Device control method, and device control unit

Abstract

To provide a device control method and the like capable of effectively controlling a device.SOLUTION: The device control method includes the following steps: acquiring location information of a user; determining details of control by an air conditioner based on the location information of the user; and executing the details of control by an air conditioner. The details of control by an air conditioner are either control to turn on the air conditioner or control to turn off the air conditioner.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present disclosure relates to an appliance control method and an appliance control device. [Background technology]

[0002] Patent Document 1 discloses that a recommended pattern generated by a recommended pattern generation unit, general statistical information, and individual statistical information are presented to a user of an electrical device. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6098273 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology disclosed in Patent Document 1 may not be able to control the device effectively.

[0005] Therefore, the present disclosure provides a device control method and the like that can effectively control devices. [Means for solving the problem]

[0006] An equipment control method according to one aspect of the present disclosure acquires user location information, determines control content for an air conditioning device based on the user location information, and executes the control content for the air conditioning device, wherein the control content for the air conditioning device is either control to turn on the power of the air conditioning device or control to turn off the power of the air conditioning device.

[0007] These comprehensive or specific aspects may be realized as a system, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM, or may be realized as any combination of a system, a method, an integrated circuit, a computer program, and a recording medium. [Effects of the Invention]

[0008] The device control method according to one aspect of the present disclosure can effectively control devices. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a hardware configuration diagram of a device control system according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating hardware of the device control device according to the embodiment. [Figure 3A] FIG. 3A is a diagram showing an example of user behavior and changes in device status. [Figure 3B] FIG. 3B is a diagram showing an example of user behavior and changes in device status. [Figure 3C] FIG. 3C is a diagram showing an example of user behavior and changes in device status. [Figure 4] FIG. 4 is a diagram showing an example of a prediction of a change in the state of a device. [Figure 5] FIG. 5 is a diagram showing an example of a prediction of a change in the state of a device. [Figure 6A] FIG. 6A is a diagram illustrating an example of a proposed rule. [Figure 6B] FIG. 6B is a diagram illustrating an example of a proposed rule. [Figure 6C] FIG. 6C is a diagram illustrating an example of a proposed rule. [Figure 7] FIG. 7 is a flowchart showing an example of a device control method according to the embodiment. [Figure 8A] FIG. 8A is a diagram showing an example of a notification. [Figure 8B] FIG. 8B is a diagram showing an example of a notification. [Figure 8C] FIG. 8C is a diagram showing an example of a notification. [Figure 8D] FIG. 8D is a diagram showing an example of a notification. [Figure 9] FIG. 9 is a flowchart showing an example of a device control method according to the modified example of the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, the embodiments will be specifically described with reference to the drawings.

[0011] The embodiments described below are all comprehensive or specific examples, and the numerical values, shapes, materials, components, the arrangement and connection of the components, steps, and the order of steps shown in the following embodiments are merely examples and are not intended to limit the scope of the claims.

[0012] In addition, the drawings are not necessarily strict illustrations, and the same reference numerals are used to designate substantially the same components in the drawings, and redundant explanations are omitted or simplified.

[0013] (Findings underlying this disclosure) If a user's response is required as a necessary condition for executing device control, the user may feel annoyed because they have to make many responses. Also, if device control is automatically executed and control is executed unintended by the user, the user may feel annoyed because they have to cancel the device control. Therefore, the present inventors have arrived at the present disclosure described below.

[0014] A device control method according to one aspect of the present disclosure includes acquiring user behavior, predicting multiple future behavioral information of the user based on the user behavior, wherein the multiple behavioral information includes multiple behaviors and predicted times for which each of the multiple behaviors is predicted to be performed, determining device control content corresponding to each of the multiple behaviors based on the multiple behaviors and the predicted times, and executing the device control content based on the predicted times.

[0015] According to this, the device control content is determined based on the user's multiple future actions and the expected duration of each action. Therefore, the device control can be automatically executed without requiring the user's reaction as a necessary condition for executing the device control. Furthermore, because the device control content to be executed is determined based on the user's multiple future actions and the expected duration of each action, it is more likely that the automatically executed device control is the control intended by the user. In this way, the device can be effectively controlled without causing the user any inconvenience.

[0016] For example, the plurality of actions may correspond to operations that change the state of a device located in a space in which the user is present.

[0017] According to this, for example, when a user returns home, an operation to change the state of the entrance light to ON is often performed, and the operations to change the state of devices required for the user's actions are somewhat predetermined. Therefore, from the user's actions, it is possible to predict the operations that will change the state of devices located in the space where the user is present, and ultimately to predict the state changes of the devices.

[0018] For example, the predicted time may be the period from the time when the plurality of behavioral information is predicted to the time when the plurality of behaviors are predicted to be executed at the latest, or the time when each of the plurality of behaviors is predicted to be executed.

[0019] In this way, the predicted time may be a period from the time when the plurality of pieces of behavioral information are predicted, or may be a timing.

[0020] For example, the predicted time may be predicted based on a time corresponding to universal time.

[0021] For example, the plurality of behavioral information may further include the probability of occurrence of each of the plurality of behaviors, and the control content of the device corresponding to each of the plurality of behaviors may be determined based on the plurality of behaviors, the predicted time, and the occurrence probability.

[0022] According to this, the control content of the device to be executed is determined based on the user's multiple future actions, the expected time for each action, and the probability of each action occurring, further increasing the likelihood that the control will be as intended by the user.

