of the user of the smart fan into time-series point ranges to obtain the corresponding relative range of the human-shaped motion of the user at each time-series point;

[0164] In an embodiment of the present invention, by setting multiple timing points (such as collecting the dynamic change range of humanoid motion once per second), the user's body motion trajectory is recorded in a specific environment. During the motion range calibration process, it is necessary to perform multiple tests under different environmental conditions (such as light changes, the presence of obstacles, etc.). By analyzing the collected data, a user humanoid motion model is established, and different motion ranges are divided, such as forward, backward, left and right movement, etc. At each timing point, the corresponding motion range is recorded and stored, and finally the corresponding relative range of user humanoid motion at each timing point is obtained.

[0165] Step S252: Obtain the intelligent sensing range of the intelligent fan, and estimate the point situation sensing intersection area for the corresponding relative range of the user humanoid movement at each time point based on the intelligent sensing range of the intelligent fan, so as to obtain the situation sensing intersection area between the user humanoid movement range and the intelligent fan sensing range at each time point;

[0166] In an embodiment of the present invention, the intelligent perception range of the smart fan is determined, which can be achieved by using sensors (such as infrared sensors, ultrasonic sensors, etc.) installed on the smart fan. The sensors monitor the surrounding environment in real time and draw a two-dimensional or three-dimensional perception space range of a perception range. Then, the relative range of the user's humanoid motion recorded at each time point is analyzed, and the intersection area estimation method is used to calculate the size of the overlapping area between the user's motion range and the smart fan's perception range. In this step, a geometric algorithm (such as a rectangular intersection algorithm or a polygon intersection algorithm) is used for precise calculation to obtain the situational awareness intersection area at each time point, and finally the situational awareness intersection area between the user's humanoid motion range and the smart fan's perception range at each time point is obtained.

[0167] Step S253: The time-series area change trend of the situation awareness intersection area between the user's human motion range and the smart fan perception range at each time-series point is judged. If the change trend of the situation awareness intersection area size at each time-series point is gradually increasing, the situation awareness state between the user and the smart fan is judged to be the smart fan user approaching perception state; if the change trend of the situation awareness intersection area size at each time-series point is gradually decreasing, the situation awareness state between the user and the smart fan is judged to be the smart fan user far away perception state.

[0168] In an embodiment of the present invention, by comparing and judging the time series area change trend for the situation awareness cross-area size at each time series point calculated in the previous step, a statistical analysis of the change trend of the cross-area size is performed through a time series analysis method to analyze the change trend of the cross-area size at each time series point. When the change trend is gradually increasing, the situation awareness state between the user and the smart fan is determined to be "close to the perception state"; and when the change trend is gradually decreasing, it is determined to be "far away from the perception state". A certain threshold can also be set. When the amplitude of the cross-area change exceeds the threshold, the corresponding state change prompt is triggered. In addition, through real-time monitoring of the cross-area change trend, user behavior can be further analyzed. For example, when the user approaches, the smart fan automatically adjusts the working mode to improve the user's comfort.

[0169] Furthermore, the present invention also provides a situation awareness system based on an intelligent fan, which is used to execute the situation awareness method based on an intelligent fan as described above. The situation awareness system based on an intelligent fan includes:

[0170] A user humanoid frame difference recognition module is used to obtain the humanoid position change data of the smart fan user and the humanoid motion posture change data of the smart fan user, and perform humanoid motion frame change recognition analysis on the humanoid position change data of the smart fan user and the humanoid motion posture change data of the smart fan user, so as to obtain the humanoid motion change data of the smart fan user;

[0171] The user human figure situation perception judgment module is used to perform Kalman filtering human figure tracking on the human figure motion change data of the smart fan user to obtain the human figure motion trajectory of the smart fan user; perform human figure motion range analysis on the human figure motion trajectory of the smart fan user to obtain the dynamic change range of the human figure motion of the smart fan user; obtain the smart fan intelligent perception range, and perform situation perception cross judgment on the dynamic change range of the human figure motion of the smart fan user based on the smart fan intelligent perception range, so as to obtain the smart fan user approaching perception state and the smart fan user far away perception state;

[0172] The user proximity sensing fan acceleration control module is used to generate a wind speed acceleration control instruction using the smart fan response and act on the built-in control system of the smart fan to perform the corresponding smart fan wind speed automatic acceleration work when the user proximity sensing state is determined.

