Method and device for estimating sleep-related time of air conditioner user

Abstract

To provide a method and a device for estimating a user's sleeping related time on an air conditioner.SOLUTION: A device for estimating a user's sleeping related time on an air conditioner can be configured so as to acquire an illuminance information log on a room, which is an air conditioning target of an air conditioner over a first predetermined period in the past, estimate a sleeping related time related to the user's sleeping in the room on the basis of the illuminance information log, and output notification information on the sleeping related time and an operation of the air conditioner to an information terminal. Also, the device can be configured so as to output a control instruction by which the air conditioner executes operation on the basis of the sleeping related time to the air conditioner.SELECTED DRAWING: Figure 1A

Description

[Technical Field]

[0001] The present disclosure relates to a method and apparatus for estimating a sleep-related time for a user of an air conditioner. [Background technology]

[0002] Conventionally, a behavior analysis device that analyzes a user's daily behavior is known, as described in Patent Document 1. This behavior analysis device analyzes the user's daily behavior based on learning behavior record data generated by the user's recording. Based on the analysis results, the behavior analysis device provides the user with content such as advertisements that are tailored to the user's lifestyle. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6338984 specification Summary of the Invention [Problem to be solved by the invention]

[0004] Conventional behavior analysis devices analyze a user's daily activities by using learning behavior record data generated by the user as training data and performing supervised learning on estimation conditions for daily activities spanning multiple time periods. However, accurately estimating daily activities with finer time granularity requires input of more detailed behavior records, which increases the burden on the user. Furthermore, the analyzed daily activities are not used to operate specific home appliances, such as air conditioners.

[0005] An object of the present disclosure is to provide a method and device for automatically estimating the sleep-related time of an air conditioner user. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems, the present disclosure provides a method and device for automatically estimating a sleep-related time for a user of an air conditioner.

[0007] A method for estimating a sleep-related time for a user of an air conditioner according to one aspect of the present disclosure includes the steps of acquiring an illuminance information log for a room that is the target of air conditioning control by the air conditioner over a first predetermined period in the past, estimating a sleep-related time related to the user's sleep in the room based on the illuminance information log, and outputting notification information regarding the sleep-related time and the operation of the air conditioner to an information terminal.

[0008] Another aspect of the method for estimating a sleep-related time for a user of an air conditioner according to the present disclosure includes the steps of acquiring an illuminance information log for a room that is the target of air conditioning control by the air conditioner over a first predetermined period in the past, estimating a sleep-related time related to the user's sleep in the room based on the illuminance information log, and outputting a control instruction to the air conditioner to operate based on the sleep-related time.

[0009] A device for estimating a sleep-related time for a user of an air conditioner according to one aspect of the present disclosure is configured to acquire an illuminance information log for a room that is the target of air conditioning control by the air conditioner over a first predetermined period in the past, estimate a sleep-related time related to the user's sleep in the room based on the illuminance information log, and output notification information regarding the sleep-related time and the operation of the air conditioner to an information terminal.

[0010] Another aspect of the present disclosure of an apparatus for estimating a user's sleep-related time of an air conditioner is configured to acquire an illuminance information log for a room that is the target of air conditioning control by the air conditioner over a past first predetermined period, estimate a sleep-related time related to the user's sleep in the room based on the illuminance information log, and output a control instruction to the air conditioner to operate based on the sleep-related time. [Effects of the Invention]

[0011] According to the present disclosure, a method and device for estimating a bedtime-related time for a user of an air conditioner can automatically estimate the bedtime-related time based on an illuminance information log. Furthermore, the estimation method and device can output a notification or control instruction related to the operation of the air conditioner based on the estimation result. [Brief explanation of the drawings]

[0012] [Figure 1A] 1 is a flowchart illustrating an overall example of a method for estimating a bedtime-related time according to the present disclosure. [Figure 1B] FIG. 1 is a schematic diagram illustrating an overall example of a method for estimating a bedtime-related time according to the present disclosure. [Figure 1C] FIG. 1 is a schematic diagram illustrating an example of a bedtime-related time according to the present disclosure. [Figure 2] FIG. 1 is a block diagram illustrating an example of a schematic configuration related to an air conditioner according to the present disclosure. [Figure 3] 1 is a flowchart showing an example of a method for estimating a sleep-related time according to the first embodiment. [Figure 4A] An example of a user interface for notification settings in the first embodiment [Figure 4B] An example of a user interface for PUSH notification in the first embodiment [Figure 5] An example of a notification user interface in the first embodiment [Figure 6] 10 is a flowchart showing an example of a method for estimating a sleep-related time according to the second embodiment. [Figure 7] Flowchart showing an example of step S200 (setting process) in the fourth embodiment [Figure 8] FIG. 13 is a schematic diagram showing an example of a first illuminance threshold value according to the fourth embodiment. [Figure 9] FIG. 13 is a schematic diagram showing an example of a second illuminance threshold value according to the fourth embodiment. [Figure 10] FIG. 10 is a schematic diagram showing an example of an illuminance increase threshold value according to the fourth embodiment. [Figure 11]Flowchart showing an example of step S300 (determination process) in the fourth embodiment [Figure 12] Flowchart showing an example of step S330 in the fourth embodiment [Figure 13A] Flowchart showing an example of a determination process in the fourth embodiment [Figure 13B] Flowchart showing an example of a determination process in the fourth embodiment [Figure 14A] FIG. 10 is a schematic diagram showing an example of label creation in the fourth embodiment. [Figure 14B] FIG. 10 is a schematic diagram showing an example of pairing in the fourth embodiment. [Figure 14C] FIG. 10 is a schematic diagram showing an example of pairing in the fourth embodiment. [Figure 14D] FIG. 10 is a schematic diagram showing an example of a sleep-related determination time in the fourth embodiment. [Figure 15] A flowchart showing an example of step S400 (estimation process) in the fourth embodiment. [Figure 16] Flowchart showing an example of step S430 in the fourth embodiment [Figure 17] 10 is a flowchart showing an example of an estimation process according to the fourth embodiment. [Figure 18A] 10 is a flowchart showing an example of estimation using the overall median in the fourth embodiment. [Figure 18B] 10 is a flowchart showing an example of estimation using a median value by day of the week in the fourth embodiment. [Figure 19A] FIG. 10 is a schematic diagram showing an example of error calculation using the overall median in the fourth embodiment. [Figure 19B] FIG. 10 is a schematic diagram showing an example of error calculation using the median value by day of the week in the fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] First, various aspects of a method and apparatus for estimating a sleep-related time for a user of an air conditioner will be described.

