Equipment management method and device, electronic equipment and readable medium

Abstract

The embodiment of the invention provides an equipment management method and device, electronic equipment and a readable medium. According to the equipment management method provided by the embodiment of the invention, health state data and space environment data of a user are acquired; inputting the health state data and the space environment data into a preset behavior prediction model, and obtaining a future equipment state predicted by the behavior prediction model; and adjusting the running state of at least one device in the space based on the future device state. Therefore, the equipment state preferred by the user can be analyzed under the condition of considering the health state of the user and the space environment, and the adjustment mode of the running state of at least one piece of equipment in the space is determined based on the equipment state preferred by the user, so that the equipment in the space can be adjusted without complex operation of the user, and the user experience is improved. And the adjustment mode is simple.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment management, and in particular to an equipment management method, an equipment management device, an electronic device and a computer-readable medium. Background Art

[0002] With the popularity of smart home devices, more and more families are controlling and managing devices through networking. However, existing smart home device management systems generally rely on preset rules or manual adjustments by users, making device adjustments difficult. Summary of the invention

[0003] The embodiments of the present invention provide a device management method, apparatus, electronic device and computer-readable storage medium to solve the problem of difficulty in adjusting household devices.

[0004] An embodiment of the present invention discloses a device management method, the method comprising:

[0005] Obtain the user's health status data and spatial environment data;

[0006] Inputting the health status data and the spatial environment data into a preset behavior prediction model to obtain a future device state predicted by the behavior prediction model;

[0007] Based on the future device state, the operating state of at least one device in the space is adjusted.

[0008] Optionally, the step of obtaining the user's health status data and spatial environment data includes:

[0009] Obtain the user's health status data through the camera and / or sensors of smart wearable devices;

[0010] The spatial environment data is obtained through at least one of a temperature and humidity sensor, a light sensor, and an external environment data source.

[0011] Optionally, the step of inputting the health status data and the spatial environment data into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model includes:

[0012] Inputting the health status data and the spatial environment data into a preset user portrait prediction model to obtain the user's environmental preference information output by the user portrait prediction model;

[0013] The environmental preference information is input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0014] Optionally, the step of inputting the environmental preference information into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model includes:

[0015] The environmental preference information and the spatial environment data are input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0016] Optionally, the step of adjusting the operating state of at least one device in the space based on the future device state includes:

[0017] Inputting the future device state into a preset reinforcement learning model to obtain a control method of at least one device in the space output by the reinforcement model;

[0018] The control method of the device is adopted to adjust the operating state of at least one device in the space.

[0019] Optionally, the method further comprises:

[0020] determining energy consumption costs based on an operating status of at least one device in the space;

[0021] Determining comfort information based on the operating status of at least one device in the space and the health status data of the user;

[0022] Determining a reward function of the reinforcement learning model based on the energy consumption cost and the comfort information;

[0023] Based on the reward function, the reinforcement learning model is adjusted.

[0024] Optionally, the method further comprises:

[0025] Receive user feedback;

[0026] The reinforcement learning model is adjusted based on the feedback information of the user.

[0027] An embodiment of the present invention further provides a device management apparatus, the device comprising:

[0028] A data acquisition module is used to obtain the user's health status data and spatial environment data;

[0029] An equipment state prediction module, used for inputting the health state data and the spatial environment data into a preset behavior prediction model to obtain a future equipment state predicted by the behavior prediction model;

[0030] The device adjustment module is used to adjust the operating state of at least one device in the space based on the future device state.

[0031] Optionally, the data acquisition module includes:

[0032] The health status acquisition submodule is used to obtain the user's health status data through the camera and / or the sensor of the smart wearable device;

[0033] The space environment acquisition submodule is used to acquire space environment data through at least one of a temperature and humidity sensor, a light sensor, and an external environment data source.

[0034] Optionally, the device state prediction module includes:

[0035] An environmental preference acquisition submodule, used to input the health status data and the spatial environment data into a preset user portrait prediction model, and obtain the user's environmental preference information output by the user portrait prediction model;

[0036] The future device state acquisition submodule is used to input the environmental preference information into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0037] Optionally, the future device status acquisition submodule includes:

[0038] The future device state prediction unit is used to input the environmental preference information and the spatial environment data into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0039] Optionally, the device adjustment module includes:

[0040] A control mode acquisition submodule, used to input the future device state into a preset reinforcement learning model to obtain a control mode of at least one device in the space output by the reinforcement model;

[0041] The operating status adjustment submodule is used to adjust the operating status of at least one device in the space by adopting the control method of the device.

