Smart home equipment control method

Abstract

The invention provides a smart home equipment control method, and belongs to the technical field of smart home, and the method comprises the steps: inputting the equipment data of equipment at the current moment into a first prediction model, and obtaining a first value of a user behavior management and control parameter outputted by the first prediction model; inputting the device data of the current moment of the to-be-controlled smart home device and the device data of the preset duration before the current moment into a second prediction model, and obtaining a second value output by the second prediction model; and issuing a first value output by the first prediction model or a second value output by the second prediction model corresponding to each group of user behavior control parameters to the to-be-controlled smart home device. The method achieves the prediction of the control behavior of the user through training two complementary prediction models, automatically completes the control of the intelligent household equipment by the user, is low in configuration difficulty, and is wide in application range. And the uplink data and the downlink data are jointly used as the input of the prediction model, so that the prediction accuracy of the prediction model on the user management and control behavior is improved.

Description

Technical Field

[0001] The present invention relates to the field of smart home technology, and in particular to a smart home scene control method. Background Art

[0002] Smart home uses residence as a platform, uses the Internet and other technologies to integrate home life-related devices, and realizes intelligent management and control through smart home scene rules.

[0003] The scene rules of smart homes refer to the ability of devices to automatically execute when triggered without manual user intervention or with minimal intervention. There are two main types of smart scene rules: one is the automation script preset by the platform or device. When the user binds the device, the device or platform automatically generates relevant scene rules and saves them on the central control device or platform; when the device meets the preset trigger conditions, the home central control device or platform will execute the preset automation action.

[0004] The second is that users can create custom rules based on their needs through the scene rule capabilities provided by the platform to manage and control home devices. The scene rule capability refers to the device scene rule elements in the object model.

[0005] Since the scenarios of preset automation scripts are simple, lack variety, and cover fewer types of devices, it is difficult to meet the diverse actual needs of users; and the configuration of user-created custom rules is complex and has a high learning cost, making it difficult for ordinary users to go deeper.

[0006] Therefore, how to realize a smart home device control method that can meet the diverse needs of users while having a simple configuration method is still an important problem that needs to be solved urgently. Summary of the invention

[0007] The present invention provides a smart home device control method to solve the defect in the prior art that it is difficult to meet the diverse needs of users through a simple configuration method, and realize the scene rules of smart home automation.

[0008] The present invention provides a smart home device control method, comprising:

[0009] Inputting the device data of the smart home device to be controlled at the current moment into the first prediction model, and obtaining a first value of at least one set of user behavior control parameters output by the first prediction model;

[0010] Inputting device data of the current moment and a preset time before the current moment of the smart home device to be controlled into the second prediction model, and obtaining a second value of the user behavior control parameter output by the second prediction model;

[0011] Sending the first value output by the first prediction model or the second value output by the second prediction model corresponding to each group of user behavior control parameters to the smart home device to be controlled;

[0012] Among them, each group of user behavior control parameters is pre-associated with the first prediction model, or pre-associated with the second prediction model; the first value or the second value is used to control the smart home device to be controlled.

[0013] The present invention also provides a smart home device control device, comprising:

[0014] An acquisition module, used to input the device data of the smart home device to be controlled at the current moment into the first prediction model, and obtain a first value of at least one set of user behavior control parameters output by the first prediction model;

[0015] It is also used to input the device data of the current moment and the preset time before the current moment of the smart home device to be controlled into the second prediction model, and obtain the second value of the user behavior control parameter output by the second prediction model;

[0016] A sending module sends the first value output by the first prediction model or the second value output by the second prediction model corresponding to each group of user behavior control parameters to the smart home device to be controlled;

[0017] Wherein, each group of user behavior control parameters is pre-associated with the first prediction model, or each group of user behavior control parameters is pre-associated with the second prediction model.

[0018] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned smart home device control methods is implemented.

[0019] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the computer program implements any of the above-mentioned smart home device control methods.

[0020] The present invention also provides a computer program product, including a computer program, which implements any of the above-mentioned smart home device control methods when executed by a processor.

[0021] A smart home device control method provided by the present invention obtains a complementary first prediction model and a second prediction model through device data training of different durations, which are used to accurately predict independent user control behaviors or user control behaviors with associated relationships; among the user behavior control parameters input by the first prediction model and the second prediction model, according to the pre-established association relationship between the user behavior control parameters and different prediction models, the first value or the second value output by the prediction model with higher prediction accuracy is selected for actual distribution, so as to realize the prediction of user control behavior; the configuration method is simple, and the accuracy of user control behavior prediction is improved by considering the influence of time on user control behavior, so as to meet the user's diversified control needs for devices, and realize a more automated smart home scene rule. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 This is one of the flow charts of the smart home device control method provided by the present invention;

[0024] Figure 2 is a schematic diagram of a prediction model used in training in the smart home device control method provided by the present invention;

[0025] Figure 3 is a schematic diagram of a physical model in the smart home device control method provided by the present invention;

[0026] Figure 4 It is a schematic diagram of the process of device data collection in the smart home device control method provided by the present invention;

[0027] Figure 5 It is a structural schematic diagram of the smart home equipment control device provided by the present invention;

[0028] Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] Combine the following Figure 1-Figure 4 Describe the smart home device control method of the present invention, such as Figure 1 , the method comprising:

[0031] Step 101: inputting the device data of the smart home device to be controlled at the current moment into the first prediction model, and obtaining the first value of at least one set of user behavior control parameters output by the first prediction model;

[0032] The smart home devices to be controlled are smart home devices that are controlled by smart home scene rules.

[0033] Optionally, the smart home devices to be controlled can be all devices connected to the home central control device or the Internet of Things platform (hereinafter referred to as the platform); or they can be devices pre-selected by the user among the devices connected to the home central control device or the platform, so that the scene rules of the smart home can better meet the user's needs when controlling the devices.

[0034] Optionally, the smart home devices to be controlled do not include some devices that may cause danger to the user, such as kettles, door and window switches, etc., in order to ensure the safety of the user.

