Air conditioning equipment and control methods for air conditioning equipment
By using radar sensors to identify user identity and detect sleep status, and dynamically acquiring silent control parameters, the problem of insufficient adaptability of silent control in air conditioning equipment is solved, achieving more efficient and intelligent silent control.
Patent Information
- Application Number
- CN202610605042.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-30
- Publication Date
- 2026-11-17
AI Technical Summary
The silent control function of air conditioning equipment relies on a single trigger condition, resulting in insufficient adaptability and difficulty in adapting to the dynamic changes of different usage scenarios.
The system uses radar sensors to collect user vital signs data, identify user categories and detect sleep states, dynamically acquire mute control parameters, and control the mute state of the sound module through multi-layered progressive judgment logic.
It improves the adaptability and intelligence of air conditioning equipment's silent control, and can flexibly configure silent periods according to the user's identity category and sleep state to meet the silent needs in diverse usage scenarios.
Smart Images

Figure CN123110745A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of air conditioning equipment technology, and in particular to an air conditioning device and a control method for the air conditioning device. Background Technology
[0002] During operation, air conditioning units typically emit beeping sounds in response to user input to confirm that the command has been received. To create a quiet indoor environment, many air conditioning units now offer silent operation control functions.
[0003] However, most of these silent control methods rely on single-dimensional triggering conditions, such as time or ambient light. Their triggering timing is relatively fixed and it is difficult to adapt to the dynamic changes of different usage scenarios. As a result, the silent control of air conditioning equipment has insufficient adaptability. Summary of the Invention
[0004] This application provides an air conditioning device and a control method for the air conditioning device to solve the problem of insufficient adaptability of current air conditioning device noise control.
[0005] In a first aspect, some embodiments provide an air conditioning device, including:
[0006] The radar sensor is configured to collect vital sign data from at least one user.
[0007] The controller, connected to the radar sensor, is configured to:
[0008] When the silent control function of the air conditioning equipment is enabled, the user's identity category is identified based on the received vital sign data;
[0009] If at least one user's identity category belongs to the target category, sleep status detection is performed on the target user belonging to the target category based on the vital sign data to obtain the sleep status detection result of the target user; wherein, the target category refers to the identity category with mute control requirements;
[0010] If, based on the sleep state detection results, it is determined that there is at least one target user in a sleep state, the mute control parameters corresponding to the mute control function are obtained; the mute control parameters include a preset mute period and a sleep mode state;
[0011] When the current time is within the preset silent period, or when the sleep mode is enabled, the sound module of the air conditioner is controlled to be in a silent state.
[0012] Technical Effects: This solution collects vital sign data from at least one user using a radar sensor. With the air conditioner's silent control function enabled, it identifies whether the user belongs to a target category requiring silent control based on this data. If at least one user is identified as belonging to the target category, the sleep state of these users is further detected based on the vital sign data. If at least one user is confirmed to be asleep, silent control parameters, including a preset silent period and sleep mode status, are acquired. The sound module is only silenced during the preset silent period or when the sleep mode is active. Thus, through a multi-layered, progressive judgment logic, the silent control of the air conditioner no longer relies on a single, fixed trigger condition. Instead, it dynamically decides based on the user's identity category, actual sleep state, and flexibly configurable silent period or sleep mode status. This better meets users' silent needs in diverse usage scenarios, significantly improving the adaptability and intelligence of the air conditioner's silent control.
[0013] Secondly, some embodiments also provide a control method for an air conditioning device, applied to the air conditioning device provided in the first aspect, the control method for the air conditioning device including:
[0014] With the silent control function of the air conditioning unit enabled, the identity category of the user is identified based on the vital sign data of at least one user collected by the radar sensor.
[0015] If at least one user's identity category belongs to the target category, sleep status detection is performed on the target user belonging to the target category based on the vital sign data to obtain the sleep status detection result of the target user; wherein, the target category refers to the identity category with mute control requirements;
[0016] If, based on the sleep state detection results, it is determined that there is at least one target user in a sleep state, the mute control parameters corresponding to the mute control function are obtained; the mute control parameters include a preset mute period and a sleep mode state;
[0017] When the current time is within the preset silent period, or when the sleep mode is enabled, the sound-emitting module in the air conditioning device is controlled to be in a silent state.
[0018] Technical Effects: This solution collects vital sign data from at least one user using a radar sensor. With the air conditioner's silent control function enabled, it identifies whether the user belongs to a target category requiring silent control based on this data. If at least one user is identified as belonging to the target category, the sleep state of these users is further detected based on the vital sign data. If at least one user is confirmed to be asleep, silent control parameters, including a preset silent period and sleep mode status, are acquired. The sound module is only silenced during the preset silent period or when the sleep mode is active. Thus, through a multi-layered, progressive judgment logic, the silent control of the air conditioner no longer relies on a single, fixed trigger condition. Instead, it dynamically decides based on the user's identity category, actual sleep state, and flexibly configurable silent period or sleep mode status. This better meets users' silent needs in diverse usage scenarios, significantly improving the adaptability and intelligence of the air conditioner's silent control. Attached Figure Description
[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0020] Figure 1 This application provides structural schematic diagrams of air conditioning equipment according to some embodiments;
[0021] Figure 2 A schematic diagram of a silent mode setting interface in a mobile terminal application provided in some embodiments of this application;
[0022] Figure 3 This application provides schematic diagrams for the mute control of a sound-generating module in some embodiments.
[0023] Figure 4 This application provides schematic diagrams illustrating device silent operation linkage in some embodiments.
[0024] Figure 5 System architecture diagrams provided for some embodiments of this application;
[0025] Figure 6 This is a schematic diagram of the noise control process provided in some embodiments of this application;
[0026] Figure 7 This application provides interactive diagrams between various functional modules in some embodiments.
[0027] Figure 8 This application provides schematic diagrams illustrating scenarios where radar sensors scan users, based on some embodiments of the present application.
[0028] Figure 9 This is a schematic flowchart of a control method for an air conditioning device provided in some embodiments of this application;
[0029] Figure 10 This is a schematic diagram of the control device for an air conditioning equipment provided in some embodiments of this application. Detailed Implementation
[0030] The embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described below do not represent all embodiments consistent with this application. They are merely examples of systems and methods consistent with some aspects of this application as detailed in the claims.
[0031] It should be noted that the brief descriptions of terms in this application are only for the convenience of understanding the embodiments described below, and are not intended to limit the embodiments of this application. Unless otherwise stated, these terms should be understood in their ordinary and common meaning.
[0032] The terms "first," "second," "third," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar or related objects or entities, and do not necessarily imply a specific order or sequence, unless otherwise specified. It should be understood that such terms are interchangeable where appropriate.
[0033] The terms “comprising” and “having”, and any variations thereof, are intended to cover but not exclude inclusion, for example, a product or device that includes a range of components is not necessarily limited to all of the components that are clearly listed, but may include other components that are not clearly listed or that are inherent to such product or device.
[0034] The term "module" refers to any known or subsequently developed hardware, software, firmware, artificial intelligence, fuzzy logic, or combination of hardware and / or software code that is capable of performing the functions associated with that element.
[0035] In practical applications of air conditioning equipment, when users operate the air conditioner (e.g., adjusting temperature, fan speed, mode, etc. via remote control, control panel, or voice commands), the air conditioner usually emits a prompt sound, such as a "beep" sound from a built-in buzzer, or a voice module announcing the operation result to provide feedback to the user. In most daily scenarios, this prompt sound helps users confirm that the command has been received. However, in quiet environments such as during sleep at night, sudden beeping or voice announcements can become a source of disturbance, affecting the user's sleep quality, especially for children, the elderly, and other people who are sensitive to sound or have light sleep, where the disturbance is more pronounced.
[0036] Currently, many air conditioning devices offer silent control functions, but most rely on single-dimensional triggering conditions, such as time or ambient light. Time-based silent control requires a user-defined time period to take effect; if a user's schedule changes temporarily, the fixed time is difficult to match with actual sleep needs. Ambient light-based silent control depends on changes in indoor brightness, but light intensity is not necessarily related to whether a user is actually asleep. These single-dimensional triggering conditions result in relatively fixed or one-sided triggering times for silent control, making it difficult to adapt to dynamic changes in different usage scenarios. This leads to insufficient adaptability in the silent control of air conditioning devices.
[0037] To address the aforementioned technical problems, embodiments of this application provide an air conditioning device, such as... Figure 1 As shown, the air conditioning device 10 includes a radar sensor 102 and a controller 104, which are communicatively connected.
