An intelligent home management method and system based on artificial intelligence

Abstract

The application relates to an intelligent home management method and system based on artificial intelligence, which comprises an intelligent home platform and multiple intelligent home devices. The intelligent home platform intelligently judges the working modes of the intelligent home devices based on an artificial intelligence model in a multi-user scenario, so as to meet the requirements of different users as much as possible.

Description

[0001] The application belongs to the field of smart home, and particularly relates to a smart home management method and system based on artificial intelligence.

[0002] The smart home is a platform based on a residence, and uses integrated wiring technology, network communication technology, security protection technology, automatic control technology, audio and video technology to integrate facilities related to home life, construct a high-efficiency residence facility and home affairs management system, and improve home safety, convenience, comfort, artistry, and realize an environmentally friendly and energy-saving living environment. The existing smart home system is not limited to the intelligentization of a single home device, but uses a unified smart home platform to control the working modes of various home devices.

[0003] In recent years, with the rise of artificial intelligence, the platform can use an artificial intelligence model to determine the next working mode of each device based on the state of each device in the current residence and other external states and user states. However, the existing artificial intelligence model usually only considers the device working mode suitable for a single user. When there are multiple users in the residence, the needs of different users may not be the same, and even may conflict. The existing artificial intelligence model has not considered how to control the device working mode when the needs of different users are different under the condition of multiple users.

[0004] In order to solve the above problems in the prior art, the application provides a smart home management method and system based on artificial intelligence.

[0005] The technical scheme adopted by the application is as follows:

[0006] A smart home management method based on artificial intelligence, the method comprising:

[0007] Step 1: collecting historical data of a smart home management system in a residence; the historical data includes working modes of various smart home devices, external data and user states;

[0008] Step 2: for any smart home device D, constructing a corresponding working mode prediction model M D , constructing a training sample set based on the historical data, and training the model M D ; wherein the model M D inputs the working modes of all smart home devices except the device D, external data and a single user code, and predicts the working mode of the device D corresponding to the user code;

[0009] ​​​Step 3: When there are multiple users in the house, the smart home platform identifies each user and determines the influence coefficient of the device D on each user in the current house based on the influence range of the device D; the influence coefficient is used to determine the influence of the device D on the location of the user;

[0010] Step 4: For any user with a non-zero influence coefficient, the model M D Calculate the probability of all working modes of the device D, and all the probabilities obtained constitute the probability matrix of the device D;

[0011] Step 5: If the working mode of the device D includes a working parameter of a non-continuous variable, determine the working parameter according to the weighting of the probability matrix by the influence coefficient;

[0012] Step 6: If the working mode of the device D also includes a working parameter of a continuous variable, determine the model M D The predicted working parameter for each user is processed as a normal distribution curve with the predicted value as the mean.

[0013] Step 7: Determine the working parameter of the continuous variable by weighting and superimposing the normal distribution curves according to the influence coefficient.

[0014] Further, the step 4 specifically includes:

[0015] Suppose there are n users User1, User2, …, User n , and the device D has m working modes, then for any user User i , obtain the model M D Output the probability of all working modes of the device D;

[0016] Let P ij be the probability of predicting the jth working mode of the device D for User i , then the probability matrix {P ij} of the device D is obtained, 1≤i≤n, 1≤j≤m.

[0017] Further, the step 5 includes:

[0018] According to the weighting of the probability matrix by the influence coefficient, the weighted probability W j of the jth working mode is calculated, that is:

[0019]

[0020] Where A i is the user User ian influence coefficient of each of the n normal distribution curves; determining a working mode ModeMax with the greatest weighted probability, wherein all the working parameters of the non-continuous variables included in the working mode ModeMax are determined as the working parameters of the device D.

[0021] Further, the step 7 comprises:

[0022] For each user User i multiplying the corresponding normal distribution curve by its influence coefficient A i to obtain a weighted normal distribution curve, adding all the n weighted normal distribution curves to obtain a final probability curve, and taking the point with the greatest vertical coordinate on the probability curve as the parameter value with the greatest probability, and taking the parameter value as the working parameter of the determined device D.

[0023] Further, the external data comprises weather data.

[0024] Further, the user state comprises a user code of a user in a current residence and a location of the user.

[0025] Further, the influence range of the device D is determined according to the device type, the residence layout and a predetermined influence range division method.

[0026] Further, all the working parameters of the device D are determined through steps 6-7, and the combination of the working parameters constitutes the predicted working mode of the device D.

[0027] Further, the smart home device comprises a television, an air conditioner, a lighting lamp, a curtain and a monitoring device.

[0028] The application further provides a smart home management system based on artificial intelligence, which comprises a smart home platform and a plurality of smart home devices, and the smart home platform can manage the smart home devices according to the above method.