[0042] Optionally, the device further comprises:

[0043] An energy consumption cost determination module, configured to determine the energy consumption cost based on the operating status of at least one device in the space;

[0044] a comfort level determination module, configured to determine comfort level information based on the operating status of at least one device in the space and the health status data of the user;

[0045] A reward function determination module, used to determine a reward function of the reinforcement learning model based on the energy consumption cost and the comfort information;

[0046] A model adjustment module is used to adjust the reinforcement learning model based on the reward function.

[0047] Optionally, the device further comprises:

[0048] A feedback information acquisition module, used to receive user feedback information;

[0049] A feedback information model adjustment module is used to adjust the reinforcement learning model based on the feedback information of the user.

[0050] The embodiment of the present invention further discloses an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0051] The memory is used to store computer programs;

[0052] The processor is used to implement the method described in the embodiment of the present invention when executing the program stored in the memory.

[0053] The embodiment of the present invention further discloses one or more computer-readable media on which instructions are stored. When executed by one or more processors, the processors are enabled to execute the method described in the embodiment of the present invention.

[0054] The embodiments of the present invention include the following advantages:

[0055] Through the device management method provided by the embodiment of the present invention, the user's health status data and space environment data are obtained; the health status data and the space environment data are input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model; based on the future device state, the operating state of at least one device in the space is adjusted. Thus, the device state preferred by the user can be analyzed while considering the user's health status and space environment, and the adjustment method of the operating state of at least one device in the space can be determined based on the device state preferred by the user, so that the adjustment of the device in the space can be completed without complicated operations by the user, and the adjustment method is simple. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a flow chart of the steps of a device management method provided in an embodiment of the present invention;

[0057] Figure 2 is a flow chart of the steps of a device management method provided in an embodiment of the present invention;

[0058] Figure 3 is a flow chart of a device management method provided in an embodiment of the present invention;

[0059] Figure 4 is a structural block diagram of a device management device provided in an embodiment of the present invention;

[0060] Figure 5 is a block diagram of an electronic device provided in an embodiment of the present invention;

[0061] Figure 6 is a schematic diagram of a computer-readable medium provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0063] Reference Figure 1 , shows a flow chart of the steps of a device management method provided in an embodiment of the present invention, which may specifically include the following steps:

[0064] Step 101, obtaining the user's health status data and spatial environment data;

[0065] In an embodiment of the present invention, the devices in the space may be managed while taking the health status of the user into consideration.

[0066] The space may refer to different areas or rooms in a home or building. These spaces may be bedrooms, living rooms, kitchens, bathrooms, offices, garages, etc. At least one device may be provided in the space, and the device may be used to adjust the environment in the space so that the space is in an environment that the user feels comfortable.

[0067] The user's health status data may refer to information related to the user's health, which may be collected through various sensors and devices. The user's health status data may help the smart home system understand the user's health status and intelligently adjust device settings and energy usage based on this information.

[0068] Generally speaking, a user's health status data may include physiological data such as heart rate, blood pressure, body temperature, blood oxygen saturation, and blood sugar level; activity data such as number of steps, amount of exercise, sleep quality, and activity trajectory; and psychological data such as stress level and emotional state.

[0069] Step 102, inputting the health status data and the spatial environment data into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model;

[0070] In an embodiment of the present invention, a behavior prediction model can be pre-set. The behavior prediction model can be used to analyze the spatial environment preferred by the user in different time periods based on the input health status data and spatial environment data, and predict the environment preferred by the user in the future, so as to obtain the future device status.

[0071] Step 103: Based on the future device state, adjust the operating state of at least one device in the space.

[0072] Afterwards, the current operating state of the device can be adjusted according to the state that the device needs to enter in the future, so that the device can be adjusted in a timely manner according to the user's preferences, thereby obtaining the user's preferred device operating state without the user having to operate the device.

[0073] Through the device management method provided by the embodiment of the present invention, the user's health status data and space environment data are obtained; the health status data and the space environment data are input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model; based on the future device state, the operating state of at least one device in the space is adjusted. Thus, the device state preferred by the user can be analyzed while considering the user's health status and space environment, and the adjustment method of the operating state of at least one device in the space can be determined based on the device state preferred by the user, so that the adjustment of the device in the space can be completed without complicated operations by the user, and the adjustment method is simple.

[0074] Reference Figure 2 , shows a flow chart of the steps of a device management method provided in an embodiment of the present invention, which may specifically include the following steps:

[0075] Step 201, obtaining the user's health status data through a camera and / or a sensor of a smart wearable device;

[0076] In an embodiment of the present invention, the devices in the space may be managed while taking the health status of the user into consideration.

[0077] The space may refer to different areas or rooms in a home or building. These spaces may be bedrooms, living rooms, kitchens, bathrooms, offices, garages, etc. At least one device may be provided in the space, and the device may be used to adjust the environment in the space so that the space is in an environment that the user feels comfortable.

