An intelligent air conditioner control method and device, an air conditioner, and a storage medium

Abstract

The application discloses an intelligent air conditioner control method and device, an air conditioner and a storage medium, and belongs to the technical field of air conditioning, and the method comprises the following steps: collecting UWB tag data of a user through UWB technology, and collecting point cloud data of the user through radar technology; performing space-time alignment processing on the UWB tag data and the point cloud data; performing adaptive Kalman filtering fusion processing on the UWB tag data and the point cloud data after the space-time alignment processing, to obtain fusion features; performing state judgment on the user by using the fusion features, to obtain a user state; performing short-time trajectory prediction on the user by using the fusion features in combination with the user state, to obtain a predicted trajectory; acquiring a current running state of an air conditioner, generating a state adjustment instruction according to the predicted trajectory, and performing running control on the air conditioner by using the state adjustment instruction. The method provided by the application can be applied to an energy-saving air conditioner, can improve the human body sensing precision of the air conditioner, and can enhance the intelligent control effect.

Description

Technical Field

[0001] This invention relates to the field of air conditioner technology, and in particular to an intelligent air conditioner control method, device, air conditioner, and storage medium. Background Technology

[0002] With the rapid iteration of IoT and AI technologies, smart home systems have broken through the traditional, rudimentary application model of simple remote control and are gradually entering a higher-level intelligent development stage characterized by "proactive sensing, intelligent judgment, and adaptive adjustment." As one of the home appliances with the highest energy consumption in modern households, optimizing the intelligent control strategy of air conditioners has key application value and practical significance for improving the comfort of the living environment, reducing household energy consumption, and creating a healthy living atmosphere.

[0003] Currently, while mainstream smart air conditioning control systems on the market have achieved basic intelligent control functions, they still have significant technical shortcomings in core areas such as human body perception, human behavior recognition, spatial positioning, and dynamic adaptive control, making it difficult to meet users' high-quality demands for refined, seamless, and personalized living environments.

[0004] Currently, most smart air conditioners use pyroelectric infrared (PIR) sensors for human detection. These sensors can only detect a wide range of human movement, with extremely low spatial positioning accuracy and significant limitations in detection performance. When users are in a static state such as sitting and reading, lying down and resting, or sleeping, PIR sensors cannot effectively distinguish between the human body and the environmental background, easily causing the air conditioning system to misjudge the absence of users and trigger a shutdown. This detection deficiency also leads to frequent start-stop cycles, causing significant fluctuations in indoor temperature, severely impacting living comfort, and exacerbating energy waste and equipment wear and tear.

[0005] To compensate for the detection limitations of PIR sensors, the industry has gradually introduced wireless positioning technologies such as Wi-Fi and Bluetooth to assist in the intelligent control of air conditioners. However, these single wireless positioning technologies are affected by factors such as indoor multipath effects, signal attenuation, and environmental obstructions, making it difficult to achieve the required positioning accuracy. This makes it impossible to support the refined control requirements of accurately pointing the air conditioner's air deflector towards the user's activity area. At the same time, these technologies are highly dependent on users carrying terminal devices such as mobile phones and smart bracelets. If users do not wear or carry such devices, the system will not be able to complete human body sensing and positioning, severely limiting the applicable scenarios and resulting in a poor intelligent experience.

[0006] Millimeter-wave radar technology, with its advantages of penetrating non-metallic obstructions, non-contact detection, and strong dynamic behavior recognition capabilities, can achieve diversified sensing functions such as human motion recognition, breathing, and heart rate monitoring, effectively making up for the detection shortcomings of traditional sensors and being gradually applied to the field of smart home appliance sensing. However, single millimeter-wave radar technology still has obvious technical defects. In complex home scenarios with long-distance detection and multiple people coexisting, it is easily interfered with by environmental clutter, spatial reflections, and other factors, making it impossible to output stable and accurate absolute spatial coordinate data. At the same time, it is difficult to accurately distinguish the location information and behavioral intentions of different users in multiple scenarios, which can easily lead to confusion in the air conditioning's airflow direction, speed, and temperature control logic, resulting in insufficient control stability and accuracy.

[0007] Ultra-wideband (UWB) technology boasts outstanding advantages such as centimeter-level ultra-high ranging accuracy, precise direction perception, and strong anti-interference capabilities, enabling stable high-precision absolute spatial positioning and effectively compensating for the coordinate drift defects of millimeter-wave radar. Currently, the application of UWB and millimeter-wave radar technologies in the field of smart air conditioning is in isolation. Existing fusion solutions in the industry only involve simple data overlay processing, failing to achieve deep integration and complementary advantages of multimodal sensing technologies. On the one hand, it fails to utilize UWB's high-precision absolute positioning data to correct the coordinate offset errors of millimeter-wave radar, thus failing to improve spatial positioning stability in complex scenarios; on the other hand, it fails to leverage the high-frequency dynamic detection data of millimeter-wave radar to compensate for the response delay problem of UWB technology in scenarios involving rapid human movement. Furthermore, existing fusion control solutions have poor scenario adaptability and lack privacy protection mechanisms designed for private home scenarios, making them difficult to adapt to diverse and demanding smart home usage scenarios.

[0008] In summary, current intelligent air conditioning sensing and control systems generally suffer from many problems such as low detection accuracy, poor scene adaptability, simple control logic, and insufficient intelligence, making it impossible to achieve accurate human body perception and proactive intelligent regulation. Summary of the Invention

[0009] This invention provides an intelligent air conditioning control method, device, air conditioner, and storage medium, aiming to improve the accuracy of human body perception in air conditioners and enhance the intelligent control effect.

[0010] In a first aspect, embodiments of the present invention provide an intelligent air conditioning control method, comprising: The user's UWB tag data is collected using UWB technology, and the user's point cloud data is collected using radar technology; wherein the user wears a UWB tag that can be collected. The UWB tag data and point cloud data are subjected to spatiotemporal alignment processing; Adaptive Kalman filtering is applied to the spatiotemporally aligned UWB tag data and point cloud data to obtain fused features. The user's state is determined by using the fusion features. By combining the user's status and using the fused features, a short-term trajectory prediction is performed on the user to obtain the predicted trajectory. The current operating status of the air conditioner is obtained, and a status adjustment command is generated based on the predicted trajectory. Then, the air conditioner is controlled by the status adjustment command.