[0023] For example, the plurality of actions correspond to operations that change the state of a device located in a space where the user is present, and the control content of the device corresponding to each of the plurality of actions may be at least one of a first control that changes the state of a first device whose state changes due to the operation without accepting the operation, a first control that changes the state of a first device whose state changes due to the operation without accepting the operation, a second control that notifies the user whether or not to change the first control, and a third control that notifies the user whether or not to perform the first control that changes the state of the first device whose state changes due to the operation.

[0024] This allows switching between performing a first control that automatically controls the device, performing a second control that automatically controls the device and notifies the user whether or not to change the first control, and performing a third control that notifies the user whether or not to perform the first control, depending on the situation.

[0025] For example, the control content of the device corresponding to an action whose occurrence probability falls within a first probability range may be determined to be the first control, the control content of the device corresponding to an action whose occurrence probability falls within a second probability range lower than the first probability range may be determined to be the first control and the second control, and the control content of the device corresponding to an action whose occurrence probability falls within a third probability range lower than the second probability range may be determined to be the third control.

[0026] According to this, the control content of the device corresponding to an action with a high probability of occurrence is likely to be what the user intended, so automatic device control can be performed. Furthermore, the control content of the device corresponding to an action with a relatively low probability of occurrence may not be what the user intended, so automatic device control can be performed and then a notification can be made as to whether or not to change the control. Furthermore, the control content of the device corresponding to an action with a low probability of occurrence is likely to be what the user did not intend, so a notification can be made as to whether or not to perform automatic device control.

[0027] For example, the control content of the device corresponding to the behavior whose predicted time falls within a first period may be determined to be the first control, the control content of the device corresponding to the behavior whose predicted time falls within a second period after the first period may be determined to be the first control and the second control, and the control content of the device corresponding to the behavior whose predicted time falls within a third period after the second period may be determined to be the third control.

[0028] According to this, for actions whose predicted time is approaching, quick control may be required, so automatic device control can be performed. Also, for actions whose predicted time is some distance away, automatic device control can be performed and then a notification can be given as to whether or not to change the control. Also, for actions whose predicted time is still far away, quick control is unlikely to be required, so a notification can be given as to whether or not to perform automatic device control.

[0029] For example, the controls of the devices corresponding to the plurality of actions whose predicted times to be performed are each within a predetermined time range may be executed at the same timing.

[0030] This allows devices corresponding to actions with similar predicted times to be controlled at the same time.

[0031] A program according to one aspect of the present disclosure is a program that causes a computer to execute the above-described device control method.

[0032] This makes it possible to provide a program that can effectively control devices.

[0033] A device control device according to one embodiment of the present disclosure includes a processor and a memory, wherein the processor (a) acquires user behavior sensed by a sensor, (b) predicts multiple pieces of future behavioral information of the user based on the user behavior, wherein the multiple pieces of behavioral information include multiple actions and predicted times for which each of the multiple actions is predicted to be performed, (c) determines device control content corresponding to each of the multiple actions by referring to rules recorded in the memory based on the multiple actions and the predicted times, and (d) executes the device control content based on the predicted times.

[0034] This makes it possible to provide a device control device that can effectively control devices.

[0035] (Embodiment) Hereinafter, embodiments will be described with reference to the drawings.

[0036] First, a device control system 1 according to the embodiment will be described with reference to FIG.

[0037] FIG. 1 is a hardware configuration diagram of a device control system 1 according to an embodiment.

[0038] As shown in Fig. 1, the device control system 1 according to the present embodiment includes a device control device 100, a sensor 200, devices 300a to 300c, and devices 400a to 400c. One example of the devices 300a to 300c and 400a to 400c is electronic equipment located within an environment. One example of the environment is an indoor space such as a home, an office, or a commercial facility. Note that the devices 300a to 300c and the devices 400a to 400c may be the same type of device.

[0039] (Device control device 100) The device control device 100 is connected to the sensor 200, the devices 300a to 300c, and the devices 400a to 400c via wired or wireless connections. The device control device 100 may be a server. Servers include physical servers and cloud servers. A cloud server is a virtual server provided via a computer network (e.g., the Internet). The device control device 100 may also be a computer located within the above environment.

[0040] The device control device 100 shown in FIG. 1 includes an information acquisition unit 101, a prediction unit 102, a control content determination unit 103, and a control instruction unit 104.

[0041] The information acquisition unit 101 acquires user behavior. The information acquisition unit 101 may acquire state information of the devices 300a to 300c and acquire the user behavior based on the state information. This is because the states of the devices 300a to 300c and the user behavior may be associated with each other, and the user behavior may be acquired from the state information of the devices 300a to 300c. For example, by acquiring state information that the entrance light is turned on, the user behavior of returning home may be acquired.

[0042] The prediction unit 102 predicts multiple pieces of future behavioral information of the user based on the user's behavior. Here, the multiple pieces of behavioral information include multiple actions and the predicted time for each of the multiple actions. The multiple actions correspond to operations that change the states of the devices 300a-300c located in the space where the user is present. For example, when the user returns home, an operation that changes the state of the entrance light to ON is often performed, and the operations that change the states of the devices 300a-300c required for the user's behavior are somewhat predetermined. Therefore, it is possible to predict operations that change the states of the devices 300a-300c located in the space where the user is present from the user's behavior. Ultimately, it is possible to predict information changes of the multiple devices 300a-300c. The prediction unit 102 may predict information changes of the devices 300a-300c based on the state information of the devices 300a-300c. The information changes of the multiple devices 300a-300c refer to state changes of the devices 300a-300c at multiple future timings or time periods. Information changes of multiple devices 300a to 300c may be, for example, status changes after 1 second, after 10 seconds, after 1 minute, or status changes within a period of 30 seconds from a certain point in time, or status changes within a period of 1 minute to 5 minutes from a certain point in time.