[0078] The user's health status data may refer to information related to the user's health, which may be collected through various sensors and devices. The user's health status data may help the smart home system understand the user's health status and intelligently adjust device settings and energy usage based on this information.

[0079] Generally speaking, a user's health status data may include physiological data such as heart rate, blood pressure, body temperature, blood oxygen saturation, and blood sugar level; activity data such as number of steps, amount of exercise, sleep quality, and activity trajectory; and psychological data such as stress level and emotional state.

[0080] In a specific implementation, a camera can be set up in the space to detect the space where the user is currently located and the user's current motion state. At the same time, the user can also wear a smart wearable device, which can be equipped with a wearable observer, so that physiological data such as heart rate, blood pressure, body temperature, blood oxygen saturation, blood sugar level, etc. can be obtained through the smart wearable device, and activity data such as number of steps, amount of exercise, sleep quality, activity trajectory, etc. can also be included. Psychological data such as stress level and emotional state can also be included.

[0081] Step 202, obtaining spatial environment data through at least one of a temperature and humidity sensor, a light sensor, and an external environment data source.

[0082] In a specific implementation, devices for obtaining space environment data can be set up in the space, such as temperature and humidity sensors, air quality sensors, light sensors and other environmental sensors, so that space environment data such as temperature, humidity, air quality, and light intensity can be obtained.

[0083] In addition, environmental data related to the external environment, such as external temperature, external humidity, external weather, etc., can also be obtained through external environmental data sources as a reference for adjusting the space environment.

[0084] In the specific implementation, an environmental context model can be first constructed, which can be used to sense the user's location and activity status through wearable devices or cameras, as well as collect the user's physical health data, determine whether the user is at home and in which room; and adjust lighting and air conditioning equipment according to external environments such as weather changes and day and night. Afterwards, it can perform data dimensionality reduction and feature extraction on the collected data to obtain data that can be used to build a personalized user portrait.

[0085] Specifically, the collected health status data and spatial environment data can first be preprocessed to remove outliers and noise data, convert data of different dimensions into a unified dimension, and fill in missing data to ensure data integrity.

[0086] Afterwards, the health status data and the spatial environment data can be reduced in dimension by methods such as principal component analysis (PCA) and linear discriminant analysis (LDA) to reduce the dimension of the data and retain the main features. For example, the health status data can be reduced in dimension to a low-dimensional feature vector, and the spatial environment data can be reduced in dimension to a low-dimensional feature vector by the principal component analysis method.

[0087] Step 203, inputting the health status data and the spatial environment data into a preset behavior prediction model to obtain the future device status predicted by the behavior prediction model;

[0088] In an embodiment of the present invention, a behavior prediction model can be pre-set. The behavior prediction model can be used to analyze the spatial environment preferred by the user in different time periods based on the input health status data and spatial environment data, and predict the environment preferred by the user in the future, so as to obtain the future device status.

[0089] In one embodiment of the present invention, the step of inputting the health status data and the spatial environment data into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model includes:

[0090] S11, inputting the health status data and the spatial environment data into a preset user portrait prediction model, and obtaining the user's environmental preference information output by the user portrait prediction model;

[0091] In a specific implementation, a user portrait prediction model can be trained first, and the user portrait prediction model can be used to predict the user's preference for the space environment under different environmental conditions and different health conditions of the user. For example, it can predict the user's preference for the space environment setting under different external weather conditions or different time periods.

[0092] In a specific implementation, a user portrait prediction model can be used to classify health status data and spatial environment data, and determine health status labels, behavior preference labels, and environmental sensitivity labels to represent user preference information. Among them, environmental sensitivity labels can be used to mark the health status or environmental status that the user focuses on, such as "poor sleep quality", "high stress state", "prefer soft lighting", etc. Behavioral preference labels can be the frequency of air conditioning use, the opening period of the heater, the lighting use time, the user's daily routine, the user's current health data, seasonal behavior changes, etc.

[0093] S12, inputting the environmental preference information into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0094] Thereafter, the environmental preference information may be input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0095] In a specific implementation, the behavior prediction model may be a long short-term memory network (LSTM), which may analyze the user's behavior patterns in the environmental preference information and analyze possible future device usage requirements.