[0029] The application has the beneficial effect that in a multi-user scenario, the working mode of each device in the smart home system is intelligently determined to meet the needs of different users as much as possible, thereby reducing the possibility of multi-user demand conflicts. BRIEF DESCRIPTION OF DRAWINGS

[0030] The drawings described herein are used to provide further understanding of the application, form a part of the application and do not constitute improper limitations on the application, and in the drawings:

[0031] Figure 1 is a basic structure diagram of the smart home management system of the application. DETAILED DESCRIPTION

[0032] The present application will be described in detail below with reference to the accompanying drawings and specific embodiments, wherein the schematic embodiments and descriptions are only used to explain the present application, but not to limit the present application.

[0033] Referring to the accompanying drawings Figure 1 which shows the basic structure diagram of the smart home management system of the present application, the system includes a smart home platform and multiple smart home devices, the smart home platform and smart home devices are arranged in the user's residence, and communicate with each other through the internal network of the home.

[0034] The smart home device can be various intelligent devices in the residence, and support unified management of the smart home system. For example, intelligent TV, air conditioner, lighting fixture, curtain, monitoring device, etc. These devices can have different working modes. Taking the air conditioner as an example, the air conditioner can have cooling mode, heating mode, dehumidification mode, sleep mode, shutdown mode, etc. And for the cooling mode of the air conditioner, each cooling temperature set can also be considered as a finer working mode, for example, in the cooling mode, the air conditioner can be set to 22 degrees or 23 degrees, which can also be considered as two working modes. The working mode of the smart home device can be actively adjusted according to the current residence state and user state, which embodies its intelligent feature.

[0035] The smart home platform is a platform for unified management of various smart home devices, which can be a server or any other type of control device inside the home, and can also be a remote server with the authorization and consent of the user. In the present application, an artificial intelligence model is run on the smart home platform, which receives the current residence state and user state to determine the working mode of each smart home device.

[0036] Based on the above smart home management system, the specific steps of the smart home management method of the present application will be described in detail below.

[0037] Step 1: Collecting the historical data of the smart home management system in the residence.

[0038] Specifically, for a residence installed with the smart home management system of the present application, the system running data, external data and user state are collected within a period of time after installation to constitute the historical data.

[0039] The system running data includes the working mode of each smart home device in the system. In specific implementation, the device manufacturer can provide a unique code for any possible working mode of the device, so the smart home platform only needs to receive and record the unique code of the working mode executed by each device and the corresponding time.

[0040] The external data may include, for example, meteorological data. This meteorological data can be queried and downloaded from the internet by the smart home platform. The external data may also include other types of relevant data; this invention does not limit this.

[0041] The user status includes the current users within the residence and their locations. To obtain user status, multiple monitoring devices can be installed within the residence. The smart home platform acquires monitoring images from each device and uses facial recognition to identify each user within the residence and their location. The smart home platform can encode each unique face as a user code, storing the user code, user location, and current time as corresponding user status data. It should be noted that the smart home platform's facial recognition and storage of user status should only be performed with the user's authorization and consent.

[0042] By collecting historical data over a period of time, the smart home platform can obtain multiple historical data points. Each historical data point includes the time, the working mode of each device at that time, the external data at that time, and the user status at that time.

[0043] Step 2: For any smart home device D, construct the corresponding working mode prediction model M. D A training sample set is constructed based on the historical data, and the model M is then tested. D Conduct training.

[0044] Specifically, after collecting sufficient historical data related to device D, the smart home platform constructs a corresponding operating mode prediction model M based on a pre-provided artificial intelligence model. D The model M D Given the operating modes of all current smart home devices (excluding device D), external data, and a single user code, predict the operating mode of device D corresponding to that user code. In fact, model M... D It outputs the probabilities of all working modes corresponding to the user's encoding, and takes the working mode with the highest probability as the prediction result. In short, this model M... D The operating mode of device D can be determined for any user based on the current system state and external data. As a single-user model, its construction and training are relatively simple, which can improve overall efficiency.

[0045] To train the model M D, the historical data related to device D needs to be collected as training samples. As described above, each piece of historical data includes time, working mode of each device at the time, external data at the time, and user state at the time; based on this, if the working mode data of device D is included in a piece of historical data, the working mode data DevicesData of each device other than device D, the working mode ModeD of device D, and the external data OtherData are obtained from the piece of historical data, and then based on the user state in the piece of historical data, the user code UserID of the user most relevant to device D at the time is obtained. Thus, a training sample <DevicesData, OtherData, UserID, ModeD> is obtained, wherein ModeD is used as a sample label.