[0038] The radar sensor 102 refers to a sensor capable of emitting and receiving electromagnetic waves, used for non-contact detection of relevant characteristics of targets. The installation location and number of radar sensors 102 can be flexibly configured according to the actual scenario. Besides being integrated inside the indoor unit of the air conditioning unit 10, they can also be placed independently of the air conditioning unit 10, for example, as an independent module placed on a bedside table, wall, or ceiling, communicating with the controller 104 via wired or wireless means. In large-area or multi-room scenarios, multiple radar sensors 102 can be deployed to work collaboratively, improving detection coverage and accuracy through data fusion. In this embodiment, the radar sensor 102 is specifically the existing SLP (Synchronous Link Positioning) millimeter-wave radar in the air conditioning unit 10. This SLP radar has the ability to collect vital sign data such as user height, body proportions, respiratory rate, and limb movement amplitude. By directly reusing the existing SLP radar in the air conditioning unit, there is no need to add additional independent radar sensor hardware, thereby effectively controlling hardware costs and system complexity while achieving refined vital sign detection.
[0039] In this embodiment, the radar sensor 102 is configured to collect vital sign data of at least one user.
[0040] Vital signs data refer to data that reflect a user's physiological state and morphological characteristics, including but not limited to at least two of the following: height information, body proportions (such as the ratio of shoulder width to height), respiratory rate, limb movement amplitude, and body posture angle. This vital signs data can be used to identify the user and determine whether the user is asleep. Limb movement amplitude refers to the magnitude of the user's limb position deviation relative to a reference position at a given moment, such as raising an arm by 5 centimeters. Optionally, in subsequent user identification and sleep status detection, the variance of limb movement amplitude can be used, which is the second-order central distance obtained by statistically calculating amplitude data collected multiple times over a period of time, reflecting the dispersion or drastic fluctuation of these amplitude values. Body posture angle refers to the angle of inclination of the user's torso axis relative to the horizontal plane, measured in degrees (°). 0° represents lying flat, 90° represents standing, and values in between represent sitting or lying on one's side. Body posture angle can help determine whether a user is asleep.
[0041] Optionally, if the user wears a smart bracelet, the vital signs data may also include data such as the user's heart rate and skin conductance collected by smart means, and the smart bracelet will report the collected vital signs data to the controller 104.
[0042] For example, radar sensor 102 transmits electromagnetic wave signals of a specific frequency and receives the echo reflected by the human body. Utilizing the frequency difference, phase difference, and time delay between the transmitted and received signals, and through principles such as the Doleptom effect and phase change, it can non-contactly measure the distance, angle, and micro-motion information of the target, i.e., the user. This allows it to extract at least one vital sign data, including height, body proportions, respiratory rate, limb movement amplitude, and body posture angle. For multiple users in a room, the radar sensor can acquire the vital sign data of each user separately and assign a unique identifier to each user, thereby distinguishing the vital sign data of different users. Radar sensor 102 reports the collected vital sign data to controller 104.
[0043] Optionally, the radar sensor 102 can continue to operate after the air conditioner's silent control function is turned on, or it can automatically start when it detects someone moving in the room, in order to reduce power consumption.
[0044] The controller 104 is the core control unit of the air conditioning unit 10, and is typically composed of a microprocessor, memory, and related peripheral circuits. The controller 104 can receive user vital sign data reported by the radar sensor 102, process, judge, and make decisions on this data according to a preset logic algorithm, and finally send a mute command to the sound-emitting module of the air conditioning unit.
[0045] In this embodiment, the controller 104 is configured to: when the silent control function of the air conditioning device 10 is enabled, identify the user's identity category based on the received vital sign data; if at least one user's identity category belongs to the target category, perform sleep state detection on the target user belonging to the target category based on the vital sign data to obtain the sleep state detection result of the target user; wherein, the target category refers to the identity category with silent control requirements; if it is determined based on the sleep state detection result that there is at least one target user in a sleeping state, obtain the silent control parameters corresponding to the silent control function; the silent control parameters include a preset silent period and a sleep mode state; if the current time is within the preset silent period, or the sleep mode state is enabled, control the sound-emitting module of the air conditioning device 10 to be in a silent state.
[0046] The silent control function refers to the automatic silent capability of the air conditioning unit 10. This function can be activated or disabled by the operator via a separate "automatic silent switch." In this embodiment, the operator refers to someone who configures the air conditioning system via a mobile application (such as activating the "automatic silent switch," setting silent periods, or activating sleep mode), typically an adult or parent. The user refers to a person detected by the radar sensor 102 who is within the air conditioning's operating space; this could be a child, the elderly, etc. The operator and the detected user can be different categories, such as a parent and a child, or they can be the same category or the same person, such as an adult.
[0047] Figure 2 The image shows the silent mode settings interface in a mobile application, which includes an "automatic silent switch". When the operator clicks the "on" button on the interface, the silent control function is activated, and the air conditioner can control the silence according to subsequent conditions; when the operator clicks the "off" button (not shown in the figure), the silent control function is deactivated, and the air conditioner will always keep making normal noise and will not automatically mute.
[0048] Identity category refers to the group type to which a user belongs, such as children, the elderly, adults, infants, post-operative recovering patients, etc. Different categories can be preset or user-defined to determine whether they require silent protection. Target category is a pre-defined identity category that requires silent control. In this embodiment, the target category includes at least one group of people with high requirements for a quiet sleep environment, such as children, the elderly, and infants. This category can be added or deleted by the operator through the application. Target user refers to a user whose identity category is determined to belong to the target category. For example, if there are both children and adults in the room, then the children are the target users.
[0049] Sleep state detection refers to the process of analyzing vital signs data to determine whether a target user is currently asleep. The output of sleep state detection may include, but is not limited to, at least one of the following: sleep state category (e.g., deep sleep, light sleep), sleep state detection time, sleep state detection duration, and confidence level of the sleep state detection results. Sleep states include both deep and light sleep states, but do not include wakefulness.
[0050] The mute control parameters are configurable parameters used to assist in deciding whether to implement mute. In this embodiment, they include preset mute periods and sleep mode states. The preset mute periods are one or more time intervals set by the operator through the application, such as 22:00-06:00, and can specify a repetition cycle (e.g., Monday to Sunday). Figure 2 As shown. The preset silent period indicates the time window during which the operator expects the monitored user to typically be asleep. Silent periods can be configured independently for different target user categories, such as 22:00-08:00 for children and 21:00-06:00 for the elderly. Sleep mode status refers to the sleep mode status of the air conditioning unit 10 itself. Sleep mode status can usually be manually activated by the operator, for example, by the operator turning on the air conditioning unit 10. Figure 2 The "Sleep Mode" in the interface shown is set to "On". When turned on, the air conditioner will automatically adjust the fan speed, temperature, etc., and this state will also be used as one of the trigger conditions for silent control.
[0051] The sound-generating module refers to the component in the air conditioning device 10 that generates operational feedback sounds, and may include, but is not limited to, at least one such as a buzzer or a voice interaction module. In this embodiment, when all multiple triggering conditions for mute control are met, all sound-generating modules will be in a mute state.
[0052] For example, the controller 104 first detects whether the silent control function of the air conditioning unit 10 is enabled. If it is not enabled, subsequent steps are not executed, and the air conditioner continues to make noise normally; if it is enabled, the controller 104 identifies the user's identity category based on the vital sign data reported by the radar sensor 102. Identity recognition can be performed using a weighted fusion method based on multiple vital sign data; alternatively, the vital sign data can be input into a pre-trained convolutional neural network or recurrent neural network to automatically extract deep features, and the network output layer can directly predict the user's identity category. The specific user identity category recognition method can be selected according to the actual situation, and this embodiment does not impose any restrictions on it.
[0053] Next, the controller 104 determines whether at least one user's identity category belongs to the target category requiring silent control, such as children or the elderly. If not, no further steps are executed, and the air conditioner continues to operate normally. If it does, the controller 104 performs sleep state detection on the target user belonging to the target category. Sleep state detection can employ a time-series probability determination method based on a hidden Markov model; it can also use a recurrent neural network to capture the dynamic patterns of vital sign data changes over time, thereby classifying and outputting the sleep state at each time point; or it can set several simple rule thresholds, such as determining deep sleep when the respiratory rate is within a certain threshold range. The specific sleep state detection method can also be selected according to the actual situation, and this embodiment does not limit it. In some embodiments, the operator can also input relevant information of the target user through a mobile terminal application to assist in identifying the user's identity. Figure 2 As shown, the application interface provides a user information management portal, where operators can add personal profiles for each target user, such as a child, including but not limited to at least one of height and age. This information is uploaded to the controller 104 as reference parameters in the identity recognition process.