[0014] A method for estimating a sleep-related time for a user of an air conditioner according to a first aspect of the present disclosure includes the steps of acquiring an illuminance information log for a room that is the target of air conditioning control by the air conditioner over a first predetermined period in the past, estimating a sleep-related time for the user to fall asleep in the room based on the illuminance information log, and outputting notification information regarding the sleep-related time and the operation of the air conditioner to an information terminal.

[0015] In a method for estimating a sleep-related time of a user of an air conditioner according to a second aspect of the present disclosure, in the first aspect, the sleep-related time may include an estimated last entry time of the user last entering the room before going to sleep. In the output step, the estimated last entry time and the notification information may be output to an information terminal a first predetermined time before the estimated last entry time.

[0016] A third aspect of the present disclosure provides a method for estimating a sleep-related time for a user of an air conditioner, including the steps of acquiring an illuminance information log for a room that is the target of air conditioning control by the air conditioner over a first predetermined period in the past, estimating a sleep-related time for the user to fall asleep in the room based on the illuminance information log, and outputting a control instruction to the air conditioner to operate based on the sleep-related time.

[0017] A fourth aspect of the present disclosure provides a method for estimating a sleep-related time of a user of an air conditioner in the third aspect, wherein the sleep-related time may include an estimated final entry time of the user's final entry into the room before going to sleep. In the output step, a control instruction to start the air conditioner may be output to the air conditioner a first predetermined time before the estimated final entry time.

[0018] A fifth aspect of the present disclosure provides a method for estimating a bedtime-related time of a user of an air conditioner, wherein the bedtime-related time may include an estimated bedtime start time when the lights in the room are finally turned off before the user goes to sleep. In the output step, if it is determined that the air conditioner is operating when a second predetermined time has elapsed from the estimated bedtime start time, a control instruction to switch the air conditioner to a sleep-related mode may be output to the air conditioner.

[0019] A method for estimating a sleep-related time of a user of an air conditioner according to a sixth aspect of the present disclosure is any one of the first to fifth aspects, wherein the illuminance information log may include time-series log data of illuminance values ​​in the room. The method for estimating a sleep-related time may further include the steps of: setting a first illuminance threshold for determining whether the room is in an unlit state based on the number of data items corresponding to each illuminance value over a first time slot related to sleep time in the illuminance information log within a first predetermined time period; and setting a second illuminance threshold for determining whether the room is in an illuminance state after an increase in the illuminance value over a second time slot related to before sleep in the illuminance information log within the first predetermined time period.

[0020] A method for estimating a sleep-related time of a user of an air conditioner according to a seventh aspect of the present disclosure may be any one of the first to sixth aspects, further including the steps of: acquiring a first illuminance threshold and a second illuminance threshold set based on an illuminance information log within a first predetermined period; determining at least one of a light-on time and a light-off time for an illuminance information log for a third time slot associated with nighttime within a past second predetermined period using the first illuminance threshold and the second illuminance threshold; and determining at least one of a last entry determination time and a sleep-start determination time for each of the third time slot within the second predetermined period based on the determined light-on time or light-off time. In the step of estimating the sleep-related time, the sleep-related time may be estimated based on the last entry determination time or the sleep-start determination time.

[0021] In the method for estimating a bedtime-related time for a user of an air conditioner of an eighth aspect according to the present disclosure, in the seventh aspect, the step of determining at least one of the final entry determination time and the bedtime start determination time may include the steps of creating a pair of determined light-on times and determined light-off times that are consecutive in chronological order, and setting the light-on time of the latest pair of pairs in a third time period as the final entry determination time for the third time period, and setting the light-off time of the latest pair as the bedtime start determination time for the third time period.

[0022] A ninth aspect of the present disclosure is a method for estimating a bedtime-related time of a user of an air conditioner, and in any one of the first to eighth aspects, the step of estimating the bedtime-related time may include a step of acquiring a portion of the bedtime-related judgment times determined based on the illuminance information log that corresponds to a third predetermined period in the past, a step of calculating the median of the acquired bedtime-related judgment times, and a step of setting the median of the bedtime-related judgment times as the bedtime-related time.

[0023] A tenth aspect of the present disclosure provides a method for estimating a bed-related time of a user of an air conditioner according to the ninth aspect, wherein the step of calculating a median of the acquired bed-related determination times may include calculating an overall median and a median for each day of the week of the acquired bed-related determination times. The step of determining the median of the bed-related determination times as the bed-related time may include the steps of calculating an error between the overall median of the bed-related determination times and the determined bed-related determination time, and an error between the median of the bed-related times by day of the week and the determined bed-related determination time, and determining the one with the smaller error as the bed-related time, either the overall median of the bed-related determination times or the median of the bed-related times by day of the week.