[0011] In a second aspect, embodiments of the present invention provide an intelligent air conditioning control device, comprising: The data acquisition unit is used to acquire the user's UWB tag data using UWB technology and the user's point cloud data using radar technology; wherein the user wears a UWB tag that can be acquired. A spatiotemporal alignment unit is used to perform spatiotemporal alignment processing on the UWB tag data and point cloud data; The data fusion unit is used to perform adaptive Kalman filter fusion processing on the UWB tag data and point cloud data after spatiotemporal alignment processing to obtain fusion features; A status determination unit is used to determine the user's status using the fusion features to obtain the user's status. The trajectory prediction unit is used to combine the user's state and perform short-term trajectory prediction on the user through the fused features to obtain the predicted trajectory. The operation control unit is used to obtain the current operating status of the air conditioner, generate a status adjustment command based on the predicted trajectory, and then use the status adjustment command to control the operation of the air conditioner.

[0012] Thirdly, embodiments of the present invention provide an air conditioner, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the intelligent air conditioning control method as described in the first aspect.

[0013] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the intelligent air conditioning control method as described in the first aspect.

[0014] This invention provides an intelligent air conditioner control method, device, air conditioner, and storage medium. The method includes: collecting UWB tag data of a user using UWB technology and point cloud data of the user using radar technology; wherein the user wears a collectable UWB tag; performing spatiotemporal alignment processing on the UWB tag data and point cloud data; performing adaptive Kalman filter fusion processing on the spatiotemporally aligned UWB tag data and point cloud data to obtain fusion features; using the fusion features to determine the user's state; combining the user state with the fusion features to predict the user's short-term trajectory to obtain a predicted trajectory; acquiring the current operating state of the air conditioner, generating a state adjustment command based on the predicted trajectory, and then using the state adjustment command to control the operation of the air conditioner. This invention deeply integrates UWB positioning technology with radar sensing technology. It leverages the high-precision absolute positioning advantage of UWB to correct radar coordinate drift errors, while utilizing radar's high-frequency dynamic detection advantage to compensate for the response delay deficiency of UWB technology in scenarios involving rapid human movement. This allows for the complementary advantages of the two technologies, effectively improving the accuracy and stability of human perception and spatial positioning in complex home environments. It enables air conditioners to achieve refined and personalized proactive intelligent control based on user location, movement trajectory, and status, enhancing user comfort and reducing unnecessary energy consumption. Furthermore, this invention designs corresponding control logic for different special scenarios (i.e., different user states), making the air conditioner more adaptable to various scenarios. This effectively solves the problems of low detection accuracy, poor scenario adaptability, and insufficient intelligence in existing intelligent air conditioner sensing and control systems. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart illustrating an intelligent air conditioning control method provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the first sub-process of an intelligent air conditioning control method provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the second sub-process of an intelligent air conditioning control method provided in an embodiment of the present invention; Figure 4 A schematic diagram of the third sub-process of an intelligent air conditioning control method provided in an embodiment of the present invention; Figure 5This is a schematic diagram of the fourth sub-process of an intelligent air conditioning control method provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the fifth sub-process of an intelligent air conditioning control method provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the sixth sub-process of an intelligent air conditioning control method provided in an embodiment of the present invention; Figure 8 This is an overall flowchart of an intelligent air conditioning control method provided in an embodiment of the present invention; Figure 9 A system architecture diagram of an intelligent air conditioning control method provided in an embodiment of the present invention; Figure 10 A scenario example diagram illustrating an intelligent air conditioning control method provided in an embodiment of the present invention; Figure 11 A schematic block diagram of an intelligent air conditioning control device provided in an embodiment of the present invention; Figure 12 This is a first sub-schematic block diagram of an intelligent air conditioning control device provided in an embodiment of the present invention; Figure 13 This is a second schematic block diagram of an intelligent air conditioning control device provided in an embodiment of the present invention; Figure 14 This is a third schematic block diagram of an intelligent air conditioning control device provided in an embodiment of the present invention; Figure 15 This is a fourth schematic block diagram of an intelligent air conditioning control device provided in an embodiment of the present invention; Figure 16 This is a fifth schematic block diagram of an intelligent air conditioning control device provided in an embodiment of the present invention; Figure 17 This is a sixth schematic block diagram of an intelligent air conditioning control device provided in an embodiment of the present invention; Figure 18 This is a schematic block diagram of an air conditioner provided in an embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0019] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0020] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0021] Please see below. Figure 1 This invention provides an intelligent air conditioning control method, specifically including steps S101 to S106.

[0022] Step S101: Collect the user's UWB tag data using UWB technology and collect the user's point cloud data using radar technology; wherein the user wears a UWB tag that can be collected. Step S102: Perform spatiotemporal alignment processing on the UWB tag data and point cloud data; Step S103: Perform adaptive Kalman filter fusion processing on the UWB tag data and point cloud data after spatiotemporal alignment processing to obtain fused features; Step S104: Use the fusion features to determine the user's state and obtain the user's state; Step S105: Combine the user status and use the fused features to predict the user's short-term trajectory to obtain the predicted trajectory; Step S106: Obtain the current operating status of the air conditioner, generate a status adjustment command based on the predicted trajectory, and then use the status adjustment command to control the operation of the air conditioner.

[0023] In this embodiment, combined with Figure 9First, UWB tag data worn by the user is collected using UWB technology, while point cloud data of the user is collected using radar technology. Next, the two types of data are spatiotemporally aligned to ensure consistency in time and space dimensions. Then, an adaptive Kalman filter algorithm is used to fuse the aligned UWB tag data and point cloud data, extracting fusion features. These fusion features are used to determine the user's current state. Based on the determined user state and the fusion features, the user's future short-term trajectory is predicted, resulting in a predicted trajectory. Finally, the current operating status of the air conditioner is obtained, and a corresponding state adjustment command is generated based on the predicted trajectory, which is then used to control the operation of the air conditioner.

[0024] This embodiment deeply integrates UWB positioning technology with radar sensing technology. It leverages the high-precision absolute positioning advantage of UWB to correct radar coordinate drift errors, while utilizing radar's high-frequency dynamic detection advantage to compensate for the response delay deficiency of UWB technology in scenarios involving rapid human movement. This allows for the complementary advantages of the two technologies, effectively improving the accuracy and stability of human perception and spatial positioning in complex home environments. It enables the air conditioner to achieve refined and personalized proactive intelligent control based on user location, movement trajectory, and status, improving user comfort and reducing unnecessary energy consumption. Furthermore, this embodiment designs corresponding control logic for different special scenarios (i.e., different user states), making the air conditioner more adaptable to various scenarios. This effectively solves the problems of low detection accuracy, poor scenario adaptability, and insufficient intelligence in existing intelligent air conditioner sensing and control systems.