[0043] The control content determination unit 103 determines the control content of the devices 400a-400c corresponding to each of the multiple actions based on the multiple actions and the estimated time included in the multiple pieces of action information. Specifically, the control content determination unit 103 determines the control content of the devices 400a-400c corresponding to each of the multiple actions based on the multiple actions and the estimated time included in the multiple pieces of action information, with reference to rules recorded in memory. Note that the control content determination unit 103 may determine the control content of the devices 400a-400c using future information changes in the multiple devices 300a-300c.

[0044] The control instruction unit 104 executes the control content of the devices 400a to 400c based on the predicted expected time. For example, the control instruction unit 104 transmits information for causing the devices 400a to 400c to execute the determined control content in accordance with the predicted expected time.

[0045] The components of the device control device 100 will be described in detail below.

[0046] FIG. 2 is a diagram illustrating hardware of the device control device 100 according to the embodiment.

[0047] As shown in FIG. 2, the device control device 100 includes a processor 1001 and a memory 1002 connected to the processor 1001. The memory 1002 is a read-only memory (ROM) and a random access memory (RAM), and can store programs to be executed by the processor 1001. The processor 1001 functions as a sequence manager and a device manager when instructions or software programs stored in the memory 1002 are executed. If the device control device 100 is a cloud server, the processor 1001 and the memory 1002 function as virtual hardware. The information acquisition unit 101, the prediction unit 102, the control content determination unit 103, and the control instruction unit 104 are realized by the processor 1001, which executes programs stored in the memory 1002. The memory 1002 may store proposed rules (described later) or modified rules (described later). The memory 1002 in which the program is recorded, the memory 1002 in which the proposed rules are recorded, and the memory 1002 in which the modified rules are recorded may be different memories.

[0048] (Sensor 200) The sensor 200 acquires status information such as the status of the devices 300a-300c in the environment and the details of operations on the devices 300a-300c. For example, if the devices 300a-300c are televisions, the status information of the televisions includes information on whether the televisions are on or off and the details of the programs being watched on the televisions. Furthermore, if the devices 300a-300c are air conditioners, the status information of the air conditioners includes information on whether the air conditioners are on or off and setting information of the air conditioners such as temperature and humidity.

[0049] The sensor 200 may be located within the environment and may be disposed inside the devices 300a to 300c. The sensor 200 may be substituted by a sensor provided for executing the functions of the devices 300a to 300c. Furthermore, the devices 300a to 300c may transmit status information of the devices 300a to 300c to the information acquisition unit 101, in which case the sensor 200 may be omitted. For example, when the devices 300a to 300c execute their functions, a log of the device execution may be transmitted to the information acquisition unit 101.

[0050] The sensor 200 may also acquire information about a user in the environment. Examples of the user information include a user ID for identifying the user, the user's location, and the user's behavior. The user's behavior may be, for example, the user's movement, the user's motion, etc.

[0051] An example of the sensor 200 is a camera, and the user's ID, user location, and user behavior are identified using camera images and reference data. The device control device 100 may identify the user's ID, user location, behavior, and action using the camera images and reference data, or a processing unit external to the device control device 100 may identify the user's ID, user location, and user behavior. Other examples of the sensor 200 include an infrared sensor, an illuminance sensor, a temperature sensor, a pressure sensor, and a distance sensor.

[0052] The sensor 200 may have a control unit and generate information about the states of the devices 300a to 300c or information about the user based on data sensed by the sensor 200. For example, the control unit included in the sensor 200 may acquire information about the amount of power of the devices 300a to 300c from the sensor 200 or the devices 300a to 300c, and determine that the devices 300a to 300c are ON when the acquired amount of power reaches or exceeds a predetermined level.

[0053] The control unit included in the sensor 200 may output the sensed data in association with the time.

[0054] (Equipment 300a~300c, equipment 400a~400c) The status information of the devices 300a to 300c is acquired by the information acquisition unit 101. Furthermore, the devices 400a to 400c receive control instructions from the control instruction unit 104 and execute the control content indicated by the control instruction. The devices 300a to 300c and the devices 400a to 400c are not distinguished from each other and may be the same device. Hereinafter, when it is not necessary to distinguish between the devices 300a to 300c, the devices 300a to 300c will also be referred to as device 300, and when it is not necessary to distinguish between the devices 400a to 400c, the devices 400a to 400c will also be referred to as device 400.

[0055] Examples of devices 300 and 400 include household electrical appliances (home appliances), home equipment, mobile terminals, speakers, etc. Examples of home appliances include microwave ovens, rice cookers, blenders, electric ovens, electric toasters, electric kettles, hot plates, induction heating (IH) cookers, roasters, bakeries, electric pressure cookers, electric waterless cookers, multi-cookers, coffee makers, refrigerators, washing machines, dishwashers, vacuum cleaners, air conditioners, air purifiers, humidifiers, hair dryers, electric fans, ion generators, TVs, and recorders. Examples of home equipment include lamps, electric shutters, electronic locks, and electric water heaters for bathtubs.

[0056] The device 300 may transmit to the information acquisition unit 101 state information relating to changes in the state of the device, such as an operation history and a control history.