[0054] If it is determined that there is at least one target user in a sleep state (including deep sleep or light sleep), the controller 104 further reads the mute control parameters corresponding to the mute control function from the local storage or the cloud, including the mute period preset by the operator and the current sleep mode status of the air conditioning device 10. If there is no target user in a sleep state, no parameter acquisition is performed.
[0055] Furthermore, the controller 104 acquires the current time and compares it with the acquired preset silent period to determine whether the current time falls within any valid period. Simultaneously, the controller 104 checks whether the air conditioner's sleep mode flag is true, i.e., whether sleep mode is activated. If at least one of the above two conditions is met, the controller 104 sends a mute command to the sound-emitting module to instruct it to be in a silent state. If neither of the above two conditions is met, no mute command is sent to the sound-emitting module, and the sound-emitting module operates normally.
[0056] In some embodiments, the controller 104 may predefine a mute mode trigger function to determine whether to execute mute, and the function may be expressed as:
[0057]
[0058] F is the mute mode trigger flag. When F is 1, it means that mute control is required, that is, the sound module is controlled to be in a mute state. When F is 0, it means that mute control is not required and the sound module works normally. The result of the silent period determination is as follows: if the current time falls within any preset silent period, then... Otherwise, it is 0. This is a sleep mode status indicator. If the sleep mode at point 10 of the air conditioner is currently enabled, then... If there is at least one target user belonging to the target category and in a sleeping state, then S = 1; otherwise, it is 0. S is a sleep status indicator for the effective target user. This is the mute control function indicator. If the mute control function is enabled, then... Otherwise, it is 0.
[0059] When the operator turns on the "auto-mute switch" in the application (i.e. There exists at least one target user belonging to the target category and in a sleeping state (i.e., S=1), and the current time is within the preset silent period (i.e., ) or the operator manually turned on the air conditioner's sleep mode (i.e. When this occurs, controller 104 automatically triggers mute control, i.e., F=1. If the operator has checked the "mute" option in the application... Figure 2 The "Automatically mute when sleep mode is enabled" setting shown in the interface means that even if sleep mode is manually enabled, the controller 104 will ignore this condition, i.e., it will be ignored. The value is always 0, and the mute control trigger relies on a preset mute period. This linkage setting allows the operator to decide whether to use sleep mode as the authorization condition for triggering mute control based on user habits. If the operator does not select the "Automatically mute when sleep mode is on" linkage setting, the controller 104 does not need to consider sleep mode when instructing the radar sensor 102 whether to scan. That is, as long as the current period is not within the preset mute period, the radar will continue to scan and monitor in order to capture the user's sleep state in a timely manner. The linkage setting refers to a set of switch options that the operator can configure in the mobile terminal application to control the dependencies between certain automatic control behaviors in the air conditioning device 10. In addition to "Automatically mute when sleep mode is on", the linkage setting may also include "Push notification when a child is detected to sleep" and "Disable all voice wake-up during mute period", etc.
[0060] When there are multiple users being detected in the room, the controller 104 calculates the probability that each user i belongs to the target category. And the probability that user i is currently asleep. The product of the two scores is taken as the overall score for user i. Then, controller 104 selects the maximum overall score from all users:
[0061]
[0062] Where n represents the number of users. If the maximum value... If the value is greater than or equal to a preset threshold (e.g., 0.5), then it is considered that there is at least one target user belonging to the target category and in a dormant state, i.e., S=1. Conversely, if the maximum value is not greater than or equal to the preset threshold, then it is considered that there is at least one target user belonging to the target category and in a dormant state. If the value is less than the threshold, then S=0. This mechanism ensures that the air conditioner enters silent mode as long as any target user in the room falls asleep, thus providing a uniformly quiet environment for all target users.
[0063] In some embodiments, if the operator believes that the silent control effect of the air conditioning device 10 does not meet their expectations, such as triggering silence when the target user is awake, or failing to silence in time after a child falls asleep, the operator can turn off the "automatic silent switch" at any time via the application. When the duration of the silence reaches a threshold (e.g., 5 minutes), the air conditioning device 10 exits silent mode, and the sound module resumes normal sound emission. Simultaneously, the controller 104 records all context data of this silent control decision and the operator's shutdown operation as a set of negative samples, and stores them in local memory or uploads them to the cloud. These negative samples will be used to optimize the weighting of multi-vital parameter fusion during subsequent identity recognition processes.
[0064] In this embodiment, a radar sensor collects the vital signs data of at least one user. With the silent control function of the air conditioner activated, the system identifies whether the user belongs to a target category requiring silent control based on the vital signs data. If at least one user is identified as belonging to the target category, the system further detects the sleep state of the target user based on the vital signs data. If it is determined that at least one target user is asleep, the system further acquires silent control parameters, including a preset silent period and sleep mode status. The sound-emitting module is only silenced when the user is currently in the preset silent period or when the sleep mode is active. Thus, through a multi-layered, progressive judgment logic, the silent control of the air conditioner no longer relies on a single, fixed trigger condition. Instead, it can make dynamic decisions based on the user's identity category, actual sleep state, and flexibly configurable silent period or sleep mode status. This better meets the silent needs of users in diverse usage scenarios, significantly improving the adaptability and intelligence of the air conditioner's silent control.
[0065] In some embodiments, the controller 104 is further configured to: acquire reference vital sign data corresponding to the target category.
[0066] The reference vital signs data are data representing the typical physiological state and morphological characteristics of the target category, and may include, but are not limited to, at least two of the following: reference height, reference body size ratio, and reference respiratory rate. This reference vital signs data can be stored in the internal memory of the controller 104 and can be modified by the operator through an application, for example, adjusting the reference height according to the child's actual height.
[0067] In some embodiments, when performing the step of identifying the user's identity category based on the received vital sign data, the controller 104 is further configured to: compare the vital sign data with reference vital sign data to obtain a data comparison result; detect a first probability that the user belongs to the target category based on the data comparison result; and determine that the user's identity category belongs to the target category if the first probability is greater than or equal to a probability threshold.
[0068] The data comparison result refers to the ratio between the vital sign data and the reference vital sign data. The first probability represents the confidence level that the current user belongs to the target category, ranging from 0 to 1; the closer to 1, the more likely the user is to belong to the target category. The probability threshold is a preset value; the user's identity is confirmed to belong to the target category only when the first probability is greater than or equal to this threshold. The probability threshold can be set according to actual conditions, for example, 0.75.
[0069] For example, after acquiring the vital sign data reported by the radar sensor 102, the controller 104 can first read the reference vital sign data corresponding to the target category from the memory. Then, it calculates the ratio between each actual vital sign data point and its corresponding reference vital sign data. By weighted fusion of these ratios, a fusion score for the user can be obtained. This fusion score reflects the degree of matching between the actual vital sign data and the reference vital sign data; the higher the score, the closer the user is to the typical vital sign of the target category. Next, the controller 104 inputs the fusion score into a non-linear activation function (such as the sigmoid function), mapping it to a probability value between 0 and 1 to obtain a first probability:
[0070]
[0071] in, Let z represent the first probability, and z represent the fusion score. When z is large... It approaches 1; when z is very small, Approaching 0.
[0072] Ultimately, controller 104 compares And the preset probability threshold c. If If so, then the current user's identity category is determined to belong to the target category. If the current user's identity category does not belong to the target category, then the controller 104 will determine that the current user's identity category does not belong to the target category. In the case of multiple users, the controller 104 will independently execute the above calculation process for each user, obtain the first probability for each user, and compare it with the threshold to identify which users belong to the target category.
[0073] In this embodiment, by acquiring reference vital sign data corresponding to the target category, the real-time collected user vital sign data is compared with the reference data, and the probability of the user belonging to the target category is calculated. The identity category is confirmed only when the probability is greater than or equal to the threshold, which effectively improves the accuracy and robustness of user identity recognition and provides a reliable data foundation for subsequent sleep detection and silent control for specific users.