[0024] An eleventh aspect of the present disclosure of an apparatus for estimating a user's sleep-related time of an air conditioner is configured to acquire an illuminance information log of a room that is the target of air conditioning control by the air conditioner over a first predetermined period in the past, estimate a sleep-related time related to the user's sleep in the room based on the illuminance information log, and output notification information regarding the sleep-related time and the operation of the air conditioner to an information terminal.

[0025] In a twelfth aspect of the present disclosure, in the device for estimating a bedtime-related time of a user of an air conditioner, in an eleventh aspect, the bedtime-related time may include an estimated final entry time of the user's final entry into the room before going to sleep. The device for estimating the bedtime-related time may be further configured, when outputting, to output the estimated final entry time and notification information to an information terminal a first predetermined time before the estimated final entry time.

[0026] A thirteenth aspect of the present disclosure of an apparatus for estimating a user's sleep-related time of an air conditioner is configured to acquire an illuminance information log of a room that is the target of air conditioning control by the air conditioner over a first predetermined period in the past, estimate a sleep-related time related to the user's sleep in the room based on the illuminance information log, and output a control instruction to the air conditioner to operate based on the sleep-related time.

[0027] In a fourteenth aspect of the present disclosure, in a thirteenth aspect of the device for estimating a bedtime-related time of a user of an air conditioner, the bedtime-related time may include an estimated final entry time of the user's final entry into the room before going to sleep. The device for estimating the bedtime-related time may be further configured, when outputting, to output a control instruction to the air conditioner to start the air conditioner a first predetermined time before the estimated final entry time.

[0028] In a fifteenth aspect of the present disclosure, in a thirteenth aspect of the device for estimating a bedtime-related time of a user of an air conditioner, the bedtime-related time may include an estimated bedtime start time when the lights in the room are finally turned off before the user goes to sleep. The device for estimating the bedtime-related time may be further configured to output a control instruction to the air conditioner to switch the air conditioner to a sleep-related mode when it is determined that the air conditioner is operating when a second predetermined time has elapsed from the estimated bedtime start time.

[0029] A sixteenth aspect of the present disclosure provides a device for estimating a sleep-related time of a user of an air conditioner, wherein the illuminance information log may include time-series log data of illuminance values ​​in the room. The device for estimating a sleep-related time may be further configured to: set a first illuminance threshold for determining whether the room is in an off state based on the number of data items corresponding to each illuminance value over a first time slot related to sleep time in the illuminance information log within a first predetermined time period; and set a second illuminance threshold for determining whether the room is in an on state based on an illuminance value after an increase when the illuminance value increases over a second time slot related to before sleeping in the illuminance information log within the first predetermined time period.

[0030] A device for estimating a bed-related time of a user of an air conditioner according to a seventeenth aspect of the present disclosure may be further configured, in any one of the eleventh to sixteenth aspects, to acquire a first illuminance threshold and a second illuminance threshold set based on an illuminance information log within a first predetermined period, determine at least one of a light-on time and a light-off time for an illuminance information log for a third time slot associated with nighttime within a past second predetermined period using the first illuminance threshold and the second illuminance threshold, and determine at least one of a last entry determination time and a sleep-start determination time for each of the third time slot within the second predetermined period based on the determined light-on time or light-off time. The device for estimating a bed-related time may be further configured, when estimating a bed-related time, to estimate the bed-related time based on the last entry determination time or the bed-start determination time.

[0031] An 18th aspect of the device for estimating a bedtime-related time for a user of an air conditioner according to the present disclosure is that in the 17th aspect, the device for estimating a bedtime-related time can be further configured to, when determining at least one of the final entry determination time and the bedtime start determination time, create a pair of determined light-on times and determined light-off times that are consecutive in chronological order, and to set the light-on time of the latest pair of pairs in a third time period as the final entry determination time for the third time period, and to set the light-off time of the latest pair as the bedtime start determination time for the third time period.

[0032] A device for estimating a bedtime-related time of a user of an air conditioner of a 19th aspect of the present disclosure may be further configured, in any one of the 11th to 18th aspects, to, when estimating a bedtime-related time, acquire a portion of the bedtime-related judgment times determined based on the illuminance information log that corresponds to a third predetermined period in the past, calculate a median of the acquired bedtime-related judgment times, and set the median of the bedtime-related judgment times as the bedtime-related time.

[0033] A device for estimating a bed-related time of a user of an air conditioner according to a twentieth aspect of the present disclosure may be further configured, in a 19th aspect, to calculate an overall median and a day-of-week median of the acquired bed-related determination times when calculating the median of the acquired bed-related determination times. When the median of the bed-related determination times is set as the bed-related time, the device for estimating a bed-related time may be further configured to calculate an error between the overall median of the bed-related determination times and the determined bed-related determination time, and an error between the day-of-week median of the bed-related times and the determined bed-related determination time, and to set the bed-related time to the overall median of the bed-related determination times or the day-of-week median of the bed-related times with the smaller error.

[0034] Each of the embodiments described below represents an example of the present disclosure or a collection of multiple examples. The numerical values, shapes, configurations, steps, and step orders shown in each of the following embodiments are examples and do not limit the present disclosure. Among the components in the following embodiment 1, components that are not recited in the independent claims showing the highest concept are described as optional components.

[0035] In each of the embodiments described below, certain elements may be modified, and other elements may be appropriately combined with any configuration, and the combined configurations will provide the respective effects. In the embodiments, the respective combinations of the respective modified configurations will provide the respective effects of the respective modified configurations.

[0036] In the following detailed description, terms such as "first," "second," etc. are used for descriptive purposes only and should not be understood as expressing or implying the relative importance or ranking of technical features. Features qualified as "first" and "second" expressly or imply the inclusion of one or more of that feature.