[0046] Among them, in order to determine the user most relevant to device D at the time, the user closest to device D can be determined as the most relevant user based on the positions of each user in the user state. If there are multiple users closest to device D, each of them can be regarded as the most relevant user, and a corresponding training sample is constructed for each of the most relevant users, thereby obtaining multiple corresponding training samples. The distance here can only be a rough estimate and does not need to be too accurate, for example, users in the same room as device D can all be regarded as the closest users.

[0047] After obtaining a sufficient set of training samples, the smart home platform trains the model M D to obtain a trained working mode prediction model. Based on the same method, for each smart home device, a corresponding working mode prediction model can be trained.

[0048] Step 3: When there are multiple users in the residence, the smart home platform identifies each user and determines the influence coefficient of device D on each user in the current residence based on the influence range of device D.

[0049] The main problem solved by the present application is the management of smart home in a multi-user scenario. If it is a single-user scenario, the model M DThe working mode of the device D is determined, which is not described in detail here. In a multi-user scenario, the smart home platform first performs face recognition on each user to determine the user code of each user. Then, the smart home platform determines the influence range of the device D according to the location of the device D. For example, if the device D is a television, its influence range is mainly in the room where the device D is located, and has little influence on adjacent rooms and no influence on non-adjacent rooms (i.e. not in the influence range of the device D). If the device D is an air conditioner, its main influence range is in the room where the device D is located, and has some influence on adjacent rooms and the influence on more distant rooms decays with distance. A more fine-grained range division can also be used, for example, the main influence range of the television can be limited to a fan-shaped area in front of it, and the influence on other areas of the room is smaller than that of the fan-shaped area. In summary, the smart home platform can determine the influence range of the device D according to the device type, the layout of the residence and the predetermined influence range division method.

[0050] Then, the smart home platform can determine the influence coefficient of the device D on each user in the current residence according to the influence range of the device D. The influence coefficient is used to determine the influence of the device D on the location of the user. For example, if a user is not in the influence range of the device D, the influence coefficient of the device D on the user is 0; if a user is in the main influence range of the device D, the influence coefficient of the device D on the user is 1; if a user is in a smaller influence range of the device D, an influence coefficient between 0 and 1 can be determined according to the influence size. The influence coefficient can be calculated according to the management strategy predetermined by the device manufacturer, or the user can set the calculation strategy of the influence coefficient.

[0051] Step 4: for any user with a non-zero influence coefficient, according to the model M D calculate the probability of all working modes of the device D.

[0052] As described above, the model M D actually outputs the probability of all working modes of the device D for a user, and then takes the working mode with the maximum probability as the prediction result. Then, assuming that there are n users User1, User2, …, User n with a non-zero influence coefficient, and the device D has m working modes, for any user User i , the model M D can output the probability of all working modes of the device D (a total of m probabilities, and the sum is 1), thereby obtaining a total of mn probability values.

[0053] Let P ij be the probability of predicting the jth working mode of the device D for User i , then the probability matrix {P ij}(1≤i≤n, 1≤j≤m).

[0054] Step 5: If the working mode of the device D includes the working parameter of the non-continuous variable, the working parameter is determined according to the weighting of the probability matrix by the influence coefficient.

[0055] The working mode described in the present application is a combination of multiple working parameters. Taking an air conditioner as an example, cooling 22 degrees, cooling 23 degrees, heating 26 degrees, and heating 27 degrees are four different working modes. They include two working parameters. The first working parameter is to select one of the cooling mode, the heating mode, the dehumidification mode, the sleep mode, and the like. The second working parameter is to select a specific temperature. The second working parameter temperature is a continuous variable that can be continuously adjusted between 16-30 degrees, i.e., the second working parameter is a continuous variable. The first working parameter is not a continuous variable, so it is a working parameter of a non-continuous variable.

[0056] First, the working parameter of the non-continuous variable is determined, and the probability matrix is weighted according to the influence coefficient, i.e., the weighted probability Wj of the jth working mode is calculated. j i.e.,

[0057]

[0058] where A i is the influence coefficient of the user User i . Thus, the working mode ModeMax with the maximum weighted probability can be determined, and all the working parameters of the non-continuous variables included in the working mode ModeMax are determined as the working parameters of the device D.

[0059] Step 6: If the working mode of the device D also includes the working parameter of the continuous variable, the model M D is used to predict the working parameter for each user.

[0060] As described above, the model M D actually outputs the probabilities of all the working modes of the device D for a user, and then the working mode with the maximum probability is taken as the prediction result. The working mode with the maximum probability also includes the working parameter of the continuous variable, which is the predicted value of the working parameter of the model M D for the user. For example, for an air conditioner, the predicted working mode for a user is cooling 22 degrees, and 22 is the predicted value for the user, which is a working parameter of a continuous variable. A normal distribution curve is established with the predicted value (such as 22) as the mean value (i.e., the mean value μ of the normal distribution is 22) and a predetermined variance.