[0074] In some embodiments, the vital sign data includes multiple sub-vital sign data, and the reference vital sign data includes multiple reference sub-vital sign data, with each sub-vital sign data corresponding to a specific reference sub-vital sign data. When the controller 104 performs the step of comparing the vital sign data with the reference vital sign data to obtain a data comparison result, it is further configured to: for each sub-vital sign data, compare the sub-vital sign data with the corresponding reference sub-vital sign data to obtain a sub-comparison result corresponding to the sub-vital sign data; obtain the fusion weight corresponding to each sub-vital sign data; and fuse the sub-comparison results based on each fusion weight to obtain a data comparison result.
[0075] The sub-vital sign data comprises the independent components of the vital sign data. In this embodiment, the sub-vital sign data includes at least two of the following: measured height, measured body proportion, measured respiratory rate, and measured limb movement amplitude. Each sub-vital sign data is extracted from the vital sign data output by the radar sensor. Reference sub-vital sign data comprises the independent components of the reference vital sign data, corresponding one-to-one with the sub-vital sign data. In this embodiment, the reference sub-vital sign data includes at least two of the following: reference height, reference body proportion, and reference respiratory rate. The reference sub-vital sign data should be adjusted accordingly for different target categories.
[0076] It should be noted that the reference sub-sign data corresponding to the measured value of limb movement amplitude is 1, because limb movement amplitude reflects the intensity of the user's limb movements in a short period of time, which belongs to the real-time behavioral state rather than the long-term physiological structural characteristics. The variation of the movement amplitude of the same user in different states far exceeds the statistical differences between different groups of people. There is no stable and discriminative reference value. In contrast, height, body size ratio, respiratory rate, etc., have statistically significant and relatively stable differences between different groups of people.
[0077] Sub-comparison results are quantified values obtained by comparing each sub-vital sign data point with its corresponding reference sub-vital sign data point. Common comparison methods include calculating ratios, but differences can also be calculated. In this embodiment, for height, body size ratio, and respiratory rate, the corresponding sub-comparison results are the ratios of measured values to reference values; for movement variance, the sub-comparison results are directly taken as measured values. The fusion weight is the importance coefficient of each sub-comparison result in the final data comparison result.
[0078] For example, the controller 104 first extracts various sub-vital characteristic data from the vital sign data: height Body proportion respiratory rate Range of motion Simultaneously, pre-stored reference sub-vital data, such as reference height, is read from memory. Body proportion respiratory rate For the range of limb movements, a reference value of 1 can be set. Next, the controller 104 calculates a sub-comparison result for each sub-vital sign data point, including: sub-comparison result 1 = Sub-comparison result 2= Comparison result 3 = Comparison result 4 = .
[0079] Then, the controller 104 obtains the fusion weights corresponding to each sub-sign data. , , , These fusion weights can be preset default values (such as 0.25) or personalized values learned from historical mute control results using machine learning algorithms (such as gradient descent). The controller 104 multiplies each sub-comparison result by its corresponding fusion weight, sums the results, and adds an optional bias term b to obtain the final data comparison result z. The calculation expression is as follows:
[0080]
[0081] The z value is used as the first probability in subsequent calculations. Input.
[0082] Optionally, after the operator manually corrects the mute control decision (e.g., by turning off the "automatic mute switch"), the controller 104 can adjust using an online gradient descent method. , , , The value of .
[0083] In this embodiment, the vital signs data are decomposed into multiple sub-vital signs data, and each is compared with the corresponding reference sub-vital signs data. Then, the comparison results of each sub-sub-sign are weighted and fused using their respective fusion weights. This integrates multi-dimensional information such as height, body shape, respiratory rate, and movement variance, avoiding the bias caused by relying on a single feature or fixed rules. This improves the accuracy and robustness of identity recognition and provides a reliable data foundation for subsequent accurate sleep detection and silent control.
[0084] In some embodiments, the controller 104 is further configured to: when there are historical silent control tasks in the air conditioning equipment, acquire the actual silent control results and silent control feedback results corresponding to the historical silent control tasks; perform deviation analysis on the historical silent control tasks based on the actual silent control results and silent control feedback results to obtain the silent control deviation corresponding to the historical silent control tasks; and adjust the initial weights corresponding to each sub-signal data according to the silent control deviation to obtain the fusion weights corresponding to each sub-signal data.
[0085] Historical mute control tasks refer to mute decision-making tasks executed within a historical period. After each decision, the controller 104 records the relevant data on which the decision was based, the decision result, and the operator's feedback behavior (such as whether the "automatic mute switch" was manually turned off or on) as a task record. These records are stored in local storage or the cloud for subsequent learning. The actual mute control result refers to the mute state (F=0 or F=1) calculated by the controller 104 based on the mute mode trigger function in the historical mute control task. The mute control feedback result refers to the mute control correction behavior initiated by the operator after mute control, such as the operator's manual adjustment (turning on or off) of the "automatic mute switch" through a mobile terminal application.
[0086] For example, if controller 104 determines that mute control is needed (i.e., F=1), but the operator manually turns off the "automatic mute switch," the feedback result is "expect to turn off mute." Conversely, if controller 104 determines that mute is not needed (i.e., F=0), and the operator manually turns on the "automatic mute switch," the feedback result is "expect to turn on mute." If the operator does not make any adjustments, the feedback result can be considered "no feedback" or "default acceptable." In this embodiment, when the mute control feedback result is opposite to the actual mute control result, the task record is considered a negative sample; when the mute control feedback result is the same as the actual mute control result, the task record is considered a positive sample for subsequent learning.
[0087] The silence control bias is used to quantitatively describe the difference between the silence control feedback result and the actual silence control result. In this embodiment, the bias can be defined as a binary value: if the actual silence control result is opposite to the silence control feedback result, the bias is 1; otherwise, the bias is 0. In the gradient descent algorithm, this bias will participate in the calculation of the loss function to guide the adjustment direction of the fusion weights. The initial weights refer to the initial fusion weight values corresponding to each sub-signature data. The initial weights can be a pre-set default value (such as 0.25) or the weights obtained after the last self-learning update.
[0088] For example, the controller 104 periodically (e.g., every 10 cumulative mute control tasks) or manually triggered by the operator, initiates a weighted self-learning process. Specifically, the controller 104 first reads all historical mute control task records that have not been learned from the memory. For each task, it extracts the actual mute control result. And silent control feedback results .like If the noise level is 1, then the task is considered to have a noise control deviation, and the deviation is recorded as 1; if If the operator provides no feedback, the deviation is recorded as 0. Then, the loss function is defined as the mean squared error plus a regularization term:
[0089]
[0090] Where L is the loss function value. N is the number of samples participating in this weight update, i.e., the number of historical silence control tasks, for example, 10. i is the sample index, and j is the weight index. This is the regularization coefficient, with a default value of 0.001, used to prevent overfitting due to excessively large weights. Then, the controller 104 calculates the loss function with respect to each weight using gradient descent. The partial derivatives of the ... Update weights:
[0091]
[0092] Where t represents the iteration step or update number of the current weight update, i.e., the nth update. This represents the weight value before the t-th update or at the beginning of the t-th iteration cycle. This represents the new weight value obtained after the t-th update, which will be used for subsequent mute decisions. Each update increments t by 1. Through continuous iteration, the fused weights gradually converge to a better value, allowing the model to adapt to user preferences.
[0093] In this embodiment, by recording historical silent control tasks and dynamically adjusting the fusion weights of each sub-signal data based on the deviation between the silent control feedback results and the actual silent control results, the air conditioning equipment can continuously learn from the user's corrective behavior and gradually adapt to the current user preferences. This significantly improves the accuracy of identity recognition and the adaptability of silent control, reduces the frequency of manual intervention by the user, and achieves an intelligent and personalized silent control experience.
[0094] In some embodiments, when the controller performs the step of detecting the sleep state of a target user belonging to a target category based on vital sign data and obtaining the sleep state detection result of the target user, it is further configured to: detect the sleep state of a target user belonging to a target category based on vital sign data and obtain a second probability that the target user is currently in a sleep state; compare the second probability with a preset sleep state threshold to obtain a comparison result; and determine the sleep state category of the target user based on the comparison result; wherein the sleep state category includes deep sleep state and light sleep state.