[0037] Overall Technical Concept Before describing specific embodiments of the method for estimating the sleep-related time of a user of an air conditioner (hereinafter also referred to as the estimation method) and the device for estimating the sleep-related time of a user of an air conditioner (hereinafter also referred to as the estimation device) according to the present disclosure, we will first explain the technical concept described in the present disclosure using one example.

[0038] The estimation method and estimation device of the present disclosure are used to estimate the bedtime-related time of a user of an air conditioner. One scenario in which the estimation method and estimation device of the present disclosure are used is when the user usually sleeps at night in a room where an air conditioner is installed, i.e., a room that is subject to air conditioning control by the air conditioner. The bedtime-related time refers to a time related to the user's sleep in the room. For example, the bedtime-related time may include the time when the user enters the room to go to bed, the time when the user turns off the lights in the room to go to bed, or a time calculated based on these times.

[0039] The estimation method and estimation device estimate a user's sleep-related time based on the illuminance information log of the room where the user sleeps, and further output notifications or control instructions related to the operation of the air conditioner based on the estimation results.

[0040] FIG. 1A is a flowchart illustrating an example of the overall estimation method, and FIG. 1B is a schematic diagram illustrating an example of the overall estimation method. The estimation device acquires a room illuminance information log covering a predetermined past period (step S100). That is, log data corresponding to a first predetermined period is acquired from the room illuminance information log. Next, the estimation device sets an illuminance threshold for determining whether the room is lit or off (step S200, setting process). Based on the set illuminance threshold, the estimation device determines a bedtime-related determination time for log data of a specific time period related to nighttime (step S300, determination process). Based on the past bedtime-related determination time, the estimation device estimates the user's future bedtime-related time (step S400, estimation process). Then, based on the estimated bedtime-related time, the estimation device outputs a notification or control instruction related to the operation of the air conditioner to the user's information terminal or the air conditioner (step S500, output process).

[0041] This exemplary estimation method includes a learning phase and an execution phase. The learning phase includes a setting process and a determination process, and the learned determination results can be stored in a database. The execution phase includes an estimation process and an output process. In the execution phase, at least a portion of the determination results stored in the database are read to estimate a future bedtime-related time for the user.

[0042] In this example, the sleep-related time includes at least one of a last entry estimated time when the user finally enters the room before going to sleep and an estimated start-to-sleep time when the user finally turns off the lights in the room before going to sleep. FIG. 1C is a schematic diagram showing an example of the sleep-related time in this example. In the example of FIG. 1C, the illuminance information log includes time-series log data of the illuminance value of the room, and the illuminance value is a numerical value detected by an illuminance sensor installed in the room. As shown in FIG. 1C, based on the illuminance value, it can be determined whether the lighting device in the room is on (on state) or off (off state) during the time period to be estimated (i.e., the specific time period related to nighttime described above). It can also be determined when the lighting device was turned on (on time) or when it was turned off (off time).

[0043] Each process can be summarized as follows:

[0044] In the setting process, the estimation device sets an illuminance threshold value for determining whether the room to be subject to air conditioning control is in a lit or unlit state, based on the illuminance information log of the room.

[0045] In the determination process, the estimation device determines the lights-on time and the lights-off time. The estimation device further determines the last lights-on time in the time period to be estimated as the last room entry determination time, and the last lights-off time in the time period to be estimated as the sleep start determination time.

[0046] In the estimation process, the estimation device estimates the user's future estimated last entry time and estimated sleep start time based on the past last entry determination time and sleep start determination time.

[0047] In the output process, the estimation device outputs a notification or control instruction related to the operation of the air conditioner based on the estimated last entry determination time and / or bedtime start determination time.

[0048] As will be described later, the learning phase and the execution phase can be executed separately. The estimation device may execute only the estimation process and the output process as long as it can acquire the pre-stored determination results. Furthermore, the setting process, determination process, estimation process, and output process do not need to be executed consecutively, and can be executed independently of each other. For example, in the learning phase, the estimation device can execute the determination process independently as long as it can acquire the illuminance information log and the pre-stored setting results. Therefore, the estimation method may include only the estimation process and the output process, or may further include at least one of the setting process and the determination process.

[0049] <<Overall outline>> FIG. 2 is a block diagram showing an example of a schematic configuration related to an air conditioner according to the present disclosure.

[0050] 2, the air conditioner 20 includes an air conditioning memory unit 21, an air conditioning control unit 22, and an air conditioning communication unit 23. The air conditioner 20 may further include at least one temperature sensor 24 to perform its functions. The air conditioner 20 may also include an illuminance sensor 25 to detect the illuminance in a room.

[0051] The air conditioner 20 can be connected to the server 10 and / or the information terminal 30 via the air conditioning communication unit 23. For example, the air conditioner 20 may be connected to the server 10 related to the air conditioner 20 and / or to an information terminal 30 such as a user's smartphone via the Internet.

[0052] The estimation device according to the present disclosure may be the server 10, the air conditioning control unit 22 of the air conditioner 20, or the information terminal 30. In the following embodiments, the server 10 is mainly described as operating as the estimation device, but this is not limited to this. When the air conditioning control unit 22 operates as the estimation device, the air conditioner 20, upon receiving a control instruction from the estimation device, appropriately performs operation in accordance with the control instruction.

[0053] Below, an overview of each component will be explained.

[0054] <Air conditioner 20> The air conditioner 20 controls the air conditioning of, for example, one room in a home, one room in an apartment or condominium, or one room for a single person. In particular, it controls the air conditioning of the room where the user usually sleeps at night. The air conditioner 20 has a cooling function and a heating function, and may also have a humidifying function, a dehumidifying function, and / or an air purifying function. These functions and operating modes can be freely combined (for example, cooling and dehumidifying function, cooling and ventilation mode, etc.).