[0096] In one embodiment of the present invention, the step of inputting the environmental preference information into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model includes:

[0097] S21, inputting the environmental preference information and the spatial environment data into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0098] In a specific implementation, in order to more accurately analyze the future device state of the user's preference, the environmental preference information and the spatial environment data may be further input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0099] In the specific implementation, the formula for using the LSTM model to predict the user's device usage behavior is:

[0100] h t =σ(W h ·x t +U h ·h t-1 +b n )

[0101] Among them, xt represents the input data at the current moment, including environmental preference information and the current spatial environment data, h t-1 represents the previous hidden state, storing past behavior information; σ is the activation function; W h ,U h ,b h is the model parameter; t Represents the predicted future equipment state. It can provide a basis for pre-scheduling and real-time control of equipment.

[0102] Step 204: Based on the future device state, adjust the operating state of at least one device in the space.

[0103] Afterwards, the current operating state of the device can be adjusted according to the state that the device needs to enter in the future, so that the device can be adjusted in a timely manner according to the user's preferences, thereby obtaining the user's preferred device operating state without the user having to operate the device.

[0104] In one embodiment of the present invention, the step of adjusting the operating state of at least one device in the space based on the future device state includes:

[0105] S31, inputting the future device state into a preset reinforcement learning model to obtain a control method of at least one device in the space output by the reinforcement model;

[0106] S32, using the control method of the device to adjust the operating state of at least one device in the space.

[0107] In a specific implementation, a reinforcement learning model can be used to analyze how the operating state of the device should be adjusted. The future device state can first be input into a preset reinforcement learning model. The reinforcement learning model can analyze the control method of at least one device in the space based on the device state that needs to be achieved at the future moment, and use the device control method to adjust the operating state of at least one device in the space. This allows users to automatically adjust the operating state of the device without actively adjusting the device.

[0108] Among them, reinforcement learning (RL) is a machine learning method that uses an agent to learn the optimal strategy in its interaction with the environment to maximize the cumulative reward. In smart home device control, reinforcement learning models can be used to analyze how the operating status of the device should be adjusted, thereby achieving automatic adjustment of the device and improving user comfort and energy efficiency.

[0109] In one embodiment of the present invention, the method further comprises:

[0110] S41, determining energy consumption cost based on an operating status of at least one device in the space;

[0111] S42, determining comfort information based on the operating status of at least one device in the space and the health status data of the user;

[0112] S43, determining a reward function of the reinforcement learning model based on the energy consumption cost and the comfort information;

[0113] S44: Adjust the reinforcement learning model based on the reward function.

[0114] The enhanced model can use comfort and energy cost as reward functions to adjust the output of the enhanced equipment, so that the adjustment of the equipment can reduce the energy cost to a certain extent while satisfying the user's preferences.

[0115] In a specific implementation, the state space of the reinforcement function can be the future device state. According to actual needs, spatial environment data, user health status data, etc. can also be used as input of the state space.

[0116] The action space of the reinforcement function can be the control method of the device (such as adjusting temperature, humidity, light intensity, device switching, etc.).

[0117] The reinforcement function can generate the corresponding action space based on the input of the state space. Afterwards, the reward function can be used to judge whether the action taken in the current state is good or bad, that is, whether the control method of the device output by the action space is appropriate.

[0118] In an embodiment of the present invention, the reward function can be determined based on the energy consumption cost and the comfort information. The energy consumption cost of the adjusted device can be determined based on the operating state of the adjusted device. The user's comfort feeling about the operating state of the adjusted device can be analyzed by the change in the user's health status data after the operating state of the device is adjusted. Thereafter, the reward function of the reinforcement learning model can be determined based on the energy consumption cost and the comfort information. The reinforcement learning model can be adjusted based on the calculation result of the reward function.

[0119] After that, after determining the new future device state, it can be input into the updated reinforcement learning model to obtain the control method of at least one device in the space output by the updated reinforcement model. And continue to analyze whether the control method of the device output in the action space is appropriate according to the reward function, and continue to adjust the reinforcement learning. In this way, the reinforcement learning model can continuously learn better device control methods during use, so that the adjusted device state can better meet the user's preferences.

[0120] In the specific implementation, the reinforcement function can be expressed as:

[0121] State transfer implementation formula: S t+1 =f(S t , A t , E t )

[0122] State space: S = {s_1, s_2, ..., s_t}, where s_t represents the state of the system at time t. (Future device state, spatial environment data, user health status data, etc.) Action space: A = {a_1, a_2, ..., a_m}, a_i represents the actions that the system can perform, such as adjusting temperature, humidity, light intensity, device switches, etc. Environmental change factors: Et, which can be external environmental factors such as outdoor temperature, outdoor humidity, light intensity, air quality, etc.

[0123] Q value update formula:

[0124] Reward function: r_t = -(energy cost + comfort penalty). Learning rate: α. Discount factor: γ.

[0125] By continuously updating the Q value, the system learns the optimal action to take in each state.