[0057] (Prediction of state changes of device 300) Next, prediction of a state change of the device 300 will be described. The prediction unit 102 predicts a plurality of pieces of user behavior information at a predetermined timing or period in the future based on the information acquired by the information acquisition unit 101. Note that predicting a plurality of pieces of future user behavior information also means predicting a state change of the device 300. This is because, as described above, the user's behavior and the state change of the device 300 are related to each other, such as the relationship between the user's behavior of returning home and the turning on of the entrance light.

[0058] 3A to 3C are diagrams showing an example of user behavior and state changes of device 300, specifically, showing an example of user behavior and state changes of device 300 from the time the user returns home until the air conditioner is turned on. From the top, the times when the user behavior or state changes of device 300 occurred are shown in chronological order.

[0059] In Figures 3A to 3C, the process from No. 1 returning home to No. 3 moving from the entrance to hallway A is the same. However, users do not necessarily behave in the same way, and therefore user behavior or state changes of the device 300 do not necessarily occur at the same time. In other words, users do not necessarily operate the device 300 in the same way every day, and the same state changes of the device 300 do not necessarily occur every day. Therefore, by using this information as learning data and learning, a learning model can be constructed that outputs user behavior (state changes of the device 300) and their occurrence probability. Information is input into this learning model, and the output result corresponds to a prediction of user behavior (state changes of the device 300). To construct the learning model, a machine learning algorithm is used, for example. User behavior or state changes of the device 300 that occurred within a certain time period, the elapsed time since the last time a state change of the device 300 to be predicted occurred, or a certain number of user behaviors or state changes of the device are input, and the model learns whether a specified state change of the device 300 will occur after a certain time period. Alternatively, it is possible to learn whether or not a state change will occur in the specified device 300, and at the same time, learn the time until the state change occurs in the specified device 300. As a machine learning algorithm capable of constructing such a learning model, for example, a logistic regression model, a decision tree model, a neural network, or the like is used.

[0060] 4 and 5 are diagrams showing examples of predictions of state changes of the device 300. The prediction examples shown in Fig. 4 and 5 are based on input information indicating the user's return home to the learning model.

[0061] 4 shows user actions (changes in the state of the device 300), their occurrence probabilities, and expected occurrence times. The expected occurrence times are an example of predicted times at which each of a plurality of user actions is predicted to be performed.

[0062] The scheduled occurrence time may be a period from the time when multiple pieces of behavioral information are predicted, such as when input information to a learning model is acquired or when the scheduled occurrence time is output using a learning model, to the time when the user's behavior (a change in the state of the device 300) is predicted to be executed at the latest. The scheduled occurrence time may also be the time when each of the user's multiple behaviors is predicted to be executed. Thus, the scheduled occurrence time may be a period or a timing. For example, the scheduled occurrence time may be predicted based on a time corresponding to universal time. Specifically, the scheduled occurrence time may be a period between times, such as from aa:0 ...

[0063] Here, an example of a state change of the device 300 may refer to a state change of the device 300 that occurs within a predetermined time (i.e., a scheduled occurrence time) based on a user's operation of the device 300. Examples of the predetermined time include 1 second, 10 seconds, 30 seconds, 1 minute, and 5 minutes. For example, the state change of the device shown in FIG. 4 illustrates a change in the state of the device 300 caused by a user's operation of the device 300. Since the lamp turns on based on a user's operation to turn on the lamp switch, "lamp ON" is listed as an item, and its occurrence probability is associated with the scheduled occurrence time. The occurrence probability is the probability that the corresponding state change of the device 300 occurs at the scheduled occurrence time. An example of the longest time for the predicted scheduled occurrence time is a relatively short time, such as 5 minutes, 10 minutes, or 30 minutes.

[0064] An example of information input to the learning model is the status of the devices 300a to 300c. Specifically, when the user returns home, each of the devices 300a to 300c is either ON or OFF. In other words, the status of the devices 300a to 300c at a certain timing is input to the learning model. In this case, for example, the output result shown in FIG. 4 is obtained in accordance with a situation in which the devices 300a to 300c are all OFF when the user returns home.

[0065] Other examples of inputs to the learning model include time, user behavior, etc. Examples of time include the time of day, day of the week, and date. Alternatively, time-series information may be used as an input to the learning model. Examples of time-series information include a history of device state changes and features calculated from the history, a history of user behavior and features calculated from the history, etc. Furthermore, the output of the learning model may be input again to the learning model to perform recursive predictions.

[0066] The predicted state change may be information indicating the user's behavior (the state of the device 300) and occurrence probability for each predetermined time (scheduled occurrence time), as shown in Fig. 5. The predicted state change shown in Fig. 5 indicates the occurrence probability for a time period less than 30 seconds from a predetermined timing (for example, the time when a plurality of pieces of behavioral information are predicted), the occurrence probability for a time period from 30 seconds to 3 minutes from the predetermined timing, and the occurrence probability for a time period from 1 minute to 5 minutes from the predetermined timing. The occurrence probability for each time period is calculated independently.

[0067] The item of the state change of the device 300 may be set not depending on the state change of all the devices 300, but depending on whether the functions of the devices 300a to 300c in the environment are controlled based on the user's operation.

[0068] For example, if the entrance lamp operates based on a motion sensor, the state of the entrance lamp does not change due to user operation but changes based on the sensing result of the motion sensor, so the state change of the entrance lamp may be deleted from the state change items of the device 300. For example, it may be omitted from the output items by deleting it from the learning data.