[0173] The user-away-from-sensing fan deceleration control module is used to generate a wind speed deceleration control instruction using the smart fan response when determining that the smart fan is in the user-away-from-sensing state, and act on the built-in control system of the smart fan to execute the corresponding smart fan wind speed automatic deceleration work.

[0174] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0175] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A situation awareness method based on intelligent fans, characterized in that: The following steps are involved: Step S1: obtaining the humanoid position change data and the humanoid motion posture change data of the smart fan user, and performing humanoid motion frame change recognition analysis on the humanoid position change data and the humanoid motion posture change data of the smart fan user to obtain the humanoid motion change data of the smart fan user; Step S2: performing Kalman filtering humanoid tracking on the humanoid motion change data of the smart fan user to obtain the humanoid motion trajectory of the smart fan user; performing humanoid motion range analysis on the humanoid motion trajectory of the smart fan user to obtain the dynamic change range of the humanoid motion of the smart fan user; obtaining the smart fan intelligent perception range, and performing situational perception cross-judgment on the dynamic change range of the humanoid motion of the smart fan user based on the smart fan intelligent perception range to obtain the smart fan user approaching perception state and the smart fan user far away perception state; Step S3: when it is determined that the smart fan is in a user approaching sensing state, the smart fan response is used to generate a wind speed acceleration control instruction and acts on the built-in control system of the smart fan to perform the corresponding smart fan wind speed automatic acceleration work; Step S4: When it is determined that the smart fan user is away from the sensing state, the smart fan response is used to generate a wind speed deceleration control instruction and act on the built-in control system of the smart fan to execute the corresponding smart fan wind speed automatic deceleration work.

2. The method for situation awareness based on intelligent fans according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Acquire the position change data of the human figure of the smart fan user and the movement posture change data of the human figure of the smart fan user; Step S12: performing time series synchronization processing on the smart fan user humanoid position change data and the smart fan user humanoid motion posture change data to obtain the user humanoid position change time series data and the user humanoid motion posture change time series data in the same time sequence dimension; Step S13: performing a user humanoid three-dimensional space conversion on the user humanoid position change time series data and the user humanoid motion posture change time series data in the same time sequence dimension to obtain a user humanoid action three-dimensional time series change space; Step S14: dividing the time frame position and posture distribution of the three-dimensional temporal change space of the user humanoid action to obtain the user humanoid position distribution sub-data and the user humanoid posture distribution sub-data corresponding to each time frame; Step S15: performing humanoid action frame-to-frame change recognition analysis on the user humanoid position distribution sub-data and the user humanoid posture distribution sub-data corresponding to each time frame to obtain the intelligent fan user humanoid action motion change data.

3. The method for situation awareness based on intelligent fans according to claim 2 is characterized in that: Step S13 includes the following steps: Step S131: extracting the position and posture change coordinate corner points of the user human figure position change time series data and the user human figure motion posture change time series data in the same time sequence dimension, and obtaining the user human figure position change coordinate corner point set and the user human figure posture change coordinate corner point set in the same time sequence dimension; Step S132: performing a three-dimensional space coordinate system geometric transformation on the user humanoid position change coordinate corner point set and the user humanoid posture change coordinate corner point set in the same time sequence dimension to obtain the user humanoid position change coordinate point set and the user humanoid posture change coordinate point set in the three-dimensional space coordinate system; Step S133: performing three-dimensional space mapping modeling of humanoid motion according to the coordinate point set of the user humanoid position change and the coordinate point set of the user humanoid posture change in the three-dimensional space coordinate system to obtain an initial three-dimensional space model of the user humanoid motion; Step S134: performing humanoid motion smoothing filtering processing on the initial three-dimensional spatial model of the user's humanoid motion to obtain a three-dimensional smooth spatial model of the user's humanoid motion; Step S135: Simulate the dynamic time series change of the three-dimensional smooth space model of the user's humanoid action to obtain the three-dimensional time series change space of the user's humanoid action.