[0061] In this way, n normal distribution curves can be established for the n users. The probability of the temperature selected by each user is described by the normal distribution curve.

[0062] Step 7: determining the working parameters of the continuous variable according to the weighted superposition of the normal distribution curves by the influence coefficient.

[0063] Specifically, for the user User i The corresponding normal distribution curve is multiplied by the influence coefficient A i to obtain a weighted normal distribution curve. All n weighted normal distribution curves are added to obtain a final probability curve. The point with the maximum ordinate on the probability curve is taken, and the corresponding abscissa is the parameter value with the maximum probability. The parameter value is taken as the working parameter of the device D.

[0064] Through steps 6-7, all working parameters of the device D can be determined. The combination of the working parameters constitutes a working mode of the device D. The working mode is taken as the working mode of the device D predicted by the present application in a multi-user scenario, so that the smart home platform can notify the device D to automatically switch to the working mode or switch to the working mode after seeking the consent of the user.

[0065] The above steps detail the intelligent home management method based on artificial intelligence. The method can intelligently determine the working mode of each device in the intelligent home system in a multi-user scenario to meet the needs of different users as much as possible.

[0066] The above only describes the preferred embodiments of the present application, and equivalent changes or modifications made according to the structure, features and principles described in the scope of the present application are included in the scope of the present application.

Claims

1. A smart home management method based on artificial intelligence, characterized in that, The method includes: Step 1: Collect historical data from the smart home management system within the residence; the historical data includes the working modes of each smart home device, external data, and user status; Step 2: For any smart home device D, construct the corresponding working mode prediction model M. D A training sample set is constructed based on the historical data, and the model M is then tested. D Training is performed; wherein, the model M D Given the operating modes, external data, and individual user codes of all smart home devices except device D, predict the operating mode of device D corresponding to the user code. Step 3: When there are multiple users in the residence, the smart home platform identifies each user and determines the influence coefficient of device D on each user in the current residence based on the influence range of device D; the influence coefficient is used to determine the influence of device D on the user's location; Step 4: For any user whose influence coefficient is not 0, according to model M D Calculate the probability of all operating modes of the device D, and the resulting probabilities constitute the probability matrix of the device D; Step 5: If the operating mode of device D includes non-continuous variable operating parameters, then determine the operating parameters based on the weighted sum of the probability matrix according to the influence coefficients; Step 6: If the operating mode of device D also includes continuous variable operating parameters, then model M... D The working parameter predicted for each user is processed into a normal distribution curve with the predicted value as the mean; Step 7: Determine the working parameters of the continuous variable by weighting the normal distribution curves according to the influence coefficients; Specifically, step 7 includes: For User i Multiply its corresponding normal distribution curve by its influence coefficient A i We obtain the weighted normal distribution curve, and sum all n weighted normal distribution curves to obtain a final probability curve. We take the point with the largest vertical coordinate on the probability curve, and the corresponding horizontal coordinate is the parameter value with the highest probability. We use this parameter value as the working parameter of the determined equipment D.

2. The method according to claim 1, characterized in that, Step 4 specifically includes: Suppose there are n users User1, User2, ..., User... n And given that device D has m working modes, then for any user User i Obtain model M D The output is the probability of all operating modes of device D; Let P ij For User i Predicting the probability of the j-th operating mode of device D yields the probability matrix {P} of device D. ij }, 1≤i≤n, 1≤j≤m.

3. The method according to claim 2, characterized in that, Step 5 includes: The weighted probability W of the j-th working mode is calculated by weighting the probability matrix according to the influence coefficients. j ,Right now: Where A i User i The influence coefficient; determine the operating mode ModeMax with the highest weighted probability, and use the operating parameters of all non-continuous variables included in this operating mode ModeMax as the operating parameters of the determined device D.

4. The method according to any one of claims 1-3, characterized in that, The external data includes meteorological data.

5. The method according to any one of claims 1-3, characterized in that, The user status includes the user code within the current residence and the user's location.

6. The method according to any one of claims 1-3, characterized in that, The scope of influence of equipment D is determined based on the equipment type, residential layout, and the predetermined scope of influence division method.

7. The method according to claim 1, characterized in that, Steps 6-7 determine all operating parameters of device D, and the combination of these operating parameters constitutes the predicted operating mode of device D.

8. The method according to claim 1, characterized in that, The smart home devices include televisions, air conditioners, lighting fixtures, curtains, and monitoring equipment.

9. An artificial intelligence-based smart home management system, characterized in that, The system includes a smart home platform and multiple smart home devices, wherein the smart home platform manages the smart home devices according to any one of claims 1-8.

Images