[0095] The second probability refers to the confidence level that the target user is currently asleep (including deep sleep or light sleep), with a value ranging from 0 to 1. The second probability is not a direct measurement but rather a posterior probability calculated by analyzing temporal vital signs such as respiratory rate, limb movement amplitude, and body posture. For example, This indicates a 92% confidence level that the target user is asleep. The sleep state threshold is a critical value used to determine whether the target user is asleep and to distinguish sleep depth. In this embodiment, two thresholds are set: a first state threshold k1 and a second state threshold k2, where the first state threshold k1 is greater than the second state threshold k2. The comparison result is obtained by comparing the second probability with the two thresholds, for example, determining which interval the second probability falls within [0, k2), [k2, k1), or [k1, 1). Sleep state categories include light sleep and deep sleep.
[0096] For example, the controller 104 organizes the vital signs data into an observation sequence in chronological order. ,in, o t ={( f resp (t),( M var (t),( P body (t))},t ∈ [1,T] .in, This represents the target user's body posture angle. Then, the controller 104 uses a pre-trained Hidden Markov Model (HMM) to calculate the posterior probability of the sleep state corresponding to this observation sequence. That is, the second probability:
[0097]
[0098] in, This indicates the observation sequence under the condition that the target user is in a sleep state. The likelihood probability. It is the prior probability, that is, the probability that the user is in a sleep state when there are no observations. It can be set according to time and historical patterns. It is an observation sequence The marginal probability of occurrence is used as a normalization factor.
[0099] After obtaining the second probability, the controller 104 compares it with two preset sleep state thresholds to determine the specific sleep state category of the target user, that is, whether the target user is in deep sleep, light sleep, or awake.
[0100] In some embodiments, when the controller performs the step of determining the sleep state category of the target user based on the comparison result, it is further configured to: determine that the target user is in a light sleep state if the comparison result indicates that the second probability is less than the first state threshold and is greater than or equal to the second state threshold; extract the target user's limb movement amplitude from the vital signs data if the comparison result indicates that the second probability is greater than or equal to the first state threshold; and determine that the target user is in a deep sleep state if the limb movement amplitude is less than the amplitude threshold.
[0101] The first state threshold is a lower probability limit used to distinguish between light sleep and deep sleep, which is preset to 0.85 in this embodiment. Deep sleep is only considered when the second probability is greater than or equal to this threshold. The second state threshold is a lower probability limit used to distinguish between wakefulness and light sleep, which is preset to 0.65 in this embodiment. Light sleep is considered when the second probability is less than the first state threshold but greater than or equal to the second state threshold. The amplitude threshold is a preset critical value for limb movement amplitude. When using limb movement amplitude variance, the amplitude threshold is the limb movement amplitude variance threshold, which is set to 0.3 cm² / s² (square centimeters per square second) in this embodiment. The target user is considered to be in the static state required for deep sleep only when the movement amplitude is less than this threshold.
[0102] For example, controller 104 calculates the second probability using the HMM. Then, the following judgment logic is executed:
[0103]
[0104] when At this point, it cannot be immediately determined that the user is in a deep sleep state; further examination of limb movement amplitude is required. Controller 104 extracts the limb movement amplitude from the current moment's vital sign data, or calculates the variance of the limb movement amplitude based on the limb movement amplitude within a recent time window, and compares it with the corresponding threshold m. If it is less than m, it indicates that the target user's limbs are almost still, with only minor displacement caused by breathing, and is determined to be in a deep sleep state. If it is greater than or equal to m, it indicates that although the probability of sleep is high, there is still significant body movement. In this case, the user should still be in light sleep or in the transition stage from deep sleep to light sleep, and therefore can be determined to be in a light sleep state.
[0105] when At this time, the controller 104 directly determines that the target user is in a light sleep state. At this time, the target user may have slight body movements or irregular breathing, but is still generally asleep. When... At that time, the controller 104 determines that the target user is awake.
[0106] In a multi-user scenario, the controller 104 needs to perform the above judgment process for each target user to obtain the sleep state category of each target user, and then make a comprehensive judgment on whether there is at least one target user who is sleeping.
[0107] Optionally, the amplitude threshold m can be adaptively adjusted according to the user's age, physical condition, or season. For example, infants and young children often have slight limb tremors during sleep, so the threshold can be appropriately relaxed, for example, to 0.5.
[0108] In this embodiment, by calculating the second probability that the target user is in a sleep state and comparing it with multiple preset sleep state thresholds, and by combining the verification with the amplitude of limb movements, it is possible to accurately distinguish between light sleep and deep sleep states, avoiding misjudgments that may be caused by relying on a single probability threshold. This provides a more refined and reliable basis for sleep state for silent control, further improving the accuracy of air conditioner silent control and user experience.
[0109] In some embodiments, the controller 104 is further configured to: acquire environmental parameters of the user's geographical location; adjust the initial state threshold used to characterize the user belonging to the target category entering a sleep state based on the environmental parameters to obtain a sleep state threshold; wherein the environmental parameters are inversely correlated with the sleep state threshold.
[0110] The environmental parameters of the user's geographical location can be understood as climate-related data for the region where the air conditioning unit 10 is located, including but not limited to at least one parameter such as outdoor ambient temperature, outdoor ambient humidity, and season. Specifically, the real-time outdoor temperature and humidity of the user's geographical location can be obtained from meteorological services through the air conditioner's built-in wireless communication module. Alternatively, the initial room temperature collected by the indoor temperature sensor before the air conditioner is turned on can be used as an approximation of the ambient temperature. Seasonal information can be automatically determined by the built-in calendar, for example, June to August is summer, and December to February is winter.
[0111] The initial state threshold refers to a preset probability value used to determine the sleep state of users belonging to the target category. In this embodiment, the initial state threshold may include a first initial state threshold and a second initial state threshold. The first initial state threshold is greater than the second initial state threshold. The values of the two initial thresholds can be determined according to the actual situation. For example, the first initial state threshold is 0.85 and the second initial state threshold is 0.65.
[0112] There is an inverse relationship between environmental parameters and sleep state thresholds; that is, the higher the environmental parameters (e.g., higher temperature, higher humidity), the lower the sleep state threshold; conversely, the lower the environmental parameters (e.g., lower temperature, lower humidity), the higher the sleep state threshold. This is because high temperature and high humidity environments, such as summer, can cause users to sleep more lightly and move more, leading to a systematically lower calculated second probability. Therefore, the threshold needs to be lowered to maintain the same sensitivity, such as reducing it from 0.65 to 0.6.
[0113] For example, before each sleep state detection, the controller 104 may first obtain environmental parameters of the target user's geographical location, including at least one such parameter such as ambient temperature, ambient humidity, and season. Then, the controller 104 calculates the adjustment amount of the sleep state threshold according to the following calculation expression:
[0114]
[0115] in, These are the sleep state thresholds after environmental parameter adjustments, including the adjusted first state threshold and the adjusted second state threshold. It is the initial state threshold, including the first initial state threshold and the second initial state threshold. This is the current ambient temperature. This refers to the current relative humidity. The value represents the seasonal factor, which is 1 for summer, 0.8 for winter, and 0.9 for spring and autumn. , , The weighting coefficients for each environmental parameter are set, with a default value of 0.05, to control the degree of influence of each parameter on the threshold adjustment.
[0116] In this embodiment, by acquiring environmental parameters of the user's geographical location and dynamically adjusting the threshold based on the inverse correlation between these parameters and the sleep state threshold, the threshold automatically decreases in environments such as high temperature and humidity or summer, thereby compensating for the underestimation of sleep probability due to environmental factors and preventing users actually in light sleep from being misjudged as awake. Conversely, the threshold increases accordingly in low temperature and dry conditions or winter. This significantly improves the adaptability of sleep state detection to different climatic conditions, thus ensuring the accuracy and reliability of silent control.
[0117] In some embodiments, such as Figure 3 As shown, when the controller 104 performs the step of controlling the sound module in the air conditioning equipment to be in a silent state, it is further configured to: send a mute command to the buzzer so that the buzzer remains silent when it receives a user operation command; and send a voice broadcast stop command to the voice interaction module so that the voice interaction module stops providing voice broadcast feedback when it receives a user operation command.
[0118] The buzzer in the air conditioning unit 10 is an electronic sound-emitting element that emits a "beep" sound. It is typically driven by the controller 104 through the output of a square wave of a certain frequency. The buzzer is used to confirm user operations, such as pressing a button, remote control commands, or at least one other type. The voice interaction module is a unit in the air conditioning unit 10 used for voice synthesis and wake-up word recognition. It typically includes a voice synthesis chip, a speaker, and a voice wake-up engine. This module can announce the operation result, such as "The temperature has been adjusted to 26 degrees," and respond to preset wake-up words.