4. The method for situation awareness based on intelligent fans according to claim 2 is characterized in that: Step S15 includes the following steps: Step S151: performing position distribution vector analysis on the user humanoid position distribution sub-data corresponding to each time frame to obtain the user humanoid position distribution space sub-vector corresponding to each time frame; Step S152: performing inter-frame distance variation statistics calculation on the user figure position distribution space sub-vectors corresponding to each time frame, and obtaining user figure position distance variation values ​​between each adjacent continuous time frame; Step S153: performing quantitative calculation of the posture angle change rate between consecutive frames for the user humanoid posture distribution sub-data corresponding to each time frame, so as to obtain the user humanoid posture angle change rate between each adjacent consecutive time frame; Step S154: performing inter-frame change recognition and analysis of humanoid motion according to the user humanoid position distance change value and the user humanoid posture angle change rate between each adjacent continuous time frame, and obtaining the intelligent fan user humanoid motion change data.

5. The method for situation awareness based on intelligent fans according to claim 4 is characterized in that: Step S153 includes the following steps: Performing posture Euler angle standardization processing on the user humanoid posture distribution sub-data corresponding to each time frame to obtain the user humanoid posture standard Euler angle corresponding to each time frame; The standard Euler angles of the user humanoid posture corresponding to each time frame are analyzed for the posture change amplitude between adjacent continuous frames to obtain the user humanoid posture Euler angle change amplitude between each adjacent continuous time frame; The rate of change of the posture angle between consecutive frames is quantitatively calculated according to the change amplitude of the Euler angle of the user humanoid posture between each adjacent continuous time frame, so as to obtain the rate of change of the user humanoid posture angle between each adjacent continuous time frame.

6. The method for situation awareness based on intelligent fans according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: performing humanoid motion state vector description analysis on the humanoid motion change data of the smart fan user to obtain a humanoid motion state description vector set of the smart fan user, wherein the humanoid motion state description vector set of the smart fan user includes a humanoid motion position state vector of the user, a humanoid motion speed state vector of the user, a humanoid motion acceleration state vector of the user, and a humanoid motion angular velocity state vector of the user; Step S22: designing a state transfer equation according to the humanoid motion state description vector set of the smart fan user to generate a user humanoid motion state transfer equation; Step S23: performing Kalman filtering humanoid tracking on the motion change data of the humanoid user of the smart fan based on the humanoid motion state transfer equation of the user to obtain the humanoid motion trajectory of the smart fan user; Step S24: performing a human motion range analysis on the human motion trajectory of the smart fan user to obtain a dynamic range of human motion change of the smart fan user; Step S25: Obtain the intelligent sensing range of the smart fan, and perform situational awareness cross-judgment on the dynamic change range of the human motion of the smart fan user based on the intelligent sensing range of the smart fan, to obtain the smart fan user approaching sensing state and the smart fan user far away sensing state.

7. The method for situation awareness based on intelligent fans according to claim 6, characterized in that: Step S23 includes the following steps: Step S231: extracting humanoid motion observation parameters from the humanoid motion change data of the smart fan user to obtain a humanoid motion change observation parameter set, wherein the humanoid motion change observation parameter set includes a humanoid motion observation position, a humanoid motion observation speed, a humanoid motion observation acceleration, and a humanoid motion observation angular velocity; Step S232: performing humanoid motion state observation estimation on the user humanoid action motion change observation parameter set based on the user humanoid motion state transfer equation to obtain the user humanoid motion state observation estimation; Step S233: performing Kalman filter gain update processing on the user humanoid motion state observation estimate based on the user humanoid action motion change observation parameter set to obtain the user humanoid motion state gain optimization estimate; Step S234: based on the user humanoid motion state gain optimization estimation, the humanoid motion trajectory of the smart fan user is tracked and reconstructed to obtain the humanoid motion trajectory of the smart fan user.