[0025] In practical applications, a complete sensing and control system can be built based on the intelligent air conditioning control method of this embodiment. Combined with... Figure 10 The system comprises a UWB positioning network, a millimeter-wave radar sensing module, a central processing unit, and an air conditioning actuator, featuring a simple hardware architecture adaptable to various home scenarios. The system achieves high-precision spatial positioning by establishing a whole-house UWB positioning network: at least four UWB anchor points are deployed at the four corners of the ceiling in main living spaces such as the living room and bedrooms, forming a three-dimensional positioning grid with no blind spots. Users can wear lightweight UWB wristbands or use the UWB function of smart terminals as mobile tags. The tags emit pulse signals at fixed intervals, which are received and analyzed by the anchor points to calculate the user's three-dimensional absolute coordinates (x, y, z) with an accuracy of ±10cm, providing data support for precise spatial positioning.

[0026] Specifically, the system can be equipped with a 77GHz millimeter-wave radar module to achieve dynamic perception and vital sign detection. This radar can adopt a 4-transmit, 4-receive antenna array structure, and can be integrated into the top of an air conditioner indoor unit or independently deployed in the indoor ceiling area, adapting to diverse installation scenarios. By periodically transmitting frequency-modulated continuous wave (FMCW), the radar can effectively receive human body reflected echoes within a range of 0.3m to 8m, accurately extracting core features such as target distance, movement speed, and micro-Doppler readings. It outputs real-time environmental point cloud data and human breathing and heartbeat waveform information, providing high-frequency dynamic data for human state recognition and trajectory prediction.

[0027] The system's core control unit employs a high-performance embedded central processing unit, internally equipped with a multi-source data fusion algorithm library and complete intelligent air conditioning control logic. It establishes data communication with a UWB base station and millimeter-wave radar module via serial ports, receiving dual-modal sensing data in real time and performing fusion calculations and logical judgments. Ultimately, it outputs precise control commands to each actuator of the air conditioning system. The supporting air conditioning actuators possess high-precision and high-response control performance: the air guide vane motor supports bidirectional fine-tuning angle adjustment, with fast response speed and high control accuracy; the compressor and electronic expansion valve support stepless speed regulation, enabling smooth dynamic adjustment of air conditioning power and airflow, ensuring the efficient and precise implementation of the entire intelligent control strategy.

[0028] In one embodiment, such as Figure 2 As shown, steps S201 to S202 are included before step S101.

[0029] Step S201: Establish a communication connection between the air conditioner and the UWB positioning network; wherein the UWB positioning network is pre-built to form a three-dimensional positioning grid covering the detection area where the user is located; and the air conditioner is equipped with a radar detection module; Step S202: Geometric calibration of the detection area based on the UWB positioning network and radar detection module.

[0030] Before collecting UWB tag positioning data and millimeter-wave radar point cloud data, this embodiment requires prior system deployment and coordinate system calibration to effectively ensure the consistency and detection accuracy of subsequent multi-source sensing data. In actual deployment, UWB anchor points can be rationally deployed according to the residential unit type and indoor activity area distribution to construct a three-dimensional positioning grid covering the entire area. Subsequently, hardware communication debugging is completed to establish a stable communication link between the air conditioner and the UWB positioning network. Simultaneously, the radar detection module is assembled and debugged to ensure that both types of sensing hardware can stably output effective detection data. Finally, based on a unified world coordinate system, the detection space of the UWB positioning network and the radar detection module is geometrically calibrated to unify the spatial coordinate reference of the dual-modal sensing data, eliminate coordinate deviation problems, and provide reliable data support for subsequent high-precision data fusion, personnel positioning, and intelligent control.

[0031] In practical applications, the system can automatically complete room geometry calibration upon initial power-on. Specifically, each UWB anchor point autonomously constructs a global spatial coordinate system through mutual ranging, while the millimeter-wave radar scans the indoor environment when no one is present, generating a static clutter map including furniture, walls, and other obstacles, thus completing environmental initialization calibration. Simultaneously, users can use the accompanying app to independently mark the specific locations of private areas such as bedrooms and bathrooms, as well as air conditioning vents, providing accurate area determination for subsequent privacy adaptive degradation strategies and direct airflow avoidance intelligent control functions.

[0032] Furthermore, after completing the geometric calibration, the system can proceed with the routine cyclic detection and data preprocessing process. First, real-time data acquisition is performed, with the system continuously running a cyclic detection mechanism to read dual-modal sensing data in parallel. Specifically, the UWB data stream reads the three-dimensional spatial coordinates (x, y, z) of all online moving tags in real time. u ,y u ,z u and instantaneous velocity vector V u Acquire high-precision absolute positioning and motion status information; real-time acquisition of environmental point cloud data from millimeter-wave radar data stream. cloud Furthermore, it analyzes and extracts the micro-motion features of the human body (D). radar , including respiratory rate f breath Heart rate amplitude A heart Amplitude of limb swing A move It captures core vital signs and motion parameters to comprehensively capture the static vital signs and dynamic activity status of the human body.

[0033] Based on this, spatiotemporal alignment of the data is completed. High-precision timestamps are added to both the UWB coordinate data stream and the radar point cloud data stream. By combining data interpolation algorithms and caching queue mechanisms, the temporal alignment of these two types of heterogeneous data can be achieved. This effectively eliminates transmission delays and timing deviations caused by differences in transmission rates and sampling frequencies between different sensors, ensuring the consistency of multi-source sensing data in both time and space. This provides a well-organized and reliable raw data foundation for subsequent high-precision data fusion, human condition assessment, and intelligent air conditioning control.

[0034] In one embodiment, such as Figure 3 As shown, step S103 includes steps S301 to S303.

[0035] Step S301: Obtain the signal strength corresponding to the UWB tag data and point cloud data respectively; Step S302: Dynamically set the fusion coefficient according to the signal strength; Step S303: Based on the fusion coefficient, perform adaptive Kalman filter fusion processing on the UWB tag data and point cloud data to obtain fused features containing the user's motion trajectory.