[0126] In one embodiment of the present invention, the method further comprises:

[0127] S41, receiving feedback information from the user;

[0128] S42: Adjust the reinforcement learning model based on the feedback information from the user.

[0129] In a specific implementation, the user can also actively provide feedback information to feedback whether the current operating state of the device is appropriate. Therefore, the reinforcement learning model can also be adjusted based on the user's feedback information, so that through the user's active adjustment, the control method of the device output by the reinforcement learning model can make the user in a more comfortable environment.

[0130] Through the device management method provided by the embodiment of the present invention, the user's health status data and space environment data are obtained; the health status data and the space environment data are input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model; based on the future device state, the operating state of at least one device in the space is adjusted. Thus, the device state preferred by the user can be analyzed while considering the user's health status and space environment, and the adjustment method of the operating state of at least one device in the space can be determined based on the device state preferred by the user, so that the adjustment of the device in the space can be completed without complicated operations by the user, and the adjustment method is simple.

[0131] Reference Figure 3 , shows a schematic flow chart of a device management method provided in an embodiment of the present invention, which may specifically include the following steps:

[0132] (1) Obtain the user's health status data through a camera and / or sensors of a smart wearable device; obtain spatial environment data through at least one of a temperature and humidity sensor, a light sensor, and an external environment data source.

[0133] In an embodiment of the present invention, the devices in the space may be managed while taking the health status of the user into consideration.

[0134] The space may refer to different areas or rooms in a home or building. These spaces may be bedrooms, living rooms, kitchens, bathrooms, offices, garages, etc. At least one device may be provided in the space, and the device may be used to adjust the environment in the space so that the space is in an environment that the user feels comfortable.

[0135] The user's health status data may refer to information related to the user's health, which may be collected through various sensors and devices. The user's health status data may help the smart home system understand the user's health status and intelligently adjust device settings and energy usage based on this information.

[0136] Generally speaking, a user's health status data may include physiological data such as heart rate, blood pressure, body temperature, blood oxygen saturation, and blood sugar level; activity data such as number of steps, amount of exercise, sleep quality, and activity trajectory; and psychological data such as stress level and emotional state.

[0137] In a specific implementation, a camera can be set up in the space to detect the space where the user is currently located and the user's current motion state. At the same time, the user can also wear a smart wearable device, which can be equipped with a wearable observer, so that physiological data such as heart rate, blood pressure, body temperature, blood oxygen saturation, blood sugar level, etc. can be obtained through the smart wearable device, and activity data such as number of steps, amount of exercise, sleep quality, activity trajectory, etc. can also be included. Psychological data such as stress level and emotional state can also be included.

[0138] In a specific implementation, devices for obtaining space environment data can be set up in the space, such as temperature and humidity sensors, air quality sensors, light sensors and other environmental sensors, so that space environment data such as temperature, humidity, air quality, and light intensity can be obtained.

[0139] In addition, environmental data related to the external environment, such as external temperature, external humidity, external weather, etc., can also be obtained through external environmental data sources as a reference for adjusting the space environment.

[0140] (2) The collected health status data and spatial environment data can be preprocessed to remove outliers and noise data, convert data of different dimensions into a unified dimension, and fill in missing data to ensure data integrity.

[0141] (3) In the specific implementation, an environmental context model can be first constructed. The environmental context model can be used to sense the user's location and activity status through wearable devices or cameras, collect the user's physical health data, determine whether the user is at home and the room he is in; adjust the lighting and air conditioning equipment according to the external environment such as weather changes and day and night. Afterwards, it can perform data dimensionality reduction and feature extraction on the collected data to obtain data that can be used to build a personalized user portrait.

[0142] Data dimensionality reduction can be performed on health status data and spatial environment data, and the dimensions of the data can be reduced and the main features can be retained through methods such as principal component analysis (PCA) and linear discriminant analysis (LDA). For example, the health status data can be reduced to a low-dimensional feature vector and the spatial environment data can be reduced to a low-dimensional feature vector through the principal component analysis method.

[0143] (4) A user portrait prediction model can be trained, which can be used to predict the user's preference for the space environment under different environmental conditions and different health conditions of the user. For example, it can predict the user's preference for the space environment under different external weather conditions or different time periods.

[0144] In a specific implementation, a user portrait prediction model can be used to classify health status data and spatial environment data, and determine health status labels, behavior preference labels, and environmental sensitivity labels to represent user preference information. Among them, environmental sensitivity labels can be used to mark the health status or environmental status that the user focuses on, such as "poor sleep quality", "high stress state", "prefer soft lighting", etc. Behavioral preference labels can be the frequency of air conditioning use, the opening period of the heater, the lighting use time, the user's daily routine, the user's current health data, seasonal behavior changes, etc.