[0119] The mute command is a control signal sent by the controller 104 to the buzzer driver circuit, such as pulling the drive enable pin low or stopping the output of a PWM (Pulse Width Modulation) wave, thereby silencing the buzzer. The voice broadcast stop command is a command sent by the controller 104 to the voice interaction module to temporarily disable any voice feedback. This command allows the voice interaction module to enter "mute mode," meaning it will still perform the corresponding action after receiving a user operation but will not broadcast any voice prompts, and may optionally stop responding to wake words.
[0120] For example, when the controller 104 determines that it needs to enter silent mode according to the aforementioned mute trigger logic, it will control the buzzer and voice interaction module to mute. Specifically, the controller 104 sends a mute command to the buzzer driver circuit. When the user subsequently performs any operation (such as pressing any button, adjusting the temperature, switching modes, etc.) through the remote control, mobile application, or air conditioner panel, the buzzer will not emit a "beep" sound, while the air conditioner will still perform the corresponding function normally. At the same time, the controller 104 sends a voice broadcast stop command to the voice interaction module through the communication bus. This command is usually a predefined operation code, such as 0x01 indicating entering silent mode and 0x02 indicating returning to normal. After receiving the command, the voice interaction module sets its internal "mute flag" to true. In this state, even if the module receives user operation events (such as temperature changes, power on / off, etc.), it will not play any voice feedback through the speaker; at the same time, if the module supports wake word detection, it can choose to continue detection without responding, or completely disable wake word detection to save power. When the air conditioner 10 exits silent mode, the controller 104 will send a recovery command to the buzzer and voice interaction module to restore the normal sound output of the buzzer and voice interaction module.
[0121] In some embodiments, the mute control behavior of the controller 104 over the voice interaction module can also be finely configured by the operator through linkage settings in a mobile terminal application. Specifically, the application can provide a switch option called "Disable all voice wake-ups during mute" (e.g., ...). Figure 2 (As shown). When this option is not checked, the voice interaction module can still speak normally even if the air conditioner is in silent mode; in this case, the mute effect only applies to the buzzer. Conversely, if the operator checks the "Disable all voice wake-up during silent mode" option, the mute effect will apply to both the buzzer and the voice interaction module.
[0122] In this embodiment, by sending a mute command to the buzzer and a voice broadcast stop command to the voice interaction module, the entire chain of air conditioner operation prompts is suppressed during the user's sleep. This not only prevents the buzzer's "beep" sound but also disables voice feedback, thereby eliminating the operation feedback noise emitted by the air conditioner itself, creating a truly undisturbed sleep environment for the user, and significantly improving the user experience.
[0123] In some embodiments, the controller 104 is further configured to: determine at least one other device that is communicatively connected to the air conditioning unit and has a prompt sound output function; and send a mute linkage command to the other device to cause the other device to turn off the prompt sound output function.
[0124] Other devices refer to electrical appliances that are communicatively connected to the air conditioning unit 10 and have the function of emitting a prompting sound when operation is completed or status changes. These devices are usually located on the same home local area network as the air conditioning unit 10. Examples include at least one washing machine, dryer, dishwasher, and robot vacuum cleaner. The prompting sound output function refers to the function of other devices emitting a "beep" sound or melodic tone through a built-in buzzer or speaker when completing preset tasks (such as washing, drying, washing finished, cleaning) or experiencing status changes (such as malfunction, standby). This function is intended to remind users, but it may become a source of interference when users are sleeping. The mute linkage command is a control message broadcast or unicast by the air conditioning controller 104 to other devices through a wireless communication protocol. This command carries a command requesting the receiving device to temporarily turn off its prompting sound output function, and usually includes at least one parameter such as the command type (i.e., mute) and the mute duration (e.g., until the child wakes up).
[0125] For example, such as Figure 4 As shown, when the air conditioner controller 104 determines that it needs to enter silent mode, in addition to controlling the sound module of the air conditioner 10 to mute, it will also actively send a mute linkage command to other devices that support linkage functions (such as other devices 1, other devices 2... other devices n) in the same home network as the air conditioner 10. Specifically, the air conditioner controller 104 periodically scans other devices in the local area network to determine which devices have the ability to receive linkage commands and support the function of turning off the prompt sound, and maintains a list of linkage devices, which may include at least one piece of information such as device type, network address, and supported command set. When the air conditioner silent mode is triggered, the controller 104 traverses each device in the linkage device list and sends a mute linkage command according to the communication protocol supported by each device. After receiving the mute linkage command, other devices can set their own prompt sound output function flag to false. After that, even if the device completes the predetermined work cycle, such as the washing machine finishing the spin cycle, it will not emit any prompt sound, but may remind the user through other means (such as flashing indicator lights, pushing notifications to the terminal, etc.). When the air conditioner exits silent mode, the controller 104 will send a "restore prompt tone" command to other devices to restore the normal prompt tone output of other devices.
[0126] Optionally, the operator can customize the list of other devices that need to be linked in the application, and can set whether to activate linkage silence. In addition, the air conditioner controller 104 can also decide whether to send linkage commands based on the user's sleep depth. For example, when the user is in deep sleep, all other devices can be linked, while during light sleep, only some noisier devices such as the washing machine can be linked.
[0127] In this embodiment, when the air conditioner meets the silent conditions, it can actively send a linkage command to other devices with prompt sound functions in the same network to turn off the prompt sound output of these devices, thereby extending the silent protection range from the air conditioner itself to the entire home environment, avoiding interference caused by prompt sounds emitted by other devices during the user's sleep, and creating a more complete and quieter sleep environment for the user.
[0128] In one specific embodiment, to more clearly illustrate the division of software and hardware modules and their collaborative relationships on which the above functions depend, the following is combined with... Figure 5 The system architecture diagram shown illustrates the various functional modules that make up the air conditioning equipment 10. Figure 4 Each module depicted is an internal component of the air conditioning unit 10 or a component connected to it in communication. The controller 104 (not shown in the figure) serves as the core decision-making unit, interacting with each module for data exchange and command transmission. Specifically, it includes an interaction module, a user status recognition module, a parameter collection module, a noise reduction decision module, and a noise control module. Among these:
[0129] The interaction module corresponds to a mobile terminal application. Operators use the mobile terminal application to turn the automatic mute master switch on or off, set the mute period, turn sleep mode on or off, input reference vital sign data, and configure various linkage options. This data is transmitted to the parameter collection module for storage. When the air conditioner is running, radar sensors continuously collect vital sign data from each user in the room, and this data is also sent to the parameter collection module for storage.
[0130] The user status recognition module obtains the vital signs data of each user from the parameter collection module, performs identity recognition and sleep status detection for the target user, and feeds back the identity recognition results and sleep status detection results to the parameter collection module for storage.
[0131] The parameter collection module gathers feedback from the user status recognition module, configuration data from the interaction module, and environmental parameters (such as outdoor temperature, humidity, and season). This module organizes and caches all parameters for other modules to access in real time.
[0132] The mute decision module is the core of this embodiment. This module reads various types of data provided by the parameter collection module and executes the mute trigger function according to the preset mute trigger function. The system performs logical operations. When the judgment condition is met, a signal indicating that the silence condition is met is sent to the silence control module. Simultaneously, this module also possesses self-learning capabilities; when the operator manually corrects the silence decision, it records negative samples and updates the fusion weights in the identity recognition model using gradient descent.
[0133] The mute control module receives feedback signals from the mute decision module and performs actual mute control on the buzzer and voice interaction module inside the air conditioning unit.
[0134] The following section uses a scenario involving silent control for children as an example to explain in detail the implementation process of silent control. Figure 6 A schematic diagram of the mute control process is shown. Figure 7 A schematic diagram illustrating the interactions between the aforementioned functional modules is shown. (Refer to...) Figure 6 , 7 As shown:
[0135] First, parents use a mobile application (the interactive module) to turn on the air conditioner's "automatic silent mode," setting the silent period to 10:00 PM to 6:00 AM the next day, and recording the height and age of their child. This configuration data is then transmitted to the parameter collection module for storage.