[0036] This embodiment utilizes an adaptive Kalman filter fusion algorithm to achieve dynamic adaptive adjustment of the dual-modal sensing weights, effectively addressing issues such as poor adaptability of fixed fusion weights, positioning drift, and trajectory jumps. The fusion coefficient α is dynamically adjusted based on real-time signal strength, using the fusion calculation formula P. final =α·P uwb +(1-α)·P radar_pred The coordinate fusion calculation is completed. When the UWB signal strength is high, the fusion coefficient α approaches 1, and the high-precision absolute coordinates of UWB can be used as the main observation value to correct the coordinate drift error caused by indoor multipath effect and environmental interference of millimeter-wave radar. When the UWB signal strength is weak, the fusion coefficient α approaches 0, and the radar predicted coordinates can be used as the core output data to ensure the continuity and stability of positioning in complex environments.

[0037] This embodiment also supports adaptive switching of detection modes, adapting to various conditions such as user wearing a tag, not wearing a tag, and UWB signal loss. When the user is not wearing a UWB tag or the UWB signal is abnormally lost, it can automatically switch to radar-dominated detection mode. If the radar detects a human breathing frequency f... breath A breathing rate greater than 10 breaths per minute is sufficient to determine if someone is present indoors, and the center of the nearest human point cloud on the radar is selected as the user's real-time location. If the detected breathing rate is less than 5 breaths per minute or there is no effective radar echo signal, the system no longer relies solely on real-time radar positioning data. Instead, it combines the last effective UWB positioning location with historical movement speed to perform trajectory smoothing extrapolation, predicting the user's current location. This effectively avoids abrupt changes in positioning trajectory, eliminates frequent changes in air conditioning deflectors and airflow direction, and ensures the stability and continuity of air conditioning airflow control.

[0038] In one embodiment, such as Figure 4 As shown, step S104 includes steps S401 to S403.

[0039] Step S401: Obtain the user's motion trajectory from the fused features; Step S402: Extract the user's movement features and vital signs based on the user's movement trajectory; wherein, the point cloud data packet contains the user's vital signs, including respiratory features and altitude features; Step S403: Combine the movement features and vital signs to determine the user's state and obtain the user state; wherein the user state includes a moving state and a stationary state.

[0040] This embodiment constructs a dual-modal behavior state determination mechanism, relying on fused positioning trajectory, motion parameters, and human physical characteristics, combined with preset thresholds and machine learning models to achieve accurate classification of user multiple states. It can accurately distinguish between various typical home states such as active movement, sitting and resting, sleeping, falling, and leaving the room. Figure 11 As shown. The specific determination logic is as follows: when the user's instantaneous movement speed V... u The speed is greater than 0.5 m / s and the limb swing amplitude A move When the value is relatively large, the user is determined to be in an active movement state; when the user's movement speed V... u When the speed is less than 0.1 m / s, the limb movement amplitude is small, and the respiratory rate is maintained in the normal range of 12 to 20 breaths / minute, the user is judged to be in a static and awake state such as sitting and reading or working; when the user's movement speed approaches 0, the static duration lasts for more than 5 minutes, the respiratory rate is kept stable at 12 to 16 breaths / minute, and the limb micro-movement amplitude is extremely low, the user is judged to be in a sleep state.

[0041] Simultaneously, this embodiment can also identify abnormal human states. When a sudden drop in the user's spatial height coordinate z (e.g., from a standing height of 1.7m to a low position of 0.5m) is detected, and the instantaneous acceleration exceeds 2g, it can be determined that the user is in an abnormal fall state, providing a basis for home safety early warning. Furthermore, if no valid UWB tag signal is detected for more than 30 seconds, and the radar does not output any valid human respiratory signs, it is determined that the user has left the room, and the air conditioner can automatically switch to energy-saving standby mode, minimizing device energy consumption while ensuring a good user experience.

[0042] In one embodiment, such as Figure 5 As shown, step S105 includes steps S501 to S503.

[0043] Step S501: Generate corresponding prediction parameters based on different user states; Step S502: Combining the user's motion trajectory and prediction parameters in the fusion features, output the position coordinate sequence within a preset time period through the trajectory prediction model; Step S503: Generate the predicted trajectory from the position coordinate sequence.

[0044] This embodiment introduces a trajectory prediction model. Based on dual-modal fusion positioning sequences, millimeter-wave radar velocity vectors, and user behavior parameters, it can achieve high-precision, short-cycle prediction of human motion trajectories, thereby supporting front-mounted, seamless intelligent air conditioning adjustment. The specific processing flow of the model is as follows: First, based on the previously determined user behavior states such as movement, sitting, and sleep, the corresponding prediction weights, smoothing coefficients, and time steps are adaptively matched to ensure prediction adaptability under different motion conditions. Then, using the time-aligned historical fusion positioning coordinate sequence as the basic feature, combined with the real-time human movement velocity vector and acceleration information output by the radar, the model is input for iterative calculation. Finally, the model outputs a continuous sequence of user position coordinates within the next second, thereby fitting a smooth short-term predicted trajectory P. next .

[0045] Based on the prediction results, proactive intelligent control to avoid direct airflow can be implemented. For example, if the predicted trajectory shows that the user is about to enter the area directly exposed to the air conditioner vent, the optimal deflection angle of the air guide can be calculated in advance, and the air guide can be driven to complete the attitude adjustment in advance. Through the proactive control logic of "predicting first and then adjusting", the air delivery path is avoided before the user reaches the area directly exposed to the airflow, achieving imperceptible intelligent adjustment of "avoiding the air before it arrives". This completely solves the problems of traditional air conditioners such as slow response, abrupt adjustment, and discomfort caused by direct airflow, further improving indoor air supply comfort and intelligent experience.

[0046] In one embodiment, such as Figure 6 As shown, step S106 includes steps S601 to S603.

[0047] Step S601: Determine the predicted area that the user will enter based on the predicted trajectory; wherein, the predicted area includes the direct blowing area of ​​the air conditioner and the non-air conditioner control area; Step S602: When it is determined that the user will enter the direct airflow area of ​​the air conditioner, a wind avoidance control command is generated, and the airflow of the air conditioner is controlled by the wind avoidance control command. Step S603: When it is determined that the user will enter the non-air conditioner control area, an energy-saving command is generated, and the air conditioner is controlled to switch to standby mode after the user enters the non-air conditioner control area through the energy-saving command.