[0145] (5) Input the environmental preference information into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0146] In a specific implementation, the behavior prediction model may be a long short-term memory network (LSTM), which may analyze the user's behavior patterns in the environmental preference information and analyze possible future device usage requirements.

[0147] In order to more accurately analyze the future device state of the user's preference, the environmental preference information and the spatial environment data may be further input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0148] In the specific implementation, the formula for using the LSTM model to predict the user's device usage behavior is:

[0149] h t =σ(W h ·x t +U h ·h t-1 +b h )

[0150] Among them, xt represents the input data at the current moment, including environmental preference information and the current spatial environment data, h t-1 represents the previous hidden state, storing past behavior information; σ is the activation function; W h ,U h ,b h is the model parameter; t Represents the predicted future equipment state. It can provide a basis for pre-scheduling and real-time control of equipment.

[0151] (6) How the current operating state of the device should be adjusted based on the state that the device needs to enter in the future, so that the device can be adjusted in a timely manner according to the user's preferences, thereby obtaining the user's preferred device operating state without the user having to operate the device.

[0152] The future device state is input into a preset reinforcement learning model to obtain a control method of at least one device in the space output by the reinforcement model; and the operating state of at least one device in the space is adjusted using the control method of the device.

[0153] In a specific implementation, a reinforcement learning model can be used to analyze how the operating state of the device should be adjusted. The future device state can first be input into a preset reinforcement learning model. The reinforcement learning model can analyze the control method of at least one device in the space based on the device state that needs to be achieved at the future moment, and use the device control method to adjust the operating state of at least one device in the space. This allows users to automatically adjust the operating state of the device without actively adjusting the device.

[0154] Among them, reinforcement learning (RL) is a machine learning method that uses an agent to learn the optimal strategy in its interaction with the environment to maximize the cumulative reward. In smart home device control, reinforcement learning models can be used to analyze how the operating status of the device should be adjusted, thereby achieving automatic adjustment of the device and improving user comfort and energy efficiency.

[0155] (7) Determine the energy consumption cost based on the operating status of at least one device in the space; determine the comfort information based on the operating status of at least one device in the space and the health status data of the user; determine the reward function of the reinforcement learning model based on the energy consumption cost and the comfort information; and adjust the reinforcement learning model based on the reward function.

[0156] The enhanced model can use comfort and energy cost as reward functions to adjust the output of the enhanced equipment, so that the adjustment of the equipment can reduce the energy cost to a certain extent while satisfying the user's preferences.

[0157] In a specific implementation, the state space of the reinforcement function can be the future device state. According to actual needs, spatial environment data, user health status data, etc. can also be used as input of the state space.

[0158] The action space of the reinforcement function can be the control method of the device (such as adjusting temperature, humidity, light intensity, device switching, etc.).

[0159] The reinforcement function can generate the corresponding action space based on the input of the state space. Afterwards, the reward function can be used to judge whether the action taken in the current state is good or bad, that is, whether the control method of the device output by the action space is appropriate.

[0160] In an embodiment of the present invention, the reward function can be determined based on the energy consumption cost and the comfort information. The energy consumption cost of the adjusted device can be determined based on the operating state of the adjusted device. The user's comfort feeling about the operating state of the adjusted device can be analyzed by the change in the user's health status data after the operating state of the device is adjusted. Thereafter, the reward function of the reinforcement learning model can be determined based on the energy consumption cost and the comfort information. The reinforcement learning model can be adjusted based on the calculation result of the reward function.

[0161] After that, after determining the new future device state, it can be input into the updated reinforcement learning model to obtain the control method of at least one device in the space output by the updated reinforcement model. And continue to analyze whether the control method of the device output in the action space is appropriate according to the reward function, and continue to adjust the reinforcement learning. In this way, the reinforcement learning model can continuously learn better device control methods during use, so that the adjusted device state can better meet the user's preferences.

[0162] In the specific implementation, the reinforcement function can be expressed as:

[0163] State transfer implementation formula: S t+1 =f(S t , A t , Et )

[0164] State space: S = {s_1, s_2, ..., s_t}, where s_t represents the state of the system at time t. (Future device state, spatial environment data, user health status data, etc.) Action space: A = {a_1, a_2, ..., a_m}, a_i represents the actions that the system can perform, such as adjusting temperature, humidity, light intensity, device switches, etc. Environmental change factors: Et, which can be external environmental factors such as outdoor temperature, outdoor humidity, light intensity, air quality, etc.

[0165] Q value update formula:

[0166] Reward function: r_t = -(energy cost + comfort penalty). Learning rate: α. Discount factor: γ.

[0167] By continuously updating the Q value, the system learns the optimal action to take in each state.