[0035] The device data of the smart home device to be controlled is data related to the smart home device to be controlled, and may include parameters corresponding to the functional status of the device itself, such as the switch status of a door lock, the brightness status of a light, or the temperature status of an air conditioner. It may also include relevant data when a smart home device to be controlled is used, such as the time when the device is used, the duration of use of the device, the external ambient temperature when the device is in use, and the external ambient temperature when the device is shut down.

[0036] Optionally, the device data may include data related to one smart home device to be controlled, or may include data related to multiple smart home devices to be controlled.

[0037] The first prediction model is pre-trained and is used to predict the user's control behavior on the smart home device based on the device data of the smart home device at a certain moment, so as to realize truly automated scene rules and reduce the user's configuration cost.

[0038] Specifically, the first prediction model outputs a first value of a user behavior control parameter, which is a control parameter of different functions of a device by the user. The first value is sent to the corresponding smart home device to be controlled, and the device is adjusted to the state predicted to be required by the user.

[0039] Optionally, each set of user behavior control parameters may include parameters corresponding to different functions of the same device, or may include parameters of different devices.

[0040] Optionally, the user behavior control parameters may include changes and invariances of function parameters corresponding to the functional state of the device. For example, the parameter corresponding to the temperature of the air conditioner at the current moment is 26°C. In the user behavior control parameters output by the first prediction model, the value of the corresponding air conditioner temperature parameter in the first value may be 25°C, that is, the temperature of the air conditioner changes after the first value is issued; or it may be 26°C, that is, the temperature of the air conditioner remains unchanged after the first value is issued.

[0041] By inputting the device data of the smart home device to be controlled at the current moment into the first prediction model, a first value of at least one set of user behavior control parameters output by the first prediction model can be obtained.

[0042] Step 102: input the device data of the current moment and the preset time before the current moment of the smart home device to be controlled into the second prediction model, and obtain the second value of the user behavior control parameter output by the second prediction model; the second prediction model is pre-trained and used to predict the user's control behavior on the smart home device through the device data at the current moment and the preset time before the current moment.

[0043] Optionally, the preset duration may be determined according to device type, time, environment, etc.

[0044] Specifically, the control behavior of the user's device is correlated with time, that is, multiple different control behaviors performed by the user on different devices or different functions of the same device within a preset time period are correlated with each other.

[0045] For example, when a user is used to opening the curtains immediately after returning home, the user's control behavior on the device includes: controlling the door lock to open, then controlling the door lock to close, and then controlling the curtains to open. In other words, from the time the user returns home to the time the user opens the curtains, there is a causal relationship between the user's multiple control behaviors on the door lock and the curtains.

[0046] Therefore, by training the second prediction model, the user's control behavior is predicted based on the device data within a period of time. When multiple control behaviors of the user on the device are correlated, the user's control behavior can be predicted more accurately.

[0047] Optionally, this application does not limit the duration of device data.

[0048] For example, the second prediction model may be trained based on the device data at multiple times within 10 consecutive minutes; or the second prediction model may be further trained based on the device data at multiple times within 24 hours.

[0049] The preset time lengths are different, and the application scope of the second prediction model is correspondingly different.

[0050] By inputting the device data of the smart home devices to be controlled at the current moment and within a preset time period before the current moment into the second prediction model, a second value of at least one set of user behavior control parameters output by the second prediction model can be obtained.

[0051] Step 103, sending the first value output by the first prediction model or the second value output by the second prediction model corresponding to each group of user behavior control parameters to the smart home device to be controlled;

[0052] Among them, each group of user behavior control parameters is pre-associated with the first prediction model, or pre-associated with the second prediction model; the first value or the second value is used to control the smart home device to be controlled.

[0053] A control parameter for user behavior is obtained based on the device data at the current moment through the first prediction model; another control parameter for user behavior is obtained based on the device data at the current moment through the second prediction model, and the value corresponding to the control parameter of user behavior is sent to the smart home device to be controlled, so that the smart home device can be controlled to adjust to the functional state predicted to be required by the user.

[0054] Since the functions of some devices or different functions of the same device are correlated in the control parameters of user behavior, the output results of the second prediction model are more accurate; the functions of some devices or different functions of the same device are independent of each other, so the output results of the first prediction model are more accurate.

[0055] Therefore, it is necessary to pre-associate each group of user behavior management parameters with the first prediction model or the second prediction model with a higher prediction accuracy.

[0056] For each group of user behavior control parameters, the first value or the second value output by the first prediction model or the second prediction model associated therewith is selected and sent to the smart home device to be controlled to control the smart home device to be controlled.

[0057] For example, the device data at the current moment is input into the first prediction model to obtain the first values ​​corresponding to device A, device B, and device C; the device data at the current moment, i.e., the preset time length before the current moment, is input into the second prediction model to obtain the second values ​​corresponding to device A, device B, and device C.

[0058] According to the pre-established association relationship, it is determined that the prediction result of the first prediction model for device A is more accurate, and it is determined that the prediction result of the second prediction model for device B and device C is more accurate, then the first value corresponding to device A, the second value corresponding to device B, and the second value corresponding to device C are sent to the corresponding devices A, B, and C.

[0059] The present invention obtains complementary first prediction models and second prediction models through equipment data training of different durations, which are used to accurately predict independent user control behaviors or user control behaviors with associated relationships; among the user behavior control parameters input by the first prediction model and the second prediction model, according to the pre-established association between the user behavior control parameters and different prediction models, the first value or the second value output by the prediction model with higher prediction accuracy is selected for actual distribution, so as to realize the prediction of user control behavior, and the configuration method is simple, and the accuracy of user control behavior prediction is improved by considering the influence of time on user control behavior, which can meet the user's diversified control needs for equipment and realize a more automated smart home scene rule.