[0048] This embodiment achieves zone-specific intelligent control based on short-term predicted trajectories, integrating multiple functions such as pre-emptive wind avoidance and direct airflow prevention, energy saving based on room departure prediction, and personalized air supply for multiple users in different zones, balancing human comfort with equipment energy efficiency. Regarding real-time wind avoidance control, it can determine the user's predicted trajectory P in real time. nextWhether the user is about to enter the direct airflow cone angle range of the air conditioner vent is determined. If it is determined that the user is about to enter the direct airflow area, the optimal deflection angle of the air guide vane can be accurately solved using the arctangent function formula θ=atan2(y-y0,x-x0)+Δ, where Δ is a preset safety offset, thereby effectively avoiding deviations in the air supply boundary. Subsequently, the air guide vane motor is driven to rotate precisely to the target angle, dynamically adjusting the air conditioner's air supply path to adaptively achieve a refined air supply effect of "wind following the person and wind avoiding blowing on the person," fundamentally eliminating the discomfort caused by direct cold air blowing.

[0049] In terms of energy saving based on user movement prediction, the system can predict user movement based on the user's direction and speed. When it detects that the user is moving quickly toward the door, it will reduce the operating frequency of the air conditioner compressor in advance to reduce the power consumption of the equipment. After confirming that the user has completely entered the non-air-conditioned control area and left the effective temperature control range, the air conditioner will automatically switch to a low-power standby mode. This effectively avoids the problems of delayed shutdown and idling energy consumption of traditional air conditioners, greatly improves the energy efficiency of air conditioner operation, and meets the low-carbon operation requirements of energy-saving air conditioners.

[0050] In addition, this embodiment can also set up personalized air supply control for multiple users in different zones. For example, when the UWB positioning network detects multiple valid user tags indoors, it can automatically divide the indoor space into virtual zones based on the three-dimensional coordinates of each user. Users in different zones can independently set personalized temperature control parameters through the terminal APP, including target temperature, air supply speed, and other preferred configurations. The system can independently adjust the air guide and sweep angles and air conditioning refrigerant flow of the corresponding zones according to the user distribution and setting requirements of each zone, realizing differentiated and personalized intelligent air and temperature control for the whole house, meeting the comfortable living needs of multiple users and multiple scenarios, and comprehensively improving the smart home interactive experience.

[0051] In one embodiment, the user state further includes a sleep state, a fall state, an animal disturbance state, and a privacy protection state; like Figure 7 As shown, step S106 further includes steps S701 to S704.

[0052] Step S701: When the user is in sleep mode, generate a mode switching command and control the air conditioner to switch from the current operating state to sleep mode. Step S702: When the user is in sleep mode, generate a fan speed control and alarm command, control the air conditioner to reduce the output fan speed, and push alarm information. Step S703: When the user status is in animal disturbance status, generate an operation status adjustment and maintenance command; Step S704: When the user is in privacy protection mode, generate instructions to reduce the collection level or stop collecting UWB tag data and point cloud data, and keep the current operating state of the air conditioner unchanged.

[0053] In addition to the aforementioned air conditioner operation control methods, this embodiment also establishes a comprehensive intelligent adaptation and safety monitoring system for various special home scenarios. This system enables sleep comfort control, fall detection and alarm, precise filtering of pet interference, and adaptive protection of privacy areas, thereby significantly improving the system's scenario adaptability, safety, and intelligence level. Combined with... Figure 8 and Figure 10 Specifically: In terms of intelligent sleep adaptation, the system can accurately determine that the user has entered a sleep state after monitoring the user's breathing rate to remain stably within the sleep threshold range for a certain period of time, and automatically trigger the air conditioner's dedicated sleep mode. The system simultaneously turns off the air conditioner's screen display and various prompts, lowers the fan speed to a silent setting, and fixes the angle of the air deflector to avoid the risk of direct airflow. At the same time, it adopts a time-sharing temperature control strategy, maintaining a suitable low temperature in the first half of the night to help users fall asleep, and gradually increasing the set temperature in the second half of the night according to the time gradient, effectively preventing users from catching a cold at night, taking into account both sleep comfort and the need for constant temperature health.

[0054] In terms of fall safety monitoring, the system relies on the high-frequency sensing capabilities of millimeter-wave radar to accurately identify abnormal falls. When the radar detects a sudden drop in vertical height exceeding 0.5m within a short period, a sudden surge in vertical velocity vector, accompanied by abnormal UWB coordinate shifts or a brief signal loss, the system determines that the user is at risk of falling. The system immediately responds with safety protection logic, shutting off the air conditioner's high-powered airflow to prevent the user from experiencing heatstroke and discomfort due to continuous strong airflow after a fall. Simultaneously, it pushes abnormal alarm information to family members' mobile devices or the community monitoring platform via Wi-Fi network, effectively adapting to home safety monitoring scenarios for elderly people and children living alone.

[0055] In terms of pet interference filtering, the system relies on dual-modal features of UWB and millimeter-wave radar to accurately identify non-human targets. On one hand, the UWB mobile tag in this embodiment is worn only by the user; pets do not have associated tag signals, which can serve as a basic basis for differentiation. On the other hand, pets possess unique characteristics such as a high breathing rate (more than 30 breaths per minute), small size, and low center of gravity, which are significantly different from human radar echo characteristics. The system incorporates a lightweight machine learning classifier based on a multilayer perceptron and is equipped with a dedicated pet behavior fingerprint database. It can compare radar echo characteristics in real time to accurately filter interference from pets such as cats and dogs, eliminating problems such as incorrect air conditioning venting adjustments and misjudgments of human presence caused by pet activity, thus improving the stability of equipment operation.

[0056] In terms of privacy security protection, the system supports an adaptive degradation strategy for privacy zones. When it is detected that a user enters a preset privacy zone such as a bedroom or a bathroom, the privacy protection mechanism is automatically activated to reduce the scanning resolution of the millimeter-wave radar, turn off the high-resolution point cloud imaging function, and only retain the basic human presence detection capability. On the premise of ensuring the normal intelligent regulation of the air conditioner, the collection range of environmental and human sensing data is minimized, so as to protect the user's home privacy from the source and eliminate the potential safety hazards of intelligent sensing in private scenarios.

[0057] In addition, this embodiment also has good functional scalability, which can realize multi-device linkage, personalized memory adaptation and contactless intelligent interaction, so as to further enhance the whole-house smart home experience. For example, in terms of multi-room linkage regulation, the UWB high-precision positioning trajectory can be relied on to predict the user's movement trend. When it is detected that the user is moving from a public area such as a living room to a private area such as a bedroom, the air conditioner in the target room can be precooled or preheated in advance, so as to achieve a seamless cross-room temperature transition and ensure a continuous comfortable experience of the whole-house environment.