[0168] In one embodiment of the present invention, the method further comprises:

[0169] (9) receiving user feedback information; and adjusting the reinforcement learning model based on the user feedback information.

[0170] In a specific implementation, the user can also actively provide feedback information to feedback whether the current operating state of the device is appropriate. Therefore, the reinforcement learning model can also be adjusted based on the user's feedback information, so that through the user's active adjustment, the control method of the device output by the reinforcement learning model can make the user in a more comfortable environment.

[0171] Through the device management method provided by the embodiment of the present invention, the user's health status data and space environment data are obtained; the health status data and the space environment data are input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model; based on the future device state, the operating state of at least one device in the space is adjusted. Thus, the device state preferred by the user can be analyzed while considering the user's health status and space environment, and the adjustment method of the operating state of at least one device in the space can be determined based on the device state preferred by the user, so that the adjustment of the device in the space can be completed without complicated operations by the user, and the adjustment method is simple.

[0172] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0173] Reference Figure 4 , shows a structural block diagram of a device management device provided in an embodiment of the present invention, which may specifically include the following modules:

[0174] The data acquisition module 401 is used to acquire the user's health status data and spatial environment data;

[0175] The device state prediction module 402 is used to input the health state data and the spatial environment data into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model;

[0176] The device adjustment module 403 is used to adjust the operating state of at least one device in the space based on the future device state.

[0177] Optionally, the data acquisition module includes:

[0178] The health status acquisition submodule is used to obtain the user's health status data through the camera and / or the sensor of the smart wearable device;

[0179] The space environment acquisition submodule is used to acquire space environment data through at least one of a temperature and humidity sensor, a light sensor, and an external environment data source.

[0180] Optionally, the device state prediction module includes:

[0181] An environmental preference acquisition submodule, used to input the health status data and the spatial environment data into a preset user portrait prediction model, and obtain the user's environmental preference information output by the user portrait prediction model;

[0182] The future device state acquisition submodule is used to input the environmental preference information into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0183] Optionally, the future device status acquisition submodule includes:

[0184] The future device state prediction unit is used to input the environmental preference information and the spatial environment data into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0185] Optionally, the device adjustment module includes:

[0186] A control mode acquisition submodule, used to input the future device state into a preset reinforcement learning model to obtain a control mode of at least one device in the space output by the reinforcement model;

[0187] The operating status adjustment submodule is used to adjust the operating status of at least one device in the space by adopting the control method of the device.

[0188] Optionally, the device further comprises:

[0189] An energy consumption cost determination module, configured to determine the energy consumption cost based on the operating status of at least one device in the space;

[0190] a comfort level determination module, configured to determine comfort level information based on the operating status of at least one device in the space and the health status data of the user;

[0191] A reward function determination module, used to determine a reward function of the reinforcement learning model based on the energy consumption cost and the comfort information;

[0192] A model adjustment module is used to adjust the reinforcement learning model based on the reward function.

[0193] Optionally, the device further comprises:

[0194] A feedback information acquisition module, used to receive user feedback information;

[0195] A feedback information model adjustment module is used to adjust the reinforcement learning model based on the feedback information of the user.

[0196] Through the device management device provided by the embodiment of the present invention, the user's health status data and space environment data are obtained; the health status data and the space environment data are input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model; based on the future device state, the operating state of at least one device in the space is adjusted. Thus, the device state preferred by the user can be analyzed while considering the user's health status and space environment, and the adjustment method of the operating state of at least one device in the space can be determined based on the device state preferred by the user, so that the adjustment of the device in the space can be completed without complicated operations by the user, and the adjustment method is simple.

[0197] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0198] In addition, an embodiment of the present invention further provides an electronic device, such as Figure 5 As shown, it includes a processor 501, a communication interface 502, a memory 503 and a communication bus 504, wherein the processor 501, the communication interface 502, and the memory 503 communicate with each other through the communication bus 504.

[0199] Memory 503, used for storing computer programs;

[0200] The processor 501 is used to execute the program stored in the memory 503, and implements the following steps:

[0201] Obtain the user's health status data and spatial environment data;

[0202] Inputting the health status data and the spatial environment data into a preset behavior prediction model to obtain a future device state predicted by the behavior prediction model;

[0203] Based on the future device state, the operating state of at least one device in the space is adjusted.

[0204] Optionally, the step of obtaining the user's health status data and spatial environment data includes:

[0205] Obtain the user's health status data through the camera and / or sensors of smart wearable devices;

[0206] The spatial environment data is obtained through at least one of a temperature and humidity sensor, a light sensor, and an external environment data source.