[0060] In the smart home device control method of the present invention, before the step of inputting the device data of the smart home device to be controlled at the current moment into the first prediction model and obtaining the first value of at least one set of user behavior control parameters output by the first prediction model, the method further includes:

[0061] The device data of the smart home device at multiple first moments are respectively used as inputs of the prediction model in multiple training processes, and the prediction model is iteratively trained multiple times to obtain a first prediction model;

[0062] The device data at the first moment is the device data at a moment in the training set.

[0063] The first prediction model is an instantaneous prediction model, which is obtained by training the device data at a moment and is used to predict user control behaviors that have no correlation or little correlation with other control behaviors.

[0064] The device data of the smart home device in the first time periods are used as inputs of the prediction model in the training process, and the prediction model is trained iteratively for multiple times to obtain a second prediction model;

[0065] The end time of the first time period is the first time, and the duration of the first time period is the preset duration.

[0066] The device data for the first period is the device data at the first moment in the training set and the predicted duration before the first moment.

[0067] The second prediction model is a short-term prediction model, which is obtained by training the device data of the current moment and the preset time before the current moment, and is used to predict user control behaviors that have a large correlation with other control behaviors.

[0068] Optionally, within the first time period, device data at multiple time points are selected as the device data for the first time period.

[0069] Optionally, the number of moments selected within the first time period is determined by the length of a preset duration.

[0070] In a feasible implementation, when the preset time length corresponding to the second prediction model is 5 minutes, device data at 10 moments can be selected within 5 minutes as input of the second prediction model; when the preset time length corresponding to the second prediction model is 24 hours, device data at 72 moments can be selected within 24 hours as input of the second prediction model.

[0071] Optionally, in this embodiment, the multiple moments selected in the first time period may be evenly distributed in the first time period, or may be unevenly distributed.

[0072] For example, when the preset time length corresponding to the second prediction model is 24 hours, device data of 24 hours can be selected from 7 am to 9 am, device data of 12 hours can be selected from 9 am to 6 pm, device data of 24 hours can be selected from 6 pm to 12 pm, and device data of 12 hours can be selected from 12 pm to 7 am the next day. That is, device data can be selected more frequently during the time when people in the family are active, so that the user control behavior output by the second prediction model can better fit the user's behavior logic.

[0073] Specifically, the first prediction model and the second prediction model are both obtained through prediction model training.

[0074] The prediction model adopts a feedforward neural network model.

[0075] The training set includes a first training set and a second training set.

[0076] When the user's control behavior occurs, the device data at the current moment is collected and put into the first training set for training the first prediction model; the device data at the current moment and multiple moments within a preset time period before the current moment are collected and put into the second training set for training the second prediction model.

[0077] The prediction model is trained using the data in the first training set to obtain a first prediction model; the prediction model is trained using the data in the second training set to obtain a second prediction model.

[0078] like Figure 2 As shown in Figure 1, the input of the prediction model training is the pre-extracted device data. After several hidden layers, that is, linear models and nonlinear functions, the dimensions are finally adjusted to the output dimensions, and the user control behavior data can be output.

[0079] Among them, the number of hidden layers and the number of neurons in each hidden layer are adjusted according to the actual input application. The specific process can be expressed as:

[0080]

[0081] The training is controlled by adjusting the learning rate, batch size and number of times of the training set, and manual adjustments are made to high parameters.

[0082] For the trained prediction model, the device data of the smart home devices are collected in the same way to obtain a first test set and a second test set. The rationality of the trained first prediction model or the second prediction model is verified through the device data in the first test set or the second test set and the corresponding actual user management and control behavior data.

[0083] It should be noted that the first prediction model and the second prediction model predict the same user control behavior, so they are both trained using the same prediction model. When the first prediction model and the second prediction model are trained, the amount of data contained in each set of device data in the first training set and the second training set is different, that is, each set of device data in the second training set includes device data at multiple moments.

[0084] Therefore, when using a feedforward neural network to train a prediction model, it is necessary to manually adjust the high parameters during the training process to obtain a complementary first prediction model and a second prediction model, and to jointly achieve the prediction of the same user management behavior data through the first prediction model and the second prediction model.

[0085] The present invention predicts the same user's control behavior data through a first prediction model and a second prediction model. When training the prediction model, the logical relationship between the user's multiple control behaviors on the device within a certain period of time is taken into account, so that the output results of the trained prediction model can be more in line with the user's actual needs and realize more detailed and automated smart home scene rules.

[0086] In the smart home device control method of the present invention, before the step of sending the first value output by the first prediction model or the second value output by the second prediction model corresponding to each group of user behavior control parameters to the smart home device to be controlled, the method further includes:

[0087] Using the device data of the smart home device at the second moment as the input of the first prediction model, and obtaining a third value of the user behavior control parameter output by the first prediction model;

[0088] Using the device data of the smart home device in the second time period as the input of the second prediction model, obtaining a fourth value of the user behavior control parameter output by the second prediction model, the end time of the second time period is the second time, and the duration of the second time period is the preset duration;

[0089] After the first prediction model and the second prediction model are trained, they each output the value of the corresponding user control behavior parameter corresponding to the same actual user control behavior. On this basis, it is necessary to establish an association between each set of user control behavior parameters and a prediction model with a higher prediction accuracy in order to determine the value of the user control behavior parameter actually issued.

[0090] The device data at the second moment in the smart home device test sample is the device data in the first test set; the device data in the second time period in the smart home device test sample, that is, the device data at the second moment and the preset time before the second moment, is the device data in the second test set.

[0091] Specifically, for the same user control behavior, the corresponding device data in the first test set is input into the first prediction model to obtain the third value of each group of user behavior control parameters output by the first prediction model.

[0092] The device data in the corresponding second test set is input into the second prediction model to obtain the fourth value of each group of user behavior control parameters output by the second prediction model.