[0058] In terms of personalized preference memory, the unique tag ID of UWB can be used to distinguish the identities of different family members and match the corresponding exclusive temperature control preferences. When an elderly user is identified, the healthy constant temperature mode is automatically activated to maintain a constant room temperature and reduce wind noise; when a child user is identified, the anti-direct-blow mode is automatically activated, adopting large-angle wide-angle soft air supply to adapt to the health and comfort needs of different groups of people.

[0059] In terms of contactless gesture interaction, relying on the ultra-high frequency dynamic capture capability of the millimeter-wave radar, the system can accurately identify simple gesture actions such as waving and making a fist within the user's preset area, and correspondingly realize operations such as air supply gear switching and temperature adjustment, without the need to contact a physical remote control or a touch panel, so as to achieve imperceptible and convenient human-computer intelligent interaction.

[0060] In general, the intelligent air conditioner control method proposed in this embodiment is based on the multi-modal sensing collaborative fusion architecture of UWB and millimeter-wave radar, which specifically solves the technical shortcomings of existing intelligent air conditioners such as low sensing control accuracy, poor adaptability, high energy consumption and poor experience, and effectively improves the accuracy, comfort, energy saving and use safety of air conditioner control. Most of the existing intelligent air conditioners adopt PIR pyroelectric infrared sensors, which can only detect human movement behaviors, cannot identify static human states such as sitting still and sleeping, and are prone to problems such as misjudging no person and stopping, frequent start and stop, and room temperature fluctuation. This embodiment utilizes the micro-Doppler effect of the millimeter-wave radar to accurately capture weak vital signs such as human respiration and heartbeat, stably identify human presence all day long, which can effectively avoid the detection blind zones and misjudgment problems of traditional sensors, ensure constant indoor temperature and comfort, and reduce the loss caused by frequent start and stop of the equipment.

[0061] Existing single Wi-Fi and Bluetooth wireless positioning technologies have limited accuracy and are susceptible to indoor signal interference, failing to meet the needs of precise directional air delivery. Furthermore, they rely on user-carried terminal devices, limiting their application scenarios. Meanwhile, single millimeter-wave radar, in complex scenarios with multiple people at long distances, suffers from unstable coordinate output and weak personnel differentiation capabilities, easily leading to chaotic air delivery control logic. To address this, this embodiment integrates the advantages of UWB high-precision absolute positioning and millimeter-wave radar high-frequency dynamic tracking, achieving bidirectional data correction. UWB corrects radar coordinate drift, while radar dynamic data compensates for UWB's rapid movement response delay, constructing real-time three-dimensional coordinates of indoor personnel. This precisely drives the air guide plate to achieve refined air delivery control, ensuring "air follows the person" and "air avoids the person," without relying on terminal devices, thus offering greater scenario adaptability.

[0062] In terms of energy-saving control, the control method provided in this embodiment is particularly suitable for energy-saving air conditioners, highly matching the core design requirements of low power consumption and precise temperature control, and can fully tap the energy-saving potential of the equipment. This embodiment can accurately identify the number, distribution, and movement trajectory of people indoors, and adaptively optimize the air conditioning air supply and temperature control strategies, avoiding redundant energy consumption caused by traditional fixed operation modes. At the same time, relying on UWB trajectory prediction capabilities, it can predict the trend of users leaving in advance, proactively reduce the air conditioning power and switch to a low-power standby mode, abandoning the traditional passive and lagging control logic, effectively reducing the ineffective high-load operation of the air conditioner, and achieving proactive and precise energy-saving management, which meets the application needs of low-carbon energy saving. In terms of privacy, it can only process radio wave data and not collect human images and facial information, avoiding the risk of privacy leakage from the root, and is suitable for use in private scenarios such as bedrooms.

[0063] Furthermore, this embodiment boasts exceptional scalability and practicality. It utilizes millimeter-wave radar to monitor human vital signs and posture in real time, accurately identifying abnormal conditions such as falls, abnormal breathing, and prolonged periods of stillness. This allows for rapid triggering of alerts and coordinated adjustments to the air conditioning system, effectively ensuring the safety of elderly people living alone and children. Simultaneously, it supports users in different areas to customize parameters such as temperature, humidity, and airflow modes, enabling personalized intelligent control throughout the house and significantly enhancing the smart home living experience and comfort.

[0064] The intelligent air conditioning control method provided in this embodiment has the following characteristics: (1) This embodiment proposes an adaptive weighted Kalman filter algorithm to achieve deep complementary fusion of UWB and millimeter-wave radar data. The algorithm uses the low-frequency, high-precision absolute coordinates output by UWB as the observation value to accurately correct the state prediction value of the millimeter-wave radar. At the same time, relying on the high-frequency human micro-motion characteristics of the millimeter-wave radar, it effectively fills the data blind spot of UWB technology in scenarios where users move quickly. Through bidirectional correction and fusion optimization of dual-modal data, this embodiment can significantly reduce user position tracking error and system response delay in complex home environments such as multiple people walking around and furniture blocking the view. It can also ensure that the air conditioner air guide can follow the user's position movement in real time and accurately, and achieve stable dynamic air supply control.

[0065] (2) Construct a dual-modal human presence determination model to completely solve the industry pain point of traditional air conditioners that "stop when a person is still". This embodiment supports adaptive switching of working modes. When the UWB signal is lost or there is no tag signal input, the system can automatically switch to the millimeter-wave radar-dominated detection mode and continuously determine the status of people in the room by extracting vital signs such as human breathing and heartbeat. At the same time, a dual-signal linkage dynamic threshold logic is set. Only when both UWB and millimeter-wave radar have no effective detection signal for a long time will the system determine that the user has left the room and execute the corresponding control strategy. This effectively identifies the user's still state such as sleeping or sitting and reading, ensuring the continuous and stable operation of the air conditioner, avoiding indoor temperature fluctuations caused by frequent start-stop, and greatly improving the comfort of living.

[0066] (3) Predictive intelligent air supply control based on high-precision fusion positioning data. In this embodiment, the precise three-dimensional coordinate data after fusion of UWB and radar is used to construct a dynamic indoor thermal distribution map in real time. Combined with the human motion velocity vector collected by millimeter-wave radar, the system accurately predicts the user's movement trajectory. When it is detected that the user is about to enter the area directly blown by the air conditioner vent, the system can calculate the optimal deflection angle in advance and actively drive the air guide plate to complete the angle adjustment, so as to achieve the predictive air supply effect of "avoiding the wind before it arrives", which solves the discomfort of direct blowing from the root and realizes refined and humanized air supply control.