[0136] When the air conditioner is running, radar sensors continuously scan the room area, such as Figure 8 As shown. Simultaneously, the silent mode decision module reads the status of the "automatic silent switch" from the parameter collection module to determine if it is on. If on, it is used as one of the trigger conditions for silent mode; if not on, the air conditioner is in non-silent mode. When the "automatic silent switch" is on, it further reads the preset silent period and obtains the current time in real time to determine if it is within that preset silent period. If within that period, it is used as one of the trigger conditions for silent mode; if not within that period, it further checks whether the air conditioner's sleep mode is on. If sleep mode is on, it is used as one of the trigger conditions for silent mode; if sleep mode is also off, silent control is not executed. At this time, the radar sensor continuously scans the room area to detect the presence of users. If no users are detected, the air conditioner is in non-silent mode, waiting for the next detection cycle; if a user is detected, it further collects the user's vital signs data, including height, body proportions, respiratory rate, variance of limb movement amplitude, and body angle, and sends this data to the parameter collection module for storage.
[0137] The user status recognition module acquires these vital sign data through the parameter collection module and performs identity recognition for each user based on this data. Specifically, it compares the measured height, measured body proportion, and measured respiratory rate with corresponding reference data to obtain the height ratio, body proportion, and respiratory rate ratio. Then, it multiplies the height ratio, body proportion, respiratory rate ratio, and variance of limb movement amplitude by their respective fusion weights and sums them. After nonlinear mapping, it obtains the probability (i.e., the first probability) that each user is a child. If the user is a child, their sleep state is further detected; if not, the air conditioner is set to non-silent mode.
[0138] After successful identity verification, the user status recognition module detects the child's sleep state. Using a Hidden Markov Model, it analyzes the continuously observed sequence of vital signs data to calculate the probability (i.e., the second probability) that the child is in a sleep state (deep sleep or light sleep). For example, assuming a probability of 0.88, greater than the first state threshold of 0.85, and a variance in limb movement amplitude of 0.1, less than the threshold of 0.3, then the child is determined to be in deep sleep. If the child is determined to be asleep, this is used as one of the trigger conditions for the silent mode; if the child is determined to be awake, the air conditioner is set to non-silent mode. The identity verification results and sleep detection results are sent to the parameter collection module for storage.
[0139] The silent mode decision module reads various data from the parameter collection module and determines whether the air conditioner's silent mode should be triggered based on this data. Assume the "automatic silent mode switch" is on (i.e....). The current time is 22:15, which is within the preset silent period (i.e.) Sleep mode is not enabled (i.e.) There exists a child in a sleeping state (S=1). Substitute the values corresponding to the above conditions into the mute trigger function. Finally, F=1 is obtained, which means that the mute condition is met, the mute mode is triggered, and a feedback signal that the mute condition is met is sent to the mute control module.
[0140] Upon receiving the signal, the mute control module sends a mute command to the buzzer, preventing it from responding to any remote control or panel operations. Simultaneously, it sends a stop voice broadcast command to the voice interaction module, preventing it from providing voice feedback and responding to wake words. At this point, the air conditioner is in a completely silent state. If the parent unchecks the "Disable all voice wake-ups during mute" option in the application, the voice interaction module can resume operation independently.
[0141] When the radar detects an increase in the variance of the child's limb movements, a change in posture to sitting, and a decrease in the probability of sleep to 0.4, the user status recognition module updates the sleep status to "awake." After the parameter collection module updates the parameters, the mute decision module recalculates F. Since S=0, F=0. At this time, the mute decision module sends a feedback signal to the mute control module indicating that the mute condition is not met. The mute control module then sends a command to the buzzer and voice interaction module to resume normal sound production.
[0142] In this embodiment, a radar sensor collects the vital signs data of at least one user. With the air conditioner's silent control function enabled, the system identifies whether the user belongs to a target category requiring silent control based on the vital signs data. If at least one user is identified as a child, the system further detects the child's sleep state based on the vital signs data. If at least one child is confirmed to be asleep, the system further acquires silent control parameters, including a preset silent period and sleep mode status. The sound-emitting module is only silenced when the user is currently in the preset silent period or when the sleep mode is enabled. Thus, through a multi-layered, progressive judgment logic, the silent control of the air conditioner no longer relies on a single, fixed trigger condition. Instead, it dynamically decides based on the user's identity category (child), actual sleep state, and flexibly configurable silent period or sleep mode status. This better meets the silent needs of children during sleep, significantly improving the adaptability and intelligence of the air conditioner's silent control.
[0143] Based on the same inventive concept, some embodiments of this application also provide a control method for an air conditioning device, applied to the aforementioned air conditioning device 10. For example... Figure 9 As shown, the method includes:
[0144] Step S902: When the silent control function of the air conditioning equipment is turned on, the user's identity category is identified based on the vital sign data of at least one user collected by the radar sensor.
[0145] Step S904: If at least one user's identity category belongs to the target category, sleep state detection is performed on the target user belonging to the target category based on vital sign data to obtain the sleep state detection result of the target user; wherein, the target category refers to the identity category with mute control requirements.
[0146] Step S906: If it is determined from the sleep state detection results that there is at least one target user in a sleep state, obtain the mute control parameters corresponding to the mute control function; the mute control parameters include the preset mute period and sleep mode state.
[0147] Step S908: When the current time is within the preset silent period or when the sleep mode is on, control the sound module in the air conditioner to be in a silent state.
[0148] In this embodiment, a radar sensor collects the vital signs data of at least one user. With the silent control function of the air conditioner activated, the system identifies whether the user belongs to a target category requiring silent control based on the vital signs data. If at least one user is identified as belonging to the target category, the system further detects the sleep state of the target user based on the vital signs data. If it is determined that at least one target user is asleep, the system further acquires silent control parameters, including a preset silent period and sleep mode status. The sound-emitting module is only silenced when the user is currently in the preset silent period or when the sleep mode is active. Thus, through a multi-layered, progressive judgment logic, the silent control of the air conditioner no longer relies on a single, fixed trigger condition. Instead, it can make dynamic decisions based on the user's identity category, actual sleep state, and flexibly configurable silent period or sleep mode status. This better meets the silent needs of users in diverse usage scenarios, significantly improving the adaptability and intelligence of the air conditioner's silent control.
[0149] In some embodiments, the control method for the air conditioning device further includes: acquiring reference vital sign data corresponding to the target category; identifying the user's identity category based on the received vital sign data, including: comparing the vital sign data with the reference vital sign data to obtain a data comparison result; detecting a first probability that the user belongs to the target category based on the data comparison result; and determining that the user's identity category belongs to the target category if the first probability is greater than or equal to a probability threshold.
[0150] In some embodiments, the vital sign data includes multiple sub-vital sign data, and the reference vital sign data includes multiple reference sub-vital sign data, with each sub-vital sign data corresponding to a specific reference sub-vital sign data. The vital sign data is compared with the reference vital sign data to obtain a data comparison result, including: for each sub-vital sign data, comparing the sub-vital sign data with the corresponding reference sub-vital sign data to obtain a sub-comparison result corresponding to the sub-vital sign data; obtaining the fusion weight corresponding to each sub-vital sign data; and fusing the sub-comparison results based on each fusion weight to obtain a data comparison result.
[0151] In some embodiments, the control method for the air conditioning equipment further includes: when the air conditioning equipment has historical silent control tasks, obtaining the actual silent control results and silent control feedback results corresponding to the historical silent control tasks; performing deviation analysis on the historical silent control tasks based on the actual silent control results and silent control feedback results to obtain the silent control deviation corresponding to the historical silent control tasks; and adjusting the initial weights corresponding to each sub-signal data according to the silent control deviation to obtain the fusion weights corresponding to each sub-signal data.
[0152] In some embodiments, sleep state detection is performed on target users belonging to the target category based on vital sign data to obtain sleep state detection results for the target users, including: performing sleep state detection on target users belonging to the target category based on vital sign data to obtain a second probability that the target user is currently in a sleep state; comparing the second probability with a preset sleep state threshold to obtain a comparison result; and determining the sleep state category of the target user based on the comparison result; wherein, the sleep state category includes deep sleep state and light sleep state.
[0153] In some embodiments, the sleep state threshold includes a first state threshold and a second state threshold, wherein the first state threshold is greater than the second state threshold; determining the sleep state category of the target user based on the comparison result includes: determining that the target user is in a light sleep state when the comparison result indicates a second probability less than the first state threshold and greater than or equal to the second state threshold; extracting the target user's limb movement amplitude from the vital signs data when the comparison result indicates a second probability greater than or equal to the first state threshold; and determining that the target user is in a deep sleep state when the limb movement amplitude is less than the amplitude threshold.
[0154] In some embodiments, the control method for the air conditioning device further includes: acquiring environmental parameters of the user's geographical location; adjusting an initial state threshold for characterizing a user belonging to a target category to enter a sleep state based on the environmental parameters, thereby obtaining a sleep state threshold; wherein the environmental parameters are inversely correlated with the sleep state threshold.