8. The method for situation awareness based on intelligent fans according to claim 7, characterized in that: Step S233 includes the following steps: Performing Kalman filtering on the user humanoid motion change observation parameter set to obtain the user humanoid motion observation noise data and the user humanoid motion observation noise filtering parameter set; Based on the user humanoid motion observation noise data, the observation noise covariance matrix of the user humanoid motion change observation parameter set is calculated to obtain the user humanoid motion observation noise covariance matrix; By introducing system noise data and calculating the Kalman gain of the user humanoid motion observation noise filter parameter set based on the user humanoid motion observation noise covariance matrix, the user humanoid motion observation noise Kalman gain is obtained; Based on the Kalman gain of the user's humanoid action motion observation noise, the user's humanoid motion state observation estimation is updated with a state gain to obtain an optimized estimate of the user's humanoid motion state gain.

9. The method for situation awareness based on intelligent fans according to claim 6, characterized in that: Step S25 includes the following steps: Step S251: calibrate and divide the dynamic range of the human-shaped motion of the user of the smart fan into time-series point ranges to obtain the relative range of the human-shaped motion of the user corresponding to each time-series point; Step S252: Obtain the intelligent sensing range of the intelligent fan, and estimate the point situation sensing intersection area for the corresponding relative range of the user humanoid movement at each time point based on the intelligent sensing range of the intelligent fan, so as to obtain the situation sensing intersection area between the user humanoid movement range and the intelligent fan sensing range at each time point; Step S253: The time-series area change trend of the situation awareness intersection area between the user's human motion range and the smart fan perception range at each time-series point is judged. If the change trend of the situation awareness intersection area size at each time-series point is gradually increasing, the situation awareness state between the user and the smart fan is judged to be the smart fan user approaching perception state; if the change trend of the situation awareness intersection area size at each time-series point is gradually decreasing, the situation awareness state between the user and the smart fan is judged to be the smart fan user far away perception state.

10. A situation awareness system based on intelligent fans, characterized in that: Used to execute the intelligent fan situation awareness method according to claim 1, the intelligent fan situation awareness system comprises: A user humanoid frame difference recognition module is used to obtain the humanoid position change data of the smart fan user and the humanoid motion posture change data of the smart fan user, and perform humanoid motion frame change recognition analysis on the humanoid position change data of the smart fan user and the humanoid motion posture change data of the smart fan user, so as to obtain the humanoid motion change data of the smart fan user; The user human figure situation perception judgment module is used to perform Kalman filtering human figure tracking on the human figure motion change data of the smart fan user to obtain the human figure motion trajectory of the smart fan user; perform human figure motion range analysis on the human figure motion trajectory of the smart fan user to obtain the dynamic change range of the human figure motion of the smart fan user; obtain the smart fan intelligent perception range, and perform situation perception cross judgment on the dynamic change range of the human figure motion of the smart fan user based on the smart fan intelligent perception range, so as to obtain the smart fan user approaching perception state and the smart fan user far away perception state; The user proximity sensing fan acceleration control module is used to generate a wind speed acceleration control instruction using the smart fan response and act on the built-in control system of the smart fan to perform the corresponding smart fan wind speed automatic acceleration work when the user proximity sensing state is determined. The user-away-from-sensing fan deceleration control module is used to generate a wind speed deceleration control instruction using the smart fan response when determining that the smart fan is in the user-away-from-sensing state, and act on the built-in control system of the smart fan to execute the corresponding smart fan wind speed automatic deceleration work.

Images