[0055] <Air conditioning memory unit 21> The air conditioning storage unit 21 is a recording medium that records various information and control programs, and may be a memory that functions as a work area for the air conditioning control unit 22. The air conditioning storage unit 21 is realized, for example, by a flash memory, a RAM (Random Access Memory), a ROM (Read Only Memory), other storage devices, or an appropriate combination of these.

[0056] When the air conditioning control unit 22 operates as an estimation device, the air conditioning storage unit 21 may store a room illuminance information log, or may acquire the illuminance information log from the server 10. When executing the estimation method, the air conditioning control unit 22 may read out only the necessary amount of log data from the illuminance information log stored in the air conditioning storage unit 21 or the server storage unit 12.

[0057] <Air conditioning control unit 22> The air conditioning control unit 22 is a controller that controls at least some of the functions of the air conditioner 20. The air conditioning control unit 22 includes a general-purpose processor such as a CPU, MPU, MCU, FPGA, DSP, or ASIC that executes programs to achieve predetermined functions. The air conditioning control unit 22 can implement various controls in the air conditioner 20 by calling and executing control programs stored in the air conditioning storage unit 21. The air conditioning control unit 22 also works in cooperation with the air conditioning storage unit 21 to read and write data stored in the air conditioning storage unit 21.

[0058] The air conditioning control unit 22 can receive various commands and setting values ​​(e.g., a command to start a specific operation mode, a command to set the temperature) from the server 10 or the information terminal 30 via the air conditioning communication unit 23. The air conditioning control unit 22 controls each component of the air conditioner 20 to perform the cooling function or heating function of the air conditioner 20 based on these setting values ​​and detected values ​​(e.g., indoor temperature) received from various sensors (e.g., temperature sensor 24). Furthermore, when the air conditioning control unit 22 operates as an estimation device, the air conditioning control unit 22 controls the operation of the air conditioner 20 based on the estimation results.

[0059] <Air Conditioning Communication Unit 23> The air conditioning communication unit 23 can also communicate with the server 10, a user's information terminal 30, etc., and can, for example, send and receive Internet packets. As described above, the air conditioning control unit 22 may cooperate with the server 10 and / or the information terminal 30 via the air conditioning communication unit 23. The air conditioning communication unit 23 may communicate and send and receive data between the server 10, the air conditioner 20, and the information terminal 30 in accordance with standards such as Wi-Fi (registered trademark), IEEE802.2, IEEE802.3, 3G, and LTE. The air conditioning communication unit 23 may communicate via the Internet, an intranet, an extranet, a LAN, ISDN, a VAN, a CATV communication network, a virtual private network, a telephone line network, a mobile communication network, a satellite communication network, infrared rays, or Bluetooth (registered trademark).

[0060] The air conditioning control unit 22 can transmit various data stored in the air conditioning memory unit 21, such as detected values ​​by various sensors, its own operating record, and control history, to the server 10 via the air conditioning communication unit 23. The air conditioning control unit 22 can also transmit detected values ​​by various sensors directly to the server 10 via the air conditioning communication unit 23.

[0061] <Temperature sensor 24> The temperature sensor 24 may include, for example, an indoor temperature sensor for detecting the indoor temperature of the room and an outdoor temperature sensor for detecting the outdoor air temperature. Information detected by the temperature sensor 24 is input and stored in the air conditioning storage unit 21, and is later used by the air conditioning control unit 22 or transmitted to the information terminal 30 or the server 10.

[0062] <Illuminance sensor 25> The illuminance sensor 25 is a sensor for detecting the illuminance received in the room. The illuminance sensor 25 converts light incident on a light-receiving element, for example, light from a lighting device 40 in the room, into an electric current to detect the ambient brightness. The illuminance sensor 25 performs detection at regular intervals, and the detected illuminance value and the date and time of detection may be stored in the air conditioning storage unit 21. The illuminance value detected by the illuminance sensor 25 is input and stored in the air conditioning storage unit 21, and is later used by the air conditioning control unit 22 or transmitted to the information terminal 30 or the server 10.

[0063] In the embodiment of FIG. 2 , the illuminance sensor 25 is mounted on the air conditioner 20, but this is not limiting. The illuminance sensor 25 may be mounted not only on the main body of the air conditioner 20 but also on, for example, another home appliance in the room or any location in the room that can receive light from the lighting device 40, or may be an independent sensor device. When performing the estimation method, the estimation device can acquire information used for estimation from the illuminance sensor 25 regardless of the location of the illuminance sensor 25. For example, when the server 10 operates as the estimation device, the server 10 may acquire illuminance values ​​from the illuminance sensor 25 via the air conditioner 20 or another home appliance, or may acquire illuminance values ​​directly from the illuminance sensor 25. When the information terminal 30 operates as the estimation device, the information terminal 30 may acquire illuminance values ​​or illuminance information logs from the illuminance sensor 25 via the server 10, or may acquire illuminance values ​​directly from the illuminance sensor 25 to create an illuminance information log.

[0064] The air conditioner 20 may further include other sensors, such as a humidity sensor and a user activity sensor. These sensors may perform detection at regular intervals, and the detected values ​​may be stored in the air conditioning storage unit 21.

[0065] <Information terminal 30> The information terminal 30 is an information terminal capable of performing data communication with the air conditioner 20. For example, the information terminal 30 may be a smartphone, cellular phone, mobile phone, tablet, wearable device, computer, etc. of the user of the air conditioner 20, equipped with a dedicated associated application 32. When the information terminal 30 operates as an estimation device, the information terminal 30 may acquire an illuminance information log from the server 10 when executing the estimation method, or may read an illuminance information log stored in the information terminal 30.