[0155] In some embodiments, the sound-generating module includes a buzzer and a voice interaction module; controlling the sound-generating module in the air conditioning unit to be in a silent state includes: sending a mute command to the buzzer so that the buzzer remains silent when it receives a user operation command; and sending a voice broadcast stop command to the voice interaction module so that the voice interaction module stops providing voice broadcast feedback when it receives a user operation command.
[0156] In some embodiments, the control method for the air conditioning device further includes: determining at least one other device that is communicatively connected to the air conditioning device and has a prompt sound output function; and sending a mute linkage command to the other device to cause the other device to turn off the prompt sound output function.
[0157] The specific execution process of the control method for the air conditioning equipment can be referred to the description of the air conditioning equipment above, and will not be repeated here. It should be understood that, although the various steps involved in the embodiments described above, unless explicitly stated herein, are not strictly limited in the order of execution, these steps can be executed in other orders. Moreover, at least some of the steps involved in the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a part of the steps or stages in other steps.
[0158] Based on the same inventive concept, this application also provides a control device for an air conditioning device to implement the control method of the air conditioning device described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more air conditioning device control device embodiments provided below can be found in the limitations of the air conditioning device control method described above, and will not be repeated here. In an exemplary embodiment, as shown in FIG10, a control device for an air conditioning device is also provided, including:
[0159] The user identification module 1002 is used to identify the user's identity category based on the vital sign data of at least one user collected by a radar sensor when the silent control function of the air conditioning equipment is turned on.
[0160] The sleep state detection module 1004 is used to detect the sleep state of a target user belonging to the target category based on vital sign data when at least one user's identity category belongs to the target category, and obtain the sleep state detection result of the target user; wherein, the target category refers to the identity category that has a mute control requirement;
[0161] The parameter acquisition module 1006 is used to acquire the mute control parameters corresponding to the mute control function when it is determined that there is at least one target user in a sleep state based on the sleep state detection results; the mute control parameters include the preset mute period and sleep mode state;
[0162] The silent state control module 1008 is used to control the sound-emitting module in the air conditioning equipment to be in a silent state when the current time is within a preset silent period or when the sleep mode is on.
[0163] The specific processing methods of each module in the control device of the aforementioned air conditioning equipment can be found in the description of the air conditioning equipment. Each module of the control device of the air conditioning equipment can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the computer device in hardware form or independent of it, or they can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.
[0164] In one embodiment, an electronic device is provided, which can serve as a control device for an air conditioning unit. This device may include a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The communication interface of the computer device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a control method for an air conditioning unit. The display screen of the electronic device may be a liquid crystal display (LCD) or an e-ink display. The input device may be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.
[0165] Those skilled in the art will understand that the structure of the electronic device described above is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.
[0166] In one exemplary embodiment, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in any of the above-described air conditioning device control method embodiments.
[0167] In one exemplary embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in any of the above-described embodiments of the control method for a heating and air conditioning device.
[0168] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in any of the above-described air conditioning device control method embodiments.
[0169] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Furthermore, any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory.
[0170] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0171] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. An air conditioning device, characterized in that, include: The radar sensor is configured to collect vital sign data from at least one user. The controller, connected to the radar sensor, is configured to: When the silent control function of the air conditioning equipment is enabled, the user's identity category is identified based on the received vital sign data; If at least one user's identity category belongs to the target category, sleep status detection is performed on the target user belonging to the target category based on the vital sign data to obtain the sleep status detection result of the target user; wherein, the target category refers to the identity category with mute control requirements; If, based on the sleep state detection results, it is determined that there is at least one target user in a sleep state, the mute control parameters corresponding to the mute control function are obtained; the mute control parameters include a preset mute period and a sleep mode state; When the current time is within the preset silent period, or when the sleep mode is enabled, the sound module of the air conditioner is controlled to be in a silent state.
2. The air conditioning equipment according to claim 1, characterized in that, The controller is also configured to: Obtain reference vital sign data corresponding to the target category; When performing the step of identifying the user's identity category based on the received vital sign data, the controller is further configured to: The vital sign data are compared with the reference vital sign data to obtain the data comparison results; Based on the data comparison results, a first probability is determined that the user belongs to the target category; If the first probability is greater than or equal to the probability threshold, the user's identity category is determined to belong to the target category.
3. The air conditioning equipment according to claim 2, characterized in that, The vital sign data includes multiple sub-vital sign data, and the reference vital sign data includes multiple reference sub-vital sign data, with each sub-vital sign data corresponding one-to-one with the reference sub-vital sign data; when the controller performs the step of comparing the vital sign data with the reference vital sign data to obtain a data comparison result, it is further configured to: For each of the sub-sign data, the sub-sign data is compared with the corresponding reference sub-sign data to obtain the sub-comparison result corresponding to the sub-sign data; Obtain the fusion weight corresponding to each of the sub-sign data; Based on the fusion weights, the sub-comparison results are fused to obtain the data comparison results.
4. The air conditioning equipment according to claim 3, characterized in that, The controller is further configured to: If the air conditioning equipment has a historical silent control task, obtain the actual silent control result and silent control feedback result corresponding to the historical silent control task; Based on the actual mute control results and the mute control feedback results, a deviation analysis is performed on the historical mute control tasks to obtain the mute control deviations corresponding to the historical mute control tasks. Based on the silence control deviation, the initial weights corresponding to each of the sub-sign data are adjusted to obtain the fusion weights corresponding to each of the sub-sign data.
5. The air conditioning equipment according to claim 1, characterized in that, When the controller performs the step of detecting the sleep state of target users belonging to the target category based on the vital sign data and obtaining the sleep state detection results of the target users, it is further configured to: Based on the vital signs data, sleep status detection is performed on target users belonging to the target category to obtain a second probability that the target user is currently in a sleep state. The second probability is compared with a preset sleep state threshold to obtain the comparison result; The sleep state category of the target user is determined based on the comparison results; wherein, the sleep state category includes deep sleep and light sleep.
6. The air conditioning equipment according to claim 5, characterized in that, The sleep state threshold includes a first state threshold and a second state threshold, wherein the first state threshold is greater than the second state threshold. When performing the step of determining the sleep state category of the target user based on the comparison result, the controller is further configured to: If the comparison result indicates that the second probability is less than the first state threshold and greater than or equal to the second state threshold, it is determined that the target user is in the light sleep state. If the comparison result indicates that the second probability is greater than or equal to the first state threshold, the amplitude of the target user's limb movements is extracted from the vital sign data. If the amplitude of the limb movement is less than the amplitude threshold, the target user is determined to be in the deep sleep state.
7. The air conditioning equipment according to claim 5, characterized in that, The controller is also configured to: Obtain the environmental parameters of the user's geographical location; Based on the environmental parameters, the initial state threshold used to characterize users belonging to the target category entering a sleep state is adjusted to obtain the sleep state threshold; wherein, the environmental parameters are inversely correlated with the sleep state threshold.
8. The air conditioning equipment according to claim 1, characterized in that, The sound-generating module includes a buzzer and a voice interaction module; When the controller performs the step of controlling the sound-emitting module in the air conditioning device to be in a silent state, it is further configured to: Send a mute command to the buzzer so that the buzzer remains silent when it receives a user operation command; Send a voice broadcast stop command to the voice interaction module so that the voice interaction module stops providing voice broadcast feedback when it receives a user operation command.
9. The air conditioning equipment according to any one of claims 1 to 8, characterized in that, The controller is also configured to: Identify at least one other device that is communicatively connected to the air conditioning unit and has a prompt tone output function; Send a mute command to the other devices to disable the notification sound output function.
10. A control method for an air conditioning device, characterized in that, The method includes: With the silent control function of the air conditioning unit enabled, the identity category of the user is identified based on the vital sign data of at least one user collected by the radar sensor. If at least one user's identity category belongs to the target category, sleep status detection is performed on the target user belonging to the target category based on the vital sign data to obtain the sleep status detection result of the target user; wherein, the target category refers to the identity category with mute control requirements; If, based on the sleep state detection results, it is determined that there is at least one target user in a sleep state, the mute control parameters corresponding to the mute control function are obtained; the mute control parameters include a preset mute period and a sleep mode state; When the current time is within the preset silent period, or when the sleep mode is enabled, the sound-emitting module in the air conditioning device is controlled to be in a silent state.