[0093] Compare the actual value of each group of user behavior control parameters corresponding to the device data at the second moment with the third value to obtain a first prediction accuracy rate of each group of user behavior control parameters;

[0094] Compare the actual value with the fourth value to obtain the second prediction accuracy of each group of user behavior control parameters;

[0095] The actual value of each group of user behavior control parameters corresponding to the user control behavior in the test set is compared with the third value to obtain the first prediction accuracy of each group of user behavior control parameters.

[0096] The first prediction accuracy rate indicates the similarity of the values ​​of each group of user behavior control parameters of the first prediction model.

[0097] Optionally, the first prediction accuracy rate may be determined based on the number of successful predictions by the first prediction model.

[0098] Due to different user control behaviors, the prediction results for the same device function can be output, that is, the same user behavior control parameter can be predicted; or for the same user control behavior, multiple device data collections are performed; therefore, for the same user behavior control parameter, there are multiple third values ​​and corresponding actual values.

[0099] All third values ​​output by the first prediction model for the same set of user behavior control parameters are compared with their corresponding actual values ​​in a feasible implementation manner, the number of successful predictions is recorded, and the value obtained by dividing the number of successful predictions by the number of all third values ​​is used as the first prediction accuracy rate.

[0100] The prediction success may be that the third value is the same as the actual value, or the third value is different from the actual value.

[0101] For example, when the user control behavior parameter is the switch parameter of the door lock, the value corresponding to the door lock opening is 1, and the value corresponding to the door lock closing is 0. In this case, the third value is the same as the actual value, indicating a successful prediction; the third value is different from the actual value, indicating a failed prediction.

[0102] For another example, when the user control behavior parameter is the temperature parameter of the air conditioner, the actual value of the air conditioner temperature is 26. The third value is different from the actual value, but the difference with the actual value is within the preset range, which may also indicate a successful prediction. For example, the preset range is ±1 of the actual value. When the third value is 25 or 27, it indicates a successful prediction. When the third value is 24, it indicates a failed prediction.

[0103] Optionally, the preset range is determined according to the type of the user control behavior parameter. For example, when the user control behavior parameter is the brightness parameter of the display, since the adjustable range of the brightness is larger than the adjustable range of the air conditioner temperature, usually 1-100, its preset range can also be larger than the preset range corresponding to the air conditioner temperature parameter, for example, ±3 of the actual value.

[0104] Optionally, a set of user behavior control parameters may be parameters corresponding to different functions of the same device, or may be parameters corresponding to different functions of different devices.

[0105] When the user behavior control parameters are parameters corresponding to different functions of the same device, the prediction success rate of each user behavior control parameter corresponding to each function of the same device under the first prediction model / second prediction model is determined, and the prediction success rates corresponding to all user behavior control parameters in the same device are averaged, and the average value is used as the first prediction success rate / second prediction success rate of the group of user behavior control parameters.

[0106] Furthermore, the actual value of each group of user behavior control parameters corresponding to the user control behavior in the test set is compared with the fourth value to obtain the second prediction accuracy of each group of user behavior control parameters.

[0107] The second prediction accuracy rate indicates the similarity of the values ​​of each group of user behavior control parameters of the second prediction model.

[0108] The method for obtaining the second prediction accuracy is substantially the same as the method for obtaining the first prediction accuracy, and therefore will not be described in detail.

[0109] Determine the larger value of the first prediction accuracy and the second prediction accuracy corresponding to each group of user behavior control parameters, and associate each group of user behavior control parameters with the first prediction model corresponding to the larger value, or with the second prediction model corresponding to the larger value.

[0110] For each group of user behavior control parameters, determine the larger value of the corresponding first prediction accuracy and the second prediction accuracy, and use the prediction model corresponding to the larger value as the prediction model corresponding to the group of user behavior control parameters, and associate the group of user behavior control parameters with the corresponding prediction model.

[0111] After all user behavior control parameters are associated with the prediction model with higher prediction accuracy, each group of user behavior control parameters is associated with the first prediction model corresponding to the larger value, or with the second prediction model corresponding to the larger value.

[0112] In a feasible implementation, the user behavior control parameters associated with the first prediction model are classified into a first type of application scenario, and the user behavior control parameters associated with the second prediction model are classified into a second type of application scenario.

[0113] When determining the value of the user behavior control parameter actually issued, the user behavior control parameter is searched in the first type of application scenario. If it exists, the first value output by the first prediction model is issued; if it does not exist, the second value output by the second prediction model is issued.

[0114] After the training of the first prediction model and the second prediction model is completed, the present invention determines the first prediction accuracy rate and the second prediction accuracy rate of each group of user behavior control parameters output by the first prediction model and the second prediction model, and associates the prediction model with higher prediction accuracy with the corresponding user behavior control parameters, so that when the user's control behavior parameters are predicted by the first prediction model and the second prediction model, the parameter value corresponding to the prediction model with higher accuracy rate is issued, so as to realize the complementarity of the first prediction model and the second prediction model and more accurately predict user needs.

[0115] In the smart home device control method of the present invention, the device data includes device status parameters, the scene mode of the device, time information and environmental information.

[0116] Specifically, the state parameters of a device describe the parameters corresponding to the functional state of the device itself at a certain moment.

[0117] There are multiple device status parameters in one device data, including different parameters corresponding to different functions of the same device, and different parameters corresponding to multiple devices.

[0118] For example, an item of equipment data includes the mode parameters and brightness parameters of a lamp, where the mode parameters of the lamp determine the type of lamp to be turned on, and the brightness parameters determine the brightness of each lamp. It also includes the mode parameters of the curtains, which determine the degree to which the curtains are opened. It can also include sensor parameters for temperature and humidity.

[0119] The device data describes the household devices corresponding to the smart home in various dimensions, reflecting the status of all smart home devices to be controlled at a certain moment.

[0120] The scene mode of the device includes the user's state at a certain moment, including resting, leaving home, sleeping, or watching a movie.

[0121] Optionally, the scene mode includes user location information, that is, the user's home information, and the user's status at a certain moment, such as whether the user is away from home, is determined through the location information.

[0122] The time information refers to the time when the device data of the smart home device to be controlled is collected.