[0066] The server 10 or the air conditioning control unit 22 can acquire settings or commands input by a user via the information terminal 30. Typically, the information terminal 30 includes a display for displaying a graphical user interface (GUI). However, when interacting with a user via a voice user interface (VUI), the information terminal 30 may include a speaker and a microphone instead of or in addition to a display.

[0067] In one embodiment, when the information terminal 30 receives notification information regarding a bedtime-related time and the operation of the air conditioner 20 from the estimation device, the information terminal 30 presents the bedtime-related time and notification information to the user via a UI. In another embodiment, the information terminal 30 creates a presentation time and a presentation notification content based on the received bedtime-related time and notification information, and then presents the presentation time and notification content to the user. The presentation time and notification content may be different from the received bedtime-related time and notification information.

[0068] <Lighting device 40> The lighting device 40 is installed in a room that is subject to air conditioning control by the air conditioner 20, and is a device that provides a light source to the room. The light source has a certain level of illuminance, and preferably has illuminance sufficient for reading and daily activities in particular. In terms of the built-in light source of the lighting device 40, it may be a light bulb (incandescent lamp) fixture, a fluorescent lamp fixture, or a high intensity discharge (HID) fixture. In terms of the style of installation in the room, the lighting device 40 may be a pendant, ceiling lamp, chandelier, spotlight, floor lamp, or table lamp.

[0069] In one embodiment, the lighting device 40 can be connected to the server 10 and / or the information terminal 30 via the Internet or other home appliances. For example, the lighting device 40 may transmit its own switch-on / switch-off information and / or the amount or luminous intensity of light it emits to the server 10 via the Internet.

[0070] <Server 10> The server 10 is a server for managing and controlling at least one air conditioner 20, but may also be used for other purposes. For example, the server 10 may be a management server of a manufacturer of the air conditioner 20 for managing at least one air conditioner 20 or for collecting data. Alternatively, the server 10 may be an application server. The server 10 may be implemented as the estimation device of the present disclosure. In the present disclosure, the server 10 includes a server storage unit 12 and a server control unit 14. The server 10 may further include a server communication unit 16 for communicating with the air conditioner 20 or the information terminal 30.

[0071] <Server storage unit 12> The server storage unit 12 is a recording medium that records various information and control programs, and may be a memory that functions as a work area for the server control unit 14. The server storage unit 12 is realized by, for example, a flash memory, an SSD (Solid State Device), a hard disk, RAM, ROM, or other storage device, or an appropriate combination of these. The server storage unit 12 may be a memory inside the server 10, or may be a storage device connected to the server 10 by wireless or wired communication.

[0072] The server storage unit 12 may store information received from the air conditioner 20 and various sensors as log data. When the server 10 operates as an estimation device, upon receiving illuminance information, the server 10 writes the received illuminance information and the date and time of detection to an illuminance information log stored in the server storage unit 12. That is, the server 10 may write information received from the illuminance sensor 25 or the lighting device 40 to the illuminance information log. The illuminance information log may include, for each fixed period of time, the illuminance value detected by the illuminance sensor 25, switch-on / switch-off information of the lighting device 40, and / or the amount of light emitted by the lighting device 40. When executing the estimation method, the server control unit 14 may read out only the necessary amount of log data from the illuminance information log stored in the server storage unit 12.

[0073] The server storage unit 12 may store information for executing the estimation method. For example, the air conditioning storage unit 21 may store at least one threshold value set (learned) in the setting process, at least one judgment condition available in the judgment process, the judgment result in the judgment process, and the estimation result in the estimation process.

[0074] The server storage unit 12 may also store a program for causing an estimation device (for example, the server 10) to execute the estimation method.

[0075] <Server control unit 14> The server control unit 14 of the server 10 is a controller that controls the entire server 10. The server control unit 14 includes a general-purpose processor such as a CPU, MPU, GPU, FPGA, DSP, or ASIC that executes programs to realize predetermined functions. The server control unit 14 can realize various controls in the server 10 by calling and executing control programs stored in the server storage unit 12. The server control unit 14 can also read and write data stored in the server storage unit 12 in cooperation with the server storage unit 12. The server control unit 14 is not limited to a unit that realizes predetermined functions through cooperation between hardware and software, and may be a hardware circuit designed specifically to realize the predetermined functions.

[0076] <Server Communication Unit 16> The server communication unit 16 cooperates with the server control unit 14 to send and receive internet packets, i.e., communicate, with the air conditioner 20 and the information terminal 30. For example, the server 10 may receive commands from the information terminal 30 via the server communication unit 16, or may send instructions to the air conditioner 20. The server communication unit 16 or the air conditioning communication unit 23 may communicate and send and receive data between the server 10, the air conditioner 20, and the information terminal 30 in accordance with standards such as Wi-Fi (registered trademark), IEEE802.2, IEEE802.3, 3G, and LTE. In addition to the internet, communication may also be with an intranet, an extranet, a LAN, ISDN, a VAN, a CATV communication network, a virtual private network, a telephone line network, a mobile communication network, a satellite communication network, infrared rays, or Bluetooth (registered trademark).

[0077] First Embodiment <Method for estimating output of notification information related to operation of air conditioner 20> Hereinafter, a first embodiment of an estimation method and an estimation device according to the present disclosure will be described in detail with reference to the accompanying drawings. Hereinafter, the estimation method and estimation device according to the present disclosure will be described using a mode in which a server 10 is used as the estimation device, but the present disclosure is not limited to this. The estimation method and estimation device (server 10) estimates a sleep-related time and outputs notification information related to the operation of an air conditioner 20 to an information terminal 30 based on the estimation result.

[0078] As described above, the bedtime-related time refers to a time related to the user's sleep in the room. For example, the bedtime-related time may include an estimated final entry time when the user finally enters the room before going to sleep and / or an estimated bedtime start time when the user finally turns off the lights in the room before going to sleep.