[0069] (Determining the details of device control) Next, the determination of the device control content will be described. The control content determination unit 103 determines the device control content corresponding to each of the user's actions based on the user's actions and the predicted time. Specifically, the control content determination unit 103 determines the device control content based on the predicted results of the state change of the device 300 as shown in Figs. 4 and 5 and the proposed rules as shown in Figs. 6A to 6C.

[0070] 6A to 6C are diagrams showing examples of proposed rules.

[0071] The proposed rule associates at least an occurrence probability threshold or an expected occurrence time threshold with a control content. In Figures 6A to 6C, the control content is divided into "notification," "control and notification," and "control," and the threshold is divided into an occurrence probability threshold and an expected occurrence time threshold.

[0072] For example, if the occurrence probability of the predicted user behavior (change in the state of the device 300) is 80% or higher, it is determined to execute a control content associated with the occurrence probability threshold (80% or higher). The occurrence probability threshold (80% or higher) is an example of a first probability range. Furthermore, if the expected time of the predicted user behavior (change in the state of the device 300) is less than 30 seconds, it is determined to execute a control content associated with the expected time threshold (less than 30 seconds). The expected time threshold (less than 30 seconds) is an example of a first period. The "control" shown in FIGS. 6A to 6C is an example of a first control that changes the state of a first device, the state of which changes due to an operation that changes the state of the device, without accepting the operation. It may be determined that the first control is executed when either the occurrence probability or the expected time meets the corresponding threshold, or it may be determined that the first control is executed when both the occurrence probability and the expected time meet the corresponding threshold.

[0073] For example, if the occurrence probability of the predicted user behavior (a change in the state of the device 300) is between 20% and 50%, a decision is made to execute a control detail associated with the occurrence probability threshold (between 20% and 50%). The occurrence probability threshold (between 20% and 50%) is an example of a third probability range. Furthermore, if the expected occurrence time of the predicted user behavior (a change in the state of the device 300) is between 3 minutes and 3 minutes, a decision is made to execute a control detail associated with the expected occurrence time threshold (between 3 minutes and 3 minutes). The expected occurrence time threshold (between 3 minutes and 3 minutes) is an example of a third period. The "notification" shown in FIGS. 6A to 6C is an example of a third control that notifies the user whether or not to execute a first control that changes the state of a first device whose state changes due to an operation that changes the state of the device. It may be decided that the third control is executed when either the occurrence probability or the expected occurrence time meets the corresponding threshold, or it may be decided that the third control is executed when both the occurrence probability and the expected occurrence time meet the corresponding threshold.

[0074] For example, if the occurrence probability of the predicted user behavior (change in the state of the device 300) is between 50% and 80%, a decision is made to execute a control content associated with the occurrence probability threshold (between 50% and 80%). The occurrence probability threshold (between 50% and 80%) is an example of a second probability range. Furthermore, if the expected occurrence time of the predicted user behavior (change in the state of the device 300) is between 30 seconds and 3 minutes, a decision is made to execute a control content associated with the expected occurrence time threshold (between 30 seconds and 3 minutes). The expected occurrence time threshold (between 30 seconds and 3 minutes) is an example of a second period. The "control and notification" shown in FIGS. 6A to 6C is an example of a first control that changes the state of a first device, the state of which changes due to an operation that changes the state of the device, without accepting the operation, and a second control that notifies the user whether to change the first control. In other words, the state of the device 300 is changed, and a notification is given to the user. It may be determined that the first control and the second control are executed when either the occurrence probability or the expected occurrence time meets a corresponding threshold, or it may be determined that the first control and the second control are executed when both the occurrence probability and the expected occurrence time meet their corresponding thresholds. The notification to the user includes a notification that the device control has been executed and a notification as to whether or not to change the executed device control. It is to be noted that the notification as to whether or not to change the device control includes a notification as to whether or not to stop the device control.

[0075] In this way, the control content determination unit 103 determines the control content of the device 400 corresponding to an action whose predicted time falls within the first period to be the first control, determines the control content of the device 400 corresponding to an action whose predicted time falls within the second period after the first period to be the first control and the second control, and determines the control content of the device 400 corresponding to an action whose predicted time falls within the third period after the second period to be the third control. Furthermore, the control content determination unit 103 determines the control content of the device 400 corresponding to an action whose occurrence probability falls within the first probability range to be the first control, determines the control content of the device 400 corresponding to an action whose occurrence probability falls within the second probability range to be the first control and the second control, and determines the control content of the device 400 corresponding to an action whose occurrence probability falls within the third probability range to be the third control. Then, the control instruction unit 104 outputs an instruction to the device 400 to execute the determined control content.

[0076] When the control content indicates a user notification, that is, when the second control or the third control is performed, for example, the current location information of the user and the location information of the device 400 having a display or a speaker may be referenced, and the user may be notified using the device 400 that is closest to the user. In this case, the current location information of the user and the location information of the device 400 having a display or a speaker may be acquired by the sensor 200. When the device 400 having a display or a speaker is fixed, the location information of the device 400 may be recorded in memory.

[0077] Note that notifications to the user may always be sent using a predetermined device. For example, as shown in FIG. 6B, a notification destination device (in other words, a device used for notification) may be specified in advance as a proposed rule. For example, if the predetermined device is a mobile terminal, and if the user's location and the mobile terminal's location are at least a predetermined distance apart, the notification may be sent by a voice reading out the user notification from the mobile terminal. An example of a voice reading out the user notification is, "Do you want to turn on the lamp in the living room?" Furthermore, if the user's location and the mobile terminal's location are less than a predetermined distance apart, the mobile terminal may be used to notify the user by displaying the notification on its display and making a sound indicating that the notification has arrived.