[0123] Optionally, the time information of the device data is recorded in universal time.

[0124] The environmental information of the device data refers to the surrounding environment information corresponding to when the device data is collected.

[0125] Optionally, the environmental information may include surrounding temperature information, humidity information, weather information, air condition information, etc. corresponding to when the device data is collected.

[0126] Optionally, the environment information of the area where the device data is located can be obtained through the information of the LBS (Location-Based Service, a service based on the user's current location information) where the device is located.

[0127] In other feasible implementations, the temperature information, humidity information and air state information of the actual environment of the device may also be obtained through temperature and humidity sensors and air quality monitors.

[0128] In the smart home device control method of the present invention, the device state parameters include the device state parameters reported by the device and the device state parameters preset by the user, and the step of obtaining the device data of the smart home device to be controlled includes:

[0129] Obtain the device status parameters reported by the device and the scene mode of the device through the reporting event model in the object model of the smart home device to be controlled;

[0130] The state parameters preset by the user are obtained through the function model in the object model of the smart home device to be controlled.

[0131] Specifically, the device state parameter reported by the device indicates the parameter of the device state at the time of device data collection. The device state parameter reported by the device is reported by the device itself when the device data is collected, that is, uplink data.

[0132] The device status parameters preset by the user are device status parameters set by the user but not yet executed, and are downlink data.

[0133] For example, the user sets the air conditioner to turn on in 20 minutes. During the time from when the user sets the air conditioner to turn on, the device data collected includes the air conditioner parameters preset by the user.

[0134] It should be noted that the device state parameter preset by the user may also be the first value or the second value actually issued by the first prediction model or the second prediction model, that is, the device state parameter preset by the user predicted by the first prediction model or the second prediction model.

[0135] By using both the uplink data and the downlink data as device status parameters and as one of the inputs of the first prediction model / the second prediction model, the accuracy of the prediction model in predicting user control behavior parameters is further improved.

[0136] Furthermore, the device status parameters reported by the device and the scene mode of the device in the device data are obtained through the reporting event model in the object model of the smart home device to be controlled.

[0137] The physical model of the smart home device to be controlled is a digital representation of entities in the physical space (sensors and devices, etc.) in the cloud. It describes the physical attributes and functions of the smart device from several dimensions, such as attributes, services, and events, as well as the information that the smart device can provide to the platform / central control.

[0138] like Figure 3 As shown, the object model includes the attributes of the object model itself, the device reporting event model (i.e. Figure 3 events in the device control model (i.e. Figure 3 Among them, the device reporting event model is used to report device-related data, and the device control function model is used to issue user behavior control parameters.

[0139] When collecting device data into training and test sets, such as Figure 4 As shown, user control behavior triggers the collection of device data.

[0140] Specifically, when user control behavior occurs, the reporting event model in the object model automatically reports the device status parameters reported by the smart home device to be controlled and the scene mode of the device to the platform or the central control device. The platform stores them in the device dimension for subsequent use.

[0141] It should be noted that the reporting event model automatically reports the device status parameters and scene mode reported by the device at the current moment, which are stored as the current state of the device; and the device status parameters and scene mode reported by the device at multiple moments within a preset time before the current moment, which are stored as historical events.

[0142] Furthermore, the current status of the device is used as the device shadow so that the platform can query the real-time status of the device.

[0143] For example, when the smart home device to be controlled is a door magnetic device, when the door magnetic device is opened, a message with a parameter of doorStatus and a value of 1 will be reported; when the door magnetic device is closed, a message with a parameter of doorStatus and a value of 0 will be reported.

[0144] For the platform, when the door magnetic device reports a message through the physical model, the parameter changes will be stored in the historical events. That is, when the closing action occurs, the parameter changes within a certain period of time and the device status parameters generated by the door magnetic opening action will be stored in the historical events, and the specific time when each historical event occurs will be recorded.

[0145] Aggregating historical events by device and time can outline the behavioral characteristics of the device over a period of time. At the same time, the platform will update the current status of the device and use it as a device shadow to query the real-time status of the device.

[0146] On this basis, when user control behavior occurs, the functional model of device control in the object model will immediately issue the device function parameters corresponding to the user control behavior, and record the user's preset status parameters as the device shadow.

[0147] It should be noted that, in this embodiment, the function model of device control sends function parameters as the trigger node of user control behavior. When the function model of device control sends function parameters, the device data is collected once through the object model.

[0148] In one possible implementation, refer to Figure 4 When the functional model of device control sends the functional parameters, the platform is triggered to collect device data. At this time, the device status parameters reported by the device and the scene mode of the device are obtained through the reporting event model (i.e. Figure 4The user location / mode data in the device is collected, where the device status parameters reported by the device at the current moment are taken as part of the device shadow, the historical events obtained are taken as device historical events, and the environmental information is obtained through LBS; the user's preset status parameters are obtained through the functional model of device management and control, also as the device shadow.

[0149] The device shadow, device historical events, user location / pattern, and environmental information data are stored in the form of snapshots as short-term behavior snapshots, and the device shadow, user location / pattern, and environmental information data are stored in the form of snapshots as instantaneous behavior snapshots.

[0150] Put the instantaneous behavior snapshots into the first training set, and the short-term behavior snapshots into the second training set.

[0151] That is, the first training set only includes all device data when manual user control occurs; while the second training set includes device data at multiple time slices before the user control action.

[0152] The data in the second training set contains historical events of the device, which can outline the state transition of the device before control. For example, when a user opens the curtains after entering the door, the door magnet is in the closed state. This does not mean that the door magnet is closed and the curtain opening operation is related, but that the state transition of the door magnet from closed to open and then to closed is causally related to the curtain opening operation.

[0153] Furthermore, when collecting device data for predicting user behavior control parameters, the device data is collected once at regular intervals through the physical model and input into the first prediction model and the second prediction model to complete the prediction of user control behavior.