[0079] 3 is a flowchart showing an example of a method for estimating a sleep-related time according to Embodiment 1. The estimation method according to Embodiment 1 includes steps S100, S400A, and S510. That is, the estimation method according to Embodiment 1 does not include a setting process and a determination process.

[0080] The server control unit 14 acquires, from the server storage unit 12, illuminance information logs for rooms that are the target of air conditioning control by the air conditioner 20 over a past first predetermined period (step S100). More specifically, the server control unit 14 reads out, from the server storage unit 12, log data of the illuminance information logs that correspond to the first predetermined period.

[0081] The first predetermined period may be determined based on the amount of data required to estimate the bedtime-related time accurately. For example, the first predetermined period may be 14 days, 20 days, 28 days, 30 days, 35 days, or 42 days. Also, by way of example, the first predetermined period refers to, but is not limited to, the first predetermined period immediately preceding the execution of the estimation method.

[0082] In one embodiment, the first predetermined period may be set according to the frequency of execution of the estimation method. For example, if the server control unit 14 estimates the bedtime-related time weekly and the determination process or estimation process described below uses a median value by day of the week or a mean absolute error for each day of the week, the first predetermined period may be a multiple of seven days.

[0083] Then, the server control unit 14 estimates the sleep-related time based on the illuminance information log (step S400A). The server control unit 14 reads at least a portion of the determination results that have already been determined and stored in the server storage unit 12 from the server storage unit 12. For example, the server control unit 14 may read only a portion of the stored determination results that corresponds to the same first predetermined period. Alternatively, the server control unit 14 may read a portion of the determination results that corresponds to another predetermined period different from the first predetermined period.

[0084] In step S400A, the server control unit 14 further estimates the sleep-related time based on the illuminance information log acquired in step S100 and the determination result acquired in step S400A. Step S400A in Fig. 3 and step S400 in Fig. 1A may be completely the same, or the same estimation process may be performed after the determination result required for estimation is acquired by different means.

[0085] Note that estimating the bedtime-related time based on the illuminance information log includes obtaining a determination result based on the illuminance information log and estimating the bedtime-related time based on the determination result. If the bedtime-related time can be estimated based only on the obtained determination result, performing step S100 to obtain the illuminance information log may be omitted.

[0086] In the output process, the server control unit 14 outputs the bedtime-related time and notification information related to the operation of the air conditioner 20 to the information terminal 30 (step S510). Examples of the estimated and output bedtime-related time may be, for example, "estimated last entry time: May 24th, 23:00" or "estimated bedtime start time: May 24th, 23:30."

[0087] Regarding the notification information, in one embodiment, the estimation device (server 10) creates notification information in the output process based on the indoor environment information of the room and the current operating status of the air conditioner 20. In one embodiment, when it is determined that the indoor environment of the room is not comfortable because the air conditioner 20 is not operating a first predetermined time before the estimated last entry time, the notification information indicates this. In step S510, the server control unit 14 outputs the estimated last entry time and the notification information to the information terminal 30 a first predetermined time before the estimated last entry time. This notification information can prompt the user to start the air conditioner 20.

[0088] Whether the indoor environment is comfortable can be determined, for example, by comparing the indoor temperature of the room with the set temperature. In one embodiment, the indoor environment information includes the indoor temperature. The server control unit 14 acquires the indoor temperature of the room and the set temperature of the air conditioner 20, and determines that the indoor environment is not comfortable when the difference between them is equal to or greater than a predetermined temperature threshold. The set temperature may be a user-set temperature input by a user, or an internal set temperature that the air conditioner 20 actually operates at for air conditioning control. In one embodiment, the indoor environment information further includes indoor humidity or indoor air quality. The server control unit 14 determines whether the indoor environment is comfortable by taking into account other parameters besides temperature, such as the difference between the indoor humidity and the set humidity, and the difference between the indoor air quality and the set air quality.

[0089] The first predetermined time is a time considered necessary to make the current indoor environment of the room comfortable. For example, the first predetermined time is the time required to lower the indoor temperature by cooling operation or the time required to raise the indoor temperature by heating operation. For example, the first predetermined time may be 15 minutes, 30 minutes, 60 minutes, or 90 minutes. To make the indoor environment of the room comfortable by the time the user goes to sleep, the server control unit 14 determines whether the room is comfortable and outputs notification information the first predetermined time before the estimated last entry time.

[0090] When the information terminal 30 receives the bedtime-related time and notification information from the estimation device, the information terminal 30 presents the bedtime-related time and notification information to the user via a push notification or the UI of the related application 32.

[0091] FIG. 4A shows an example of a user interface for notification settings. Screen SC1 shown in FIG. 4A displays a notification settings screen. Display area R1 displays the comfortable temperature range set by the user. Display area R2 displays the scheduled date and time for the next notification (e.g., notification for the current day or the following day). The displayed time is the first predetermined time before the estimated last entry time. Display area R3 displays a toggle button for enabling / disabling the notification function.

[0092] If the notification function is disabled by the user, even if the server control unit 14 determines that "the difference between the room temperature and the set temperature is equal to or greater than the predetermined temperature threshold and the air conditioner is not operating" a first predetermined time before the estimated time of last entry, the server control unit 14 will not output notification information. Alternatively, in this case, the server control unit 14 outputs notification information to the information terminal 30, but the information terminal 30 does not present the notification information to the user.

[0093] When the notification function is enabled, the server control unit 14 queries the air conditioner 20 for operation information of the air conditioner 20 and indoor environment information of the room a first predetermined time before the estimated last entry time. Suppose that, based on the received operation information and indoor environment information, it is determined that "the difference between the room temperature and the set temperature is equal to or greater than a predetermined temperature threshold, and the air conditioner is not operating." In this case, the server control unit 14 may output notification information indicating that the room is not comfortable, and may further include a message urging the user to start the air conditioner 20 in the notification information.