[0067] (4) Establish a full-process privacy protection mechanism to balance perception accuracy and privacy scene security. This embodiment only processes pure sensor data such as distance, speed, phase, and point cloud throughout the process, without collecting or storing any human image information, thus avoiding the risk of image data leakage from the source. At the same time, a privacy area adaptive degradation strategy is proposed. The system can automatically identify preset private areas such as bedrooms and bathrooms. When the system detects that the user is in a privacy scene, it automatically adjusts the radar working state, reduces the scanning resolution, and turns off the high-resolution imaging function, retaining only the low-power basic human presence detection capability. This ensures normal intelligent control of the air conditioner while comprehensively protecting the user's home privacy.

[0068] (5) Construct a dedicated pet interference identification mechanism to effectively filter non-human target interference. This solution combines the missing features of UWB tags and the differences in millimeter-wave radar echoes, and relies on the unique characteristics of pets such as small size, low center of gravity, and fast breathing rate to train a dedicated machine learning classifier and establish a pet behavior fingerprint database. Through intelligent classification and identification, it can accurately distinguish between human and pet targets, automatically filter interference signals generated by pet activities, significantly reduce the probability of false triggering and misjudgment of air conditioning control, and improve the stability and accuracy of air conditioning operation in complex home scenarios.

[0069] Figure 11 This is a schematic block diagram of an intelligent air conditioning control device 1100 provided in an embodiment of the present invention. The device 1100 includes: The data acquisition unit 1101 is used to acquire the user's UWB tag data through UWB technology and the user's point cloud data through radar technology; wherein the user wears a UWB tag that can be acquired. Spatiotemporal alignment unit 1102 is used to perform spatiotemporal alignment processing on the UWB tag data and point cloud data; The data fusion unit 1103 is used to perform adaptive Kalman filter fusion processing on the UWB tag data and point cloud data after spatiotemporal alignment processing to obtain fusion features; The status determination unit 1104 is used to determine the user's status using the fusion features to obtain the user's status. The trajectory prediction unit 1105 is used to combine the user's state and perform short-term trajectory prediction on the user through the fused features to obtain the predicted trajectory. The operation control unit 1106 is used to obtain the current operating status of the air conditioner, generate a status adjustment command based on the predicted trajectory, and then use the status adjustment command to control the operation of the air conditioner.

[0070] In one embodiment, such as Figure 12 As shown, the data acquisition unit 1101 includes: The communication connection unit 1201 is used to establish a communication connection between the air conditioner and the UWB positioning network; wherein the UWB positioning network is pre-built to form a three-dimensional positioning grid covering the detection area where the user is located; and the air conditioner is equipped with a radar detection module. The geometric calibration unit 1202 is used to perform geometric calibration on the detection area based on the UWB positioning network and radar detection module.

[0071] In one embodiment, such as Figure 13 As shown, the data fusion unit 1103 includes: The intensity acquisition unit 1301 is used to acquire the signal intensity corresponding to the UWB tag data and the point cloud data respectively; The coefficient setting unit 1302 is used to dynamically set the fusion coefficient according to the signal strength; The fusion processing unit 1303 is used to perform adaptive Kalman filter fusion processing on the UWB tag data and point cloud data based on the fusion coefficient to obtain fusion features containing the user's motion trajectory.

[0072] In one embodiment, such as Figure 14 As shown, the state determination unit 1104 includes: The trajectory acquisition unit 1401 is used to acquire the user's motion trajectory from the fused features; The feature extraction unit 1402 is used to extract the user's movement features and vital signs features based on the user's movement trajectory; wherein, the point cloud data packet contains the user's vital signs features, which include respiratory features and altitude features; The status acquisition unit 1403 is used to combine the movement features and vital signs features to determine the user's status and obtain the user status; wherein, the user status includes a moving status and a stationary status.

[0073] In one embodiment, such as Figure 15 As shown, the trajectory prediction unit 1105 includes: The parameter generation unit 1501 is used to generate corresponding prediction parameters according to different user states; The sequence output unit 1502 is used to combine the user motion trajectory and prediction parameters in the fusion features and output a sequence of position coordinates within a future preset time period through the trajectory prediction model. The trajectory generation unit 1503 is used to generate the predicted trajectory from the position coordinate sequence.

[0074] In one embodiment, such as Figure 16 As shown, the operation control unit 1106 includes: The area prediction unit 1601 is used to determine the predicted area that the user will enter based on the predicted trajectory; wherein, the predicted area includes the direct blowing area of ​​the air conditioner and the non-air conditioner control area; The first instruction generation unit 1602 is used to generate a wind avoidance control instruction when it is determined that a user will enter the direct blowing area of ​​the air conditioner, and to control the air blowing of the air conditioner through the wind avoidance control instruction. The second instruction generation unit 1603 is used to generate an energy-saving instruction when it is determined that a user will enter the non-air conditioner control area, and to control the air conditioner to switch to standby mode after the user enters the non-air conditioner control area through the energy-saving instruction.

[0075] In one embodiment, the user state further includes a sleep state, a fall state, an animal disturbance state, and a privacy protection state; like Figure 17 As shown, the operation control unit 1106 further includes: The third instruction generation unit 1701 is used to generate a mode switching instruction when the user is in a sleep state, and control the air conditioner to switch from the current operating state to sleep mode. The fourth instruction generation unit 1702 is used to generate wind speed control and alarm instructions when the user is in sleep mode, and to control the air conditioner to reduce the output wind speed and push alarm information. The fifth instruction generation unit 1703 is used to generate an operation status adjustment and maintenance instruction when the user status is in an animal disturbance state. The sixth instruction generation unit 1704 is used to generate instructions to reduce the collection level or stop collecting UWB tag data and point cloud data when the user is in a privacy protection state, while keeping the current operating state of the air conditioner unchanged.

[0076] Since the embodiments of the apparatus and the embodiments of the method correspond to each other, please refer to the description of the embodiments of the method for the embodiments of the apparatus, which will not be repeated here.

[0077] Please see Figure 18 The present invention also provides an air conditioner 1800, which is a device with wireless and wired communication capabilities.