[0207] Optionally, the step of inputting the health status data and the spatial environment data into a preset behavior prediction model to obtain the future device status predicted by the behavior prediction model includes:

[0208] Inputting the health status data and the spatial environment data into a preset user portrait prediction model to obtain the user's environmental preference information output by the user portrait prediction model;

[0209] The environmental preference information is input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0210] Optionally, the step of inputting the environmental preference information into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model includes:

[0211] The environmental preference information and the spatial environment data are input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

[0212] Optionally, the step of adjusting the operating state of at least one device in the space based on the future device state includes:

[0213] Inputting the future device state into a preset reinforcement learning model to obtain a control method of at least one device in the space output by the reinforcement model;

[0214] The control method of the device is adopted to adjust the operating state of at least one device in the space.

[0215] Optionally, the method further comprises:

[0216] determining energy consumption costs based on an operating status of at least one device in the space;

[0217] Determining comfort information based on the operating status of at least one device in the space and the health status data of the user;

[0218] Determining a reward function of the reinforcement learning model based on the energy consumption cost and the comfort information;

[0219] Based on the reward function, the reinforcement learning model is adjusted.

[0220] Optionally, the method further comprises:

[0221] Receive feedback from users;

[0222] The reinforcement learning model is adjusted based on the feedback information of the user.

[0223] Through the electronic device provided by the embodiment of the present invention, the user's health status data and spatial environment data are obtained; the health status data and the spatial environment data are input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model; based on the future device state, the operating state of at least one device in the space is adjusted. In this way, the device state preferred by the user can be analyzed while considering the user's health status and spatial environment, and the adjustment method of the operating state of at least one device in the space can be determined based on the device state preferred by the user, so that the adjustment of the device in the space can be completed without complicated operations by the user, and the adjustment method is simple.

[0224] The communication bus mentioned in the above terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0225] The communication interface is used for communication between the above terminal and other devices.

[0226] The memory may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0227] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0228] like Figure 6 As shown, in another embodiment provided by the present invention, a computer-readable storage medium 601 is also provided, in which instructions are stored. When the computer-readable storage medium is run on a computer, the computer executes the device management method described in the above embodiment.

[0229] In another embodiment of the present invention, a computer program product including instructions is provided. When the computer program product is run on a computer, the computer executes the device management method described in the above embodiment.

[0230] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk Solid State Disk (SSD)), etc.

[0231] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0232] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0233] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A device management method, characterized in that: include: Obtain the user's health status data and spatial environment data; Inputting the health status data and the spatial environment data into a preset behavior prediction model to obtain a future device state predicted by the behavior prediction model; Based on the future device state, the operating state of at least one device in the space is adjusted.

2. The method according to claim 1, characterized in that The step of obtaining the user's health status data and spatial environment data includes: Obtain the user's health status data through the camera and / or sensors of smart wearable devices; The spatial environment data is obtained through at least one of a temperature and humidity sensor, a light sensor, and an external environment data source.

3. The method according to claim 1, characterized in that The step of inputting the health status data and the spatial environment data into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model includes: Inputting the health status data and the spatial environment data into a preset user portrait prediction model to obtain the user's environmental preference information output by the user portrait prediction model; The environmental preference information is input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

4. The method according to claim 3, characterized in that The step of inputting the environmental preference information into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model includes: The environmental preference information and the spatial environment data are input into a preset behavior prediction model to obtain the future device state predicted by the behavior prediction model.

5. The method according to claim 1, characterized in that: The step of adjusting the operating state of at least one device in the space based on the future device state includes: Inputting the future device state into a preset reinforcement learning model to obtain a control method of at least one device in the space output by the reinforcement model; The control method of the device is adopted to adjust the operating state of at least one device in the space.

6. The method according to claim 5, characterized in that The method further comprises: determining energy consumption costs based on an operating status of at least one device in the space; Determining comfort information based on the operating status of at least one device in the space and the health status data of the user; Determining a reward function of the reinforcement learning model based on the energy consumption cost and the comfort information; Based on the reward function, the reinforcement learning model is adjusted.

7. The method according to claim 5, characterized in that The method further comprises: Receive user feedback; The reinforcement learning model is adjusted based on the feedback information of the user.

8. A device management device, characterized in that: include: A data acquisition module is used to obtain the user's health status data and spatial environment data; An equipment state prediction module, used for inputting the health state data and the spatial environment data into a preset behavior prediction model to obtain a future equipment state predicted by the behavior prediction model; The device adjustment module is used to adjust the operating state of at least one device in the space based on the future device state.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; The memory is used to store computer programs; The processor is used to implement the method according to any one of claims 1 to 7 when executing the program stored in the memory.

10. A computer-readable medium having instructions stored thereon, which, when executed by one or more processors, cause the processors to perform the method according to any one of claims 1 to 7.

Images