[0078] Furthermore, if the occurrence probability or expected occurrence time falls within a third control threshold for notification, i.e., a third probability range or a third period, a notification is sent to the user asking about control execution, etc., and therefore a user reaction to the notification is required. Therefore, in order to increase the likelihood that the user will notice the notification, the notification may be sent both by displaying it on the display and by a voice reading out the user notification.

[0079] In addition, if the occurrence probability or expected occurrence time falls within the first control threshold for equipment control, i.e., the first probability range or the first period, notification to the user is not required, but a notification indicating that control has been executed after the control has been executed, or information summarizing the control execution contents for a specified period, may be sent.

[0080] 6C, the proposed rule may include a device state condition. If the device state condition is met in addition to each threshold, the control content is determined. For example, if the occurrence probability of the predicted user behavior (state change of device 300) is 20% or more and less than 50%, and further if the living room lamp is OFF, the control content is determined to be the third control. In other words, even if the occurrence probability of the predicted user behavior (state change of device 300) is 20% or more and less than 50%, if the living room lamp is ON, the control content does not have to be determined to be the third control.

[0081] Next, the processing of the device control device 100 configured as above will be described.

[0082] Fig. 7 is a flowchart showing an example of a device control method according to the embodiment. Note that the device control method is a method executed by the device control device 100, and therefore Fig. 7 is also a flowchart showing an example of the operation of the device control device 100 according to the embodiment.

[0083] (Step S101) The information acquiring unit 101 acquires user behavior (state information of the device 300). As an example of user behavior, consider the user returning home. For example, if the device 300 includes a lamp at the entrance and lamps arranged in the house, it is conceivable that all the lamps are OFF for a predetermined period or more while the user is out. Thereafter, when the user returns home, a state in which the entrance lamp is ON and the other lamps are OFF is acquired. In this way, state information (state change) of the device 300 in which the entrance lamp is ON and the other lamps are OFF after all the lamps are OFF for a predetermined period or more, i.e., the user behavior of the user returning home, is acquired. For example, the information acquiring unit 101 may also acquire, as the user behavior, the time when the state in which all the lamps are OFF changes from the state in which all the lamps are OFF to the state in which the entrance lamp is ON and the other lamps are OFF.

[0084] (Step S102) The prediction unit 102 predicts information about a plurality of future user behaviors (state changes of the plurality of devices 300) using a learning model and a user behavior in which all lamps are OFF for a predetermined period of time or more, and then the entrance lamp is ON while the other lamps are OFF. For example, as shown in FIG. 4, state changes of the plurality of devices 300 after the user returns home are predicted. The prediction unit 102 may perform prediction at predetermined time intervals. Examples of the predetermined time intervals include 30 seconds, 1 minute, and 5 minutes. In other words, as shown in FIG. 4, information about state changes of the plurality of devices 300 may be obtained at predetermined time intervals. Alternatively, the prediction unit 102 may perform prediction when the next state change of the device 300 occurs. A new state change of the device 300 is also predicted as the timing when the user will move on to the next behavior, and is therefore considered suitable for executing the control described below.

[0085] (Step S103) The control content determination unit 103 determines the content of device control based on the predicted result of the state change of the device 300 and the proposed rule. Since the entrance light ON in Fig. 4 has an occurrence probability of 100% and an expected occurrence time of less than 30 seconds, which corresponds to the occurrence probability threshold of 80% or more and the expected occurrence time threshold of less than 30 seconds corresponding to the "control" (i.e., the first control) of the proposed rule shown in Fig. 6A, the control content determination unit 103 determines to execute the control of turning on the entrance light. For example, since the expected occurrence time is less than 30 seconds, the control content determination unit 103 determines to execute the control of turning on the entrance light immediately. 4 has an occurrence probability of 70%, and an expected occurrence time of 30 seconds or more and less than 3 minutes, which corresponds to the occurrence probability threshold of 50% or more and less than 80% and the expected occurrence time threshold of 30 seconds or more and less than 3 minutes, which correspond to the "control and notification" (i.e., the first control and the second control) of the proposed rule shown in FIG. 6A, so the control content determination unit 103 decides to execute the control of turning on the bathroom lamp and to issue a notification confirming cancellation of turning on the bathroom lamp. For example, since the expected occurrence time is 30 seconds or more and less than 3 minutes, the control content determination unit 103 decides to execute the control of turning on the bathroom lamp 30 seconds later.

[0086] (Step S104) The control instruction unit 104 outputs the content determined by the control content determination unit 103 to the device 400. For example, the control instruction unit 104 immediately outputs an instruction to turn on the entrance lamp, and 30 seconds later outputs an instruction to turn on the bathroom lamp and a notification to confirm cancellation of turning on the bathroom lamp.

[0087] The control instruction unit 104 may, for example, directly output an ON instruction to each of the entrance lamp and the bathroom lamp, or if the entrance lamp and the bathroom lamp are in an environment controlled by a cloud server, output an ON instruction to the cloud server for the entrance lamp and the bathroom lamp.

[0088] The notification to cancel turning on the bathroom light is to identify device 400 having a speaker closest to the user located at the entrance, and as shown in FIG. 8A, device 400 issues a notification such as "The bathroom light has been turned on. Do you want to cancel it?". Note that if an instruction to cancel the setting is received from the user after the notification by device 400, control instruction unit 104 outputs an instruction to turn off the bathroom light. FIG. 8B also shows a similar display example.