[0078] The air conditioner 1800 includes a processor 1802, a memory, and a network interface 1805 connected via a system bus 1801. The memory may include a non-volatile storage medium 1803 and internal memory 1804.

[0079] The non-volatile storage medium 1803 can store an operating system 18031 and a computer program 18032. When the computer program 18032 is executed, it causes the processor 1802 to execute an intelligent air conditioning control method.

[0080] The processor 1802 provides computing and control capabilities to support the operation of the entire air conditioner 1800.

[0081] The internal memory 1804 provides an environment for the operation of the computer program 18032 in the non-volatile storage medium 1803. When the computer program 18032 is executed by the processor 1802, the processor 1802 can execute an intelligent air conditioning control method.

[0082] This network interface 1805 is used for network communication with other devices. Those skilled in the art will understand that... Figure 18The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the air conditioner 1800 to which the present invention is applied. The specific air conditioner 1800 may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0083] The processor 1802 is used to run a computer program 18032 stored in a memory to implement any embodiment of the above-described intelligent air conditioning control method.

[0084] It should be understood that, in this embodiment of the invention, the processor 1802 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0085] It will be understood by those skilled in the art that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program may be stored in a storage medium, which is a computer-readable storage medium. The computer program is executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0086] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed, can perform the steps provided in the above embodiments. The storage medium may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0087] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

[0088] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A smart air conditioning control method, characterized in that, include: The user's UWB tag data is collected using UWB technology, and the user's point cloud data is collected using radar technology; wherein the user wears a UWB tag that can be collected. The UWB tag data and point cloud data are subjected to spatiotemporal alignment processing; Adaptive Kalman filtering is applied to the spatiotemporally aligned UWB tag data and point cloud data to obtain fused features. The user's state is determined by using the fusion features. By combining the user's status and using the fused features, a short-term trajectory prediction is performed on the user to obtain the predicted trajectory. The current operating status of the air conditioner is obtained, and a status adjustment command is generated based on the predicted trajectory. Then, the air conditioner is controlled by the status adjustment command.

2. The intelligent air conditioning control method according to claim 1, characterized in that, Prior to the steps of collecting the user's UWB tag data via UWB technology and collecting the user's point cloud data via radar technology, the following steps are included: A communication connection is established between the air conditioner and the UWB positioning network; wherein the UWB positioning network is pre-built to form a three-dimensional positioning grid that fully covers the detection area where the user is located; and the air conditioner is equipped with a radar detection module; The detection area is geometrically calibrated based on the UWB positioning network and radar detection module.

3. The intelligent air conditioning control method according to claim 1, characterized in that, The adaptive Kalman filter fusion process is performed on the UWB tag data and point cloud data after spatiotemporal alignment to obtain fused features, including: Obtain the signal strength corresponding to each of the UWB tag data and point cloud data; The fusion coefficient is dynamically set based on the signal strength. Based on the fusion coefficients, the UWB tag data and point cloud data are subjected to adaptive Kalman filter fusion processing to obtain fused features containing the user's motion trajectory.

4. The intelligent air conditioning control method according to claim 3, characterized in that, The step of using the fused features to determine the user's state and obtain the user's state includes: Obtain the user's motion trajectory from the fused features; The user's movement features and vital signs are extracted based on the user's movement trajectory; wherein, the point cloud data packet contains the user's vital signs, including respiratory features and altitude features; The user's state is determined by combining the movement characteristics and vital signs; wherein, the user state includes a moving state and a stationary state.

5. The intelligent air conditioning control method according to claim 3, characterized in that, The step of combining user status and using the fused features to predict the user's short-term trajectory to obtain the predicted trajectory includes: Generate corresponding prediction parameters based on different user states; Combining the user's motion trajectory and prediction parameters in the fusion features, the trajectory prediction model outputs a sequence of location coordinates within a preset time period in the future; The position coordinate sequence is used to generate the predicted trajectory.

6. The intelligent air conditioning control method according to claim 1, characterized in that, The step of obtaining the current operating status of the air conditioner, generating a status adjustment command based on the predicted trajectory, and then using the status adjustment command to control the operation of the air conditioner includes: The predicted trajectory determines the predicted area the user will enter; wherein, the predicted area includes the area directly blown by the air conditioner and the area not controlled by the air conditioner. When it is determined that a user will enter the direct airflow area of ​​the air conditioner, a wind avoidance control command is generated, and the air conditioner is controlled to blow air through the wind avoidance control command; When it is determined that a user will enter the non-air conditioner control area, an energy-saving command is generated, and the air conditioner is controlled to switch to standby mode after the user enters the non-air conditioner control area through the energy-saving command.

7. The intelligent air conditioning control method according to claim 4, characterized in that, The user status also includes sleep status, fall status, animal disturbance status, and privacy protection status; The step of obtaining the current operating status of the air conditioner, generating a status adjustment command based on the predicted trajectory, and then using the status adjustment command to control the operation of the air conditioner further includes: When the user is in sleep mode, a mode switching command is generated, and the air conditioner is controlled to switch from the current operating state to sleep mode. When the user is in sleep mode, a fan speed control and alarm command is generated, and the air conditioner is controlled to reduce the output fan speed and an alarm message is pushed. When the user is in an animal disturbance state, a running status adjustment and maintenance command is generated. When the user is in privacy protection mode, an instruction is generated to reduce the collection level or stop collecting UWB tag data and point cloud data, while keeping the current operating state of the air conditioner unchanged.

8. An intelligent air conditioning control device, characterized in that, include: The data acquisition unit is used to acquire the user's UWB tag data using UWB technology and the user's point cloud data using radar technology; wherein the user wears a UWB tag that can be acquired. A spatiotemporal alignment unit is used to perform spatiotemporal alignment processing on the UWB tag data and point cloud data; The data fusion unit is used to perform adaptive Kalman filter fusion processing on the UWB tag data and point cloud data after spatiotemporal alignment processing to obtain fusion features; A status determination unit is used to determine the user's status using the fusion features to obtain the user's status. The trajectory prediction unit is used to combine the user's state and perform short-term trajectory prediction on the user through the fused features to obtain the predicted trajectory. The operation control unit is used to obtain the current operating status of the air conditioner, generate a status adjustment command based on the predicted trajectory, and then use the status adjustment command to control the operation of the air conditioner.

9. An air conditioner, characterized in that, The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the intelligent air conditioning control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the intelligent air conditioning control method as described in any one of claims 1 to 7.

Images