[0154] Optionally, the interval can be of uniform duration, such as collecting device data every five minutes; or it can be of non-uniform duration, for example, when the user is an adult, device data is actually collected at a lower frequency during his working hours and sleeping hours, and device data is collected at a higher interval during his resting time at home.

[0155] Optionally, the interval may also be determined according to the composition of the user's family members. For example, if there are students among the user's family members, the device data is collected once at a relatively high interval during the winter and summer vacations.

[0156] Furthermore, when collecting device data for predicting user behavior control parameters, since the prediction of user behavior control parameters essentially relies on the platform / central control device to perform control operations on smart home devices, for dangerous operations of some devices, such as kettles and door and window switches, etc., which involve control behaviors of user safety, the platform can be prohibited from automatically collecting the device data corresponding to the control behavior in the object model, so as to avoid issuing parameters corresponding to more dangerous control actions based on the prediction model trained based on the device data.

[0157] It should also be noted that since the management and control of smart home devices through prediction models is uncontrollable, users need to be given full right to know and right to choose before using the first prediction model and the second prediction model.

[0158] The present invention uses user behavior control parameters as both one of the inputs and one of the outputs of the prediction model. When the prediction model is used to predict user control behavior, the control behavior required by the user can be predicted more accurately; and the rapid collection of device data is achieved through the physical model.

[0159] In the smart home device control method of the present invention, before the step of inputting the device data of the smart home device to be controlled at the current moment into the first prediction model and obtaining the first value of at least one set of user behavior control parameters output by the first prediction model, the method further includes:

[0160] Map device data into preset ranges.

[0161] Since the physical model includes the values ​​of various functional parameters of the smart home devices to be controlled and the value range corresponding to the value of each functional parameter, collecting the uplink and downlink numbers of the smart home devices to be controlled as device data based on the physical model can conveniently pre-process the collected data.

[0162] Specifically, for uplink and downlink parameters, the types and values ​​of various parameters are inconsistent, and the preprocessing process needs to homogenize them and map them to the same range.

[0163] Optionally, the size of the same range is determined according to the type of the smart home device to be controlled. The value range of the parameter corresponding to each type of device is different, and the size of the same range changes accordingly.

[0164] Optionally, the parameters in the device data are normalized by min-max scaling to map them to the same preset range:

[0165]

[0166] Where x is the scaled data, v is the value of the original parameter, and v minis the minimum value of the original parameter, v max It is the maximum value of the original parameter. After processing the parameters in the device data in the above manner, the corresponding values ​​of each parameter can be mapped to the range of [0,1].

[0167] Optionally, Z-score standardization or normalization may be used to homogenize the device data by simply mapping it to the same preset range.

[0168] Optionally, when the value corresponding to a parameter in the device data is discrete data, the discrete value may be first converted into a continuous value, which is then subjected to a homogenization process, and then the corresponding value within the range is converted into a selectable discrete value.

[0169] The present invention can pre-process the collected data more conveniently by homogenizing the data collected by the physical model to facilitate the training of the prediction model, while improving the processing speed when the first prediction model and the second prediction model are used.

[0170] The smart home device control device provided by the present invention is described below. The smart home device control device described below and the smart home device control method described above can be referenced to each other.

[0171] like Figure 5 As shown, the device includes an acquisition module 501 and a delivery module 502, wherein:

[0172] The acquisition module 501 is used to input the device data of the smart home device to be controlled at the current moment into the first prediction model, and obtain the first value of at least one set of user behavior control parameters output by the first prediction model;

[0173] The smart home devices to be controlled are smart home devices that are controlled by smart home scene rules.

[0174] Optionally, the smart home devices to be controlled can be all devices connected to the home central control device or platform; or they can be devices pre-selected by the user among the devices connected to the home central control device or platform, so that the scene rules of the smart home can better meet user needs when controlling the devices.

[0175] Optionally, the smart home devices to be controlled do not include some devices that may cause danger to the user, such as kettles, door and window switches, etc., in order to ensure the safety of the user.

[0176] The device data of the smart home device to be controlled is data related to the smart home device to be controlled, and may include parameters corresponding to the functional status of the device itself, such as the switch status of a door lock, the brightness status of a light, or the temperature status of an air conditioner. It may also include relevant data when a smart home device to be controlled is used, such as the time when the device is used, the duration of use of the device, the external ambient temperature when the device is in use, and the external ambient temperature when the device is shut down.

[0177] Optionally, the device data may include data related to one smart home device to be controlled, or may include data related to multiple smart home devices to be controlled.

[0178] The first prediction model is pre-trained and is used to predict the user's control behavior on the smart home device based on the device data of the smart home device at a certain moment, so as to realize truly automated scene rules and reduce the user's configuration cost.

[0179] Specifically, the first prediction model outputs a first value of a user behavior control parameter, which is a control parameter of different functions of a device by the user. The first value is sent to the corresponding smart home device to be controlled, and the device is adjusted to the state predicted to be required by the user.

[0180] Optionally, each set of user behavior control parameters may include parameters corresponding to different functions of the same device, or may include parameters of different devices.

[0181] Optionally, the user behavior control parameters may include changes and invariances of function parameters corresponding to the functional state of the device. For example, the parameter corresponding to the temperature of the air conditioner at the current moment is 26°C. In the user behavior control parameters output by the first prediction model, the value of the corresponding air conditioner temperature parameter in the first value may be 25°C, that is, the temperature of the air conditioner changes after the first value is issued; or it may be 26°C, that is, the temperature of the air conditioner remains unchanged after the first value is issued.

[0182] By inputting the device data of the smart home device to be controlled at the current moment into the first prediction model, a first value of at least one set of user behavior control parameters output by the first prediction model can be obtained.

[0183] It is also used to input the device data of the current moment and the preset time before the current moment of the smart home device to be controlled into the second prediction model, and obtain the second value of the user behavior control parameter output by the second prediction model;

[0184] The second prediction model is pre-trained and is used to predict the user's control behavior on the smart home device based on the device data at the current moment and within a preset time period before the current moment.