[0089] The device 400 is controlled based on the instruction output by the control instruction unit 104. In this way, future user behavior (e.g., operation of the device 300) is predicted, and device control or device control suggestions are executed according to the prediction accuracy. In particular, by switching between executing automatic control and prompting the user depending on the predicted situation, it becomes possible to execute device control while confirming the user's intention. Furthermore, prompting the user includes confirmation of whether to execute control and whether to cancel it. For example, if the user's response is made a necessary condition for executing control, it is possible to reduce the possibility that the user will be annoyed by having to make many responses. Note that, although an example has been described in which the control content is determined based on the expected occurrence time and occurrence probability, the occurrence probability does not necessarily have to be used when determining the control content.

[0090] When receiving a cancellation instruction from the user, the control instruction unit 104 may update the parameters of the proposed rules shown in Figures 6A to 6C. For example, the parameter may be updated from 50% or more and less than 80% to 60% or more and less than 80%.

[0091] Although the example has been described in which, after control that changes the state of device 400 is performed, notification as to whether or not to cancel the control is given, notification other than whether or not to cancel the control may be given. For example, notification may be given as to whether or not to perform control different from the control that changed the state of device 400. For example, a lamp may be controlled to be turned on at a certain brightness, and notification may be given as to whether or not to change the brightness to a different brightness.

[0092] Furthermore, the control instruction unit 104 may simultaneously execute the control of the devices 400 corresponding to a plurality of actions whose predicted times to be performed are each within a predetermined time range. In other words, the control of the devices 400 corresponding to actions whose predicted times are close to each other may be executed together at the same time.

[0093] (Device 400 Notification) Device 400 may have a display unit, and notification information may be presented by the display unit. In the case of the third control, in which only notification is provided, the notification information may include an indication indicating that device control is executable and an execution indication. For example, the execution indication is a button or icon, etc., that causes device control to be executed, and an instruction to execute device control is output when the execution indication (button, icon, etc.) is operated. In the case of the second control, in which control and notification are provided, the notification information may include an indication indicating that control has been executed and a cancel indication. For example, the cancel indication is a button or icon, etc., that causes the executed device control to be canceled, and an instruction to cancel the executed device control is output when the cancel indication (button, icon, etc.) is operated. An example of the display unit is a display or a speaker. When the display unit is a display, the display unit displays text information as shown in FIGS. 8A to 8D. The "YES" indication shown in FIGS. 8A to 8C and the "Cancel" indication shown in FIG. 8D are examples of the cancel indication. If the presentation unit is a speaker, the presentation unit outputs audio information such as "The bathroom light has been turned on. Do you want to cancel it?" If multiple control contents to be executed at the same timing are determined, the presentation unit may notify the multiple control contents collectively, as shown in Figures 8C and 8D.

[0094] (Variation) The control content determination unit 103 may modify the selected control content using the proposed rules shown in FIGS. 6A to 6C.

[0095] Fig. 9 is a flowchart showing an example of a device control method according to a modified example of the embodiment. Like Fig. 7, Fig. 9 is also a flowchart showing an example of the operation of the device control device 100 according to a modified example of the embodiment. The flowchart shown in Fig. 9 includes steps S1031 to S1034 instead of step S103 in the flowchart shown in Fig. 7. S101, S102, and S104 in the flowchart shown in Fig. 9 are the same as those in Fig. 7, and therefore will not be described here.

[0096] (S1031) The control content determination unit 103 selects the content of device control based on the proposed rule and the predicted result of the state change of the device 300. The method of selecting the content of device control is the same as the method of determining the content of device control in S103 of the flowchart shown in FIG.

[0097] (S1032) The control content determination unit 103 refers to the modification rule and determines whether or not to modify the selected control content. If the control content is to be modified, the process proceeds to S1033, and if the control content is not to be modified, the process proceeds to S1034. The modification rule will be described in detail later.

[0098] (S1033) The control content determination unit 103 refers to the modification rule and modifies the selected control content.

[0099] (S1034) If modification is necessary based on the modification rule, the control content determination unit 103 determines the modified content as the control content.If modification is not necessary based on the modification rule, the control content determination unit 103 determines the content selected based on the proposed rule as the control content.

[0100] Next, the modification rules will be described. For example, there are a first modification rule, a second modification rule, and a third modification rule.

[0101] The first modification rule indicates that the control content will not be determined or executed within a predetermined time after the control content is determined or executed. For example, if the same control content is selected again before a predetermined time has elapsed since the control content was determined, the control content may be modified to not execute the selected control content, and the selected control content may not be executed (in other words, the modified control content, not executing the selected control content, may be executed). However, if the user moves to a location different from the location where the control content was executed immediately before, the first modification rule may not be applied.

[0102] The second modification rule indicates that when the control content selected based on a predetermined input is a notification (i.e., the third control), the notification is not executed. The predetermined input is, for example, the user's actions such as waking up or returning home. For example, after waking up or returning home, the user often performs predetermined actions in succession, and frequent notifications may be annoying. Therefore, in such a case, the modification is made to not execute the third control, which is notification only, and the selected control content (the third control) does not have to be executed. In other words, in such a case, control and notification (i.e., the second control), or only control (i.e., the first control) may be executed.