[0094] 4B is an example of a user interface for a push notification. In the screen SC2 shown in FIG. 4B, a push notification is displayed in the display area R6, which reads, "The room temperature before going to sleep is outside the comfortable temperature range." When the user clicks on this push notification, the information terminal 30 may transition from the current display screen to another screen displaying more detailed information, or may launch a related application 32 to display more detailed information.

[0095] In one embodiment, as described below, the estimation device outputs a control instruction to the air conditioner 20 based on the estimated bedtime start time. The air conditioner 20 controls its own operation based on the control instruction, and, for example, automatically switches to sleep mode at the received estimated bedtime start time. This automatic switching function may also be displayed on the screen SC1 shown in FIG. 4A. For example, a toggle button for enabling / disabling the automatic switching function is displayed in the display area R4 of FIG. 4A. The time when the automatic switching is scheduled to occur, i.e., the estimated bedtime start time, is displayed in the display area R5 of FIG. 4A.

[0096] The estimation process of the estimation method may be performed every day to estimate the bedtime-related time for the current day, or may collectively estimate and store the bedtime-related times for several future days (including the current day). For example, the server control unit 14 may estimate the bedtime-related times every day, every three days, every seven days, or every ten days. When estimating for several days, the server control unit 14 may output only the bedtime-related times and notification information for the current day to the information terminal 30 in the output process. Alternatively, the server control unit 14 may collectively output the estimated bedtime-related times to the information terminal 30, and further output the notification information for the current day to the information terminal 30.

[0097] In one embodiment, if the setting process and / or the determination process are still learning, a message to that effect is presented to the user via the information terminal 30. For example, if an appropriate threshold has not yet been set in the setting process, or if the number of determination results available for estimation is still insufficient, the server control unit 14 causes the information terminal 30 to present a message indicating "learning." FIG. 5 shows an example of a notification user interface in the first embodiment. On the GUI screen SC3 of the information terminal 30 shown in FIG. 5, the words "learning" are displayed in the display area R7, but the bedtime-related time and notification information are not displayed.

[0098] When the air conditioning control unit 22 or the information terminal 30 is executed as an estimation device, the estimation device acquires an illuminance information log from the server 10 or the air conditioner 20, performs an estimation method, and outputs bedtime-related times, etc. to the information terminal 30.

[0099] In one embodiment, the estimation device (e.g., the air conditioning control unit 22, the server 10, or the information terminal 30) has a program used to execute the ventilation control method as described above. The program causes the estimation device to execute the estimation method.

[0100] The estimation method and estimation device disclosed herein can acquire an illuminance information log for a room that is the target of air conditioning control, and automatically estimate a user's sleep-related time based on the acquired illuminance information log. Therefore, the estimation method and estimation device can accurately estimate the estimated last room entry time or the estimated sleep start time without requiring any input. Furthermore, the estimation method and estimation device can output a notification related to the operation of an air conditioner to an information terminal based on the estimation result, urging the user to operate the air conditioner appropriately.

[0101] Second Embodiment <Method for outputting control instructions based on estimated last entry time> In the second embodiment, the estimation method and estimation device estimate a sleep-related time and output a control instruction to the air conditioner 20 to operate based on the estimation result to the air conditioner 20. Below, the estimation method and estimation device of the present disclosure will be described using an aspect in which the server 10 executes as the estimation device, but the present disclosure is not limited to this.

[0102] First, it is preferable to obtain the user's consent in advance because the estimation device will perform automatic control on the air conditioner 20. For example, the user's consent can be obtained in advance by using a toggle button for switching between enabling and disabling the automatic control function, as in the display area R4 of the setting screen SC1 shown in Fig. 4A.

[0103] 6 is a flowchart showing an example of a method for estimating a sleep-related time according to Embodiment 2. The estimation method according to Embodiment 2 includes steps S100, S400A, and S520. That is, the estimation method according to Embodiment 2 does not perform a setting process or a determination process.

[0104] The server control unit 14 estimates the bed-related time by executing steps S100 and S400A similar to those in the first embodiment. The difference from the first embodiment is in the output process. In the second embodiment, the server control unit 14 outputs a control instruction to the air conditioner 20 to operate the air conditioner 20 based on the bed-related time (step S520). The bed-related time in the second embodiment includes the estimated final entry time of the user's final entry into the room before going to sleep.

[0105] In step S520, the server control unit 14 may output a control instruction to the air conditioner 20 to start the air conditioner 20 a first predetermined time before the estimated last entry time. For example, if the server control unit 14 determines that the indoor environment of the room is not comfortable, it outputs a control instruction to operate the air conditioner 20 so as to make the indoor environment of the room comfortable before the bedtime-related time is reached. More specifically, the server control unit 14 outputs a control instruction when it determines that the difference between the indoor temperature or indoor humidity of the room and the set temperature or set humidity is equal to or greater than a predetermined threshold and that the air conditioner 20 is not operating.

[0106] The control instruction may include, in addition to starting the air conditioner 20, a designation of an operation mode (for example, cooling mode, heating mode, or dehumidification mode), a designation of a set temperature, and / or a designation of a set humidity. When the air conditioning control unit 22 receives a control instruction from the server 10 via the air conditioning communication unit 23, it controls the air conditioning of the room in accordance with the control instruction to make the indoor environment of the room comfortable before the bedtime-related time is reached.

[0107] In one embodiment, if the user has given prior consent to the automatic control function, the server control unit 14 determines whether to output a control instruction based on the user's location. When the server control unit 14 determines that the conditions for issuing a control instruction as described above are met, it acquires the user's location