[0185] Optionally, the preset duration may be determined according to device type, time, environment, etc.

[0186] Specifically, the control behavior of the user's device is correlated with time, that is, multiple different control behaviors performed by the user on different devices or different functions of the same device within a preset time period are correlated with each other.

[0187] For example, when a user is used to opening the curtains immediately after returning home, the user's control behavior on the device includes: controlling the door lock to open, then controlling the door lock to close, and then controlling the curtains to open. In other words, from the time the user returns home to the time the user opens the curtains, there is a causal relationship between the user's multiple control behaviors on the door lock and the curtains.

[0188] Therefore, by training the second prediction model, the user's control behavior is predicted based on the device data within a period of time. When multiple control behaviors of the user on the device are correlated, the user's control behavior can be predicted more accurately.

[0189] Optionally, this application does not limit the duration of device data.

[0190] For example, the second prediction model may be trained based on the device data at multiple times within 10 consecutive minutes; or the second prediction model may be further trained based on the device data at multiple times within 24 hours.

[0191] The preset time lengths are different, and the application scope of the second prediction model is correspondingly different.

[0192] By inputting the device data of the smart home devices to be controlled at the current moment and within a preset time period before the current moment into the second prediction model, a second value of at least one set of user behavior control parameters output by the second prediction model can be obtained.

[0193] The sending module 502 is used to send the first value output by the first prediction model or the second value output by the second prediction model corresponding to each group of user behavior control parameters to the smart home device to be controlled;

[0194] Wherein, each group of user behavior control parameters is pre-associated with the first prediction model, or each group of user behavior control parameters is pre-associated with the second prediction model.

[0195] Each set of user behavior control parameters is pre-associated with the first prediction model, or pre-associated with the second prediction model.

[0196] A control parameter for user behavior is obtained based on the device data at the current moment through the first prediction model; another control parameter for user behavior is obtained based on the device data at the current moment through the second prediction model.

[0197] Since the functions of some devices or different functions of the same device are correlated in the control parameters of user behavior, the output results of the second prediction model are more accurate; the functions of some devices or different functions of the same device are independent of each other, so the output results of the first prediction model are more accurate.

[0198] Therefore, it is necessary to pre-associate each group of user behavior management parameters with the first prediction model or the second prediction model with a higher prediction accuracy.

[0199] For each group of user behavior control parameters, the first value or the second value output by the first prediction model or the second prediction model associated therewith is selected and sent to the smart home device to be controlled.

[0200] For example, the device data at the current moment is input into the first prediction model to obtain the first values ​​corresponding to device A, device B, and device C; the device data at the current moment, i.e., the preset time length before the current moment, is input into the second prediction model to obtain the second values ​​corresponding to device A, device B, and device C.

[0201] According to the pre-established association relationship, it is determined that the prediction result of the first prediction model for device A is more accurate, and it is determined that the prediction result of the second prediction model for device B and device C is more accurate, then the first value corresponding to device A, the second value corresponding to device B, and the second value corresponding to device C are sent to the corresponding devices A, B, and C.

[0202] The present invention obtains complementary first prediction models and second prediction models through equipment data training of different durations, which are used to accurately predict independent user control behaviors or user control behaviors with associated relationships; among the user behavior control parameters input by the first prediction model and the second prediction model, according to the pre-established association between the user behavior control parameters and different prediction models, the first value or the second value output by the prediction model with higher prediction accuracy is selected for actual distribution, so as to realize the prediction of user control behavior, and the configuration method is simple, and the accuracy of user control behavior prediction is improved by considering the influence of time on user control behavior, which can meet the user's diversified control needs for equipment and realize a more automated smart home scene rule.

[0203] The smart home device control device of the present invention further includes a training module for:

[0204] The device data of the smart home device at multiple first moments are respectively used as inputs of the prediction model in multiple training processes, and the prediction model is iteratively trained multiple times to obtain a first prediction model;

[0205] The device data of the smart home device in the first time periods are used as inputs of the prediction model in the training process, and the prediction model is trained iteratively for multiple times to obtain a second prediction model;

[0206] The end time of the first time period is the first time, and the duration of the first time period is the preset duration.

[0207] In the smart home device control device of the present invention, the training module is also used for:

[0208] Using the device data of the smart home device at the second moment as the input of the first prediction model, and obtaining a third value of the user behavior control parameter output by the first prediction model;

[0209] Using the device data of the smart home device in the second time period as the input of the second prediction model, obtaining a fourth value of the user behavior control parameter output by the second prediction model, the end time of the second time period is the second time, and the duration of the second time period is the preset duration;

[0210] Compare the actual value of each group of user behavior control parameters corresponding to the device data at the second moment with the third value to obtain a first prediction accuracy rate of each group of user behavior control parameters;

[0211] Compare the actual value with the fourth value to obtain the second prediction accuracy of each group of user behavior control parameters;

[0212] Determine the larger value of the first prediction accuracy and the second prediction accuracy corresponding to each group of user behavior control parameters, and associate each group of user behavior control parameters with the first prediction model corresponding to the larger value, or with the second prediction model corresponding to the larger value.

[0213] The smart home device control device of the present invention further includes a collection module for:

[0214] Obtain the device status parameters reported by the device and the scene mode of the device through the reporting event model in the object model of the smart home device to be controlled;

[0215] The state parameters preset by the user are obtained through the function model in the object model of the smart home device to be controlled.

[0216] The smart home device control device of the present invention further includes a preprocessing module for mapping device data into the same range.

[0217] Figure 6 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 6As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620 and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630, input the device data of the smart home device to be controlled at the current moment into the first prediction model, and obtain the first value of at least one group of user behavior control parameters output by the first prediction model; input the device data of the smart home device to be controlled at the current moment and the preset time before the current moment into the second prediction model, and obtain the second value of the user behavior control parameter output by the second prediction model; according to the first prediction model or the second prediction model corresponding to any group of user behavior control parameters, the first value of any group of user behavior control parameters output by the first prediction model or the second value of any group of user behavior control parameters output by the second prediction model is sent to the smart home device to be controlled; wherein each group of user behavior control parameters is pre-associated with the first prediction model or pre-associated with the second prediction model.

[0218] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0219] On the other hand, the present invention also provides a computer program product, including a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the smart home device control method provided by the above methods, the method including: inputting the device data of the smart home device to be controlled at the current moment into a first prediction model, and obtaining a first value of at least one set of user behavior control parameters output by the first prediction model; inputting the device data of the smart home device to be controlled at the current moment and a preset time before the current moment into a second prediction model, and obtaining a second value of the user behavior control parameter output by the second prediction model; according to the first prediction model or the second prediction model corresponding to any group of user behavior control parameters, the first value of any group of user behavior control parameters output by the first prediction model or the second value of any group of user behavior control parameters output by the second prediction model is sent to the smart home device to be controlled; wherein each group of user behavior control parameters is pre-associated with the first prediction model, or is pre-associated with the second prediction model.

[0220] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the smart home device control method provided by the above-mentioned methods, the method comprising: inputting the device data of the smart home device to be controlled at the current moment into a first prediction model, and obtaining a first value of at least one set of user behavior control parameters output by the first prediction model; inputting the device data of the smart home device to be controlled at the current moment and a preset time before the current moment into a second prediction model, and obtaining a second value of the user behavior control parameter output by the second prediction model; according to the first prediction model or the second prediction model corresponding to any set of user behavior control parameters, the first value of any set of user behavior control parameters output by the first prediction model or the second value of any set of user behavior control parameters output by the second prediction model is sent to the smart home device to be controlled; wherein each set of user behavior control parameters is pre-associated with the first prediction model, or is pre-associated with the second prediction model.

[0221] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0222] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on this understanding, the above technical solution, in essence or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for a computer device to execute the methods described in each embodiment or some parts of the embodiments.

[0223] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart home device control method, characterized in that: include: Inputting the device data of the smart home device to be controlled at the current moment into the first prediction model, and obtaining a first value of at least one set of user behavior control parameters output by the first prediction model; Inputting device data of the current moment of the smart home device to be controlled and a preset time before the current moment into a second prediction model, and obtaining a second value of the user behavior control parameter output by the second prediction model; Sending the first value output by the first prediction model or the second value output by the second prediction model corresponding to each group of the user behavior control parameters to the smart home device to be controlled; Among them, each group of user behavior control parameters is pre-associated with the first prediction model, or pre-associated with the second prediction model; the first value or the second value is used to control the smart home device to be controlled.

2. The smart home device control method according to claim 1, characterized in that: Before the step of inputting the current device data of the smart home device to be controlled into the first prediction model to obtain the first value of at least one set of user behavior control parameters output by the first prediction model, the method further includes: The device data of the smart home device at multiple first moments are respectively used as inputs of the prediction model in multiple training processes, and the prediction model is iteratively trained multiple times to obtain a first prediction model; The device data of the smart home device in the first time periods are used as inputs of the prediction model in the training process, and the prediction model is trained iteratively for multiple times to obtain a second prediction model; The end time of the first time period is the first time, and the duration of the first time period is a preset duration.

3. The smart home device control method according to claim 2, characterized in that: Before the step of sending the first value output by the first prediction model or the second value output by the second prediction model corresponding to each group of the user behavior control parameters to the smart home device to be controlled, the step further includes: Using device data of the smart home device at the second moment as input of the first prediction model, and obtaining a third value of the user behavior control parameter output by the first prediction model; Using the device data of the smart home device in the second time period as the input of the second prediction model, obtaining the fourth value of the user behavior control parameter output by the second prediction model, the end time of the second time period is the second time, and the duration of the second time period is a preset duration; Compare the actual value of each group of user behavior control parameters corresponding to the device data at the second moment with the third value to obtain a first prediction accuracy rate of each group of user behavior control parameters; Compare the actual value with the fourth value to obtain a second prediction accuracy rate for each group of the user behavior control parameters; Determine the larger value of the first prediction accuracy and the second prediction accuracy corresponding to each group of the user behavior control parameters, and associate each group of the user behavior control parameters with the first prediction model corresponding to the larger value, or with the second prediction model corresponding to the larger value.

4. The smart home device control method according to any one of claims 1 to 3, characterized in that: The device data includes device status parameters, the scene mode of the device, time information and environment information.

5. The smart home device control method according to claim 4, characterized in that: The device status parameters include device status parameters reported by the device and device status parameters preset by the user, and the step of obtaining device data of the smart home device to be controlled includes: Acquire the device status parameters reported by the device and the scene mode of the device through the reporting event model in the object model of the smart home device to be controlled; The state parameters preset by the user are obtained through the function model in the object model of the smart home device to be controlled.

6. The smart home device control method according to any one of claims 1 to 3, characterized in that: Before the step of inputting the device data of the smart home device to be controlled at the current moment into the first prediction model and obtaining the first value of at least one set of user behavior control parameters output by the first prediction model, the method further includes: Map the device data into the same range.

7. A smart home equipment control device, characterized in that: include: An acquisition module, configured to input device data of the smart home device to be controlled at a current moment into a first prediction model, and acquire a first value of at least one set of user behavior control parameters output by the first prediction model; Also used for inputting device data of the current moment of the smart home device to be controlled and a preset time before the current moment into the second prediction model, and obtaining the second value of the user behavior control parameter output by the second prediction model; A sending module sends the first value output by the first prediction model or the second value output by the second prediction model corresponding to each group of the user behavior control parameters to the smart home device to be controlled; Among them, each group of user behavior control parameters is pre-associated with the first prediction model, or each group of user behavior control parameters is pre-associated with the second prediction model.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the smart home device control method as described in any one of claims 1 to 7 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the smart home device control method according to any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the smart home device control method according to any one of claims 1 to 7 is implemented.

Images