[0103] The third modification rule indicates that the selected control content is not executed if the predicted content changes little over time. For example, when a multi-step prediction is performed, if there is no change in the predicted content for a predetermined number of steps among the predicted multiple steps, the control content may not be executed. For example, when a 10-step prediction is performed, if there is no change in the prediction for three or more steps, the control content may not be executed. Furthermore, the state change of the device 300 may be weighted more heavily when the expected time of the state change is close to the present than when the expected time of the state change is far from the present. For example, a coefficient of 1.0 may be assigned to the change in the first three steps, a coefficient of 0.5 may be assigned to the change in the fourth through tenth steps, the coefficient of the changed step may be multiplied by 1, and the coefficient of the unchanged step may be multiplied by 0. Whether the change in the predicted content is little may be determined based on whether the sum of the multiplications for each step is equal to or greater than a predetermined value.

[0104] As described above, the control details of the device 400 are determined based on the user's multiple future actions (state changes of the device 300) and the predicted time for each action (each state change), and are executed. Therefore, the user's reaction is not an essential condition for executing the control of the device 400, and the control of the device 400 can be executed automatically. Furthermore, because the control details of the device 400 to be executed are determined based on the user's multiple future actions (state changes of the device 300) and the predicted time for each action (each state change), it is more likely that the automatically executed control of the device 400 will be the control intended by the user. In this way, the device 400 can be controlled effectively.

[0105] (Other embodiments) While the device control method and device control device 100 according to one or more aspects of the present disclosure have been described above based on the embodiments, the present disclosure is not limited to these embodiments. As long as they do not deviate from the spirit of the present disclosure, various modifications conceivable by those skilled in the art to each embodiment and configurations constructed by combining components of different embodiments may also be included within the scope of one or more aspects of the present disclosure.

[0106] For example, the present disclosure can be realized as a program that causes a processor (computer) to execute steps included in a device control method. Furthermore, the present disclosure can be realized as a non-transitory computer-readable recording medium, such as a CD-ROM, on which the program is recorded.

[0107] For example, when the present disclosure is realized as a program (software), each step is performed by running the program using hardware resources such as a computer's CPU, memory, input / output circuits, etc. In other words, each step is performed by the CPU acquiring data from memory or input / output circuits, etc., performing calculations on the data, and outputting the calculation results to memory or input / output circuits, etc.

[0108] In the above embodiment, each component included in the device control device 100 may be configured with dedicated hardware, or may be realized by executing a software program suitable for each component. Each component may also be realized by a program execution unit such as a CPU or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory.

[0109] Some or all of the functions of the device control device 100 according to the above embodiment are typically realized as an LSI, which is an integrated circuit. These may be implemented individually on a single chip, or some or all of them may be integrated on a single chip. Furthermore, the integrated circuit is not limited to an LSI, and may be implemented using a dedicated circuit or a general-purpose processor. An FPGA (Field Programmable Gate Array), which can be programmed after LSI manufacture, or a reconfigurable processor, which can reconfigure the connections and settings of circuit cells within an LSI, may also be used. [Industrial Applicability]

[0110] The present disclosure can be applied to a system that automatically controls electronic devices in a facility. [Explanation of symbols]

[0111] 1. Equipment control system 100 Equipment control device 101 Information Acquisition Department 102 Prediction Department 103 Control content determination unit 104 Control instruction section 200 sensors 300, 300a, 300b, 300c equipment 400, 400a, 400b, 400c equipment 1001 processor 1002 memory

Claims

1. Obtain the user's location information, determining control details for the air conditioner based on the location information of the user; Execute the control content of the air conditioner, The control content of the air conditioner is either control to turn on the power of the air conditioner or control to turn off the power of the air conditioner. Equipment control methods.

2. Get the temperature sensor information, acquiring information on whether the air conditioner is on or off based on information from the temperature sensor; The device control method according to claim 1 .

3. Get the temperature sensor information, acquiring temperature setting information for the air conditioner based on information from the temperature sensor; The device control method according to claim 1 .

4. Further acquiring temperature information in the environment by the temperature sensor; determining whether the temperature information and the temperature setting information of the air conditioner satisfy predetermined conditions; If the predetermined conditions are met, determining the control content of the air conditioner; The device control method according to claim 3.

5. Temperature sensors acquire temperature information in the environment, Acquire the status of the air conditioner; determining whether the temperature information and the state of the air conditioner satisfy predetermined conditions; If the predetermined conditions are met, determining the control content of the air conditioner; The device control method according to claim 1 .

6. Acquire status information of the air conditioners in the environment and operation details of the air conditioners; determining whether the status information of the air conditioner and the operation details satisfies a predetermined condition; If the predetermined conditions are met, determining the control content of the air conditioner; The device control method according to claim 1 .

7. Furthermore, the control content of the air conditioner is determined based on the information on whether the air conditioner is on or off or the setting information. The device control method according to claim 3.

8. predicting the user's behavior based on the user's location information; determining control details for the air conditioner based on the behavior prediction; The device control method according to claim 1 .

9. notifying the user that the control content of the air conditioner has been executed; The device control method according to claim 1 .

10. the user's location information is information indicating that the user is returning home, The control content of the air conditioner is control to turn on the power supply of the air conditioner. The device control method according to claim 1 .

11. a processor and a memory, The processor: Obtain the user's location information, determining control details for the air conditioner based on the location information of the user; Execute the control content of the air conditioner, The control content of the air conditioner is either control to turn on the power of the air conditioner or control to turn off the power of the air conditioner. Equipment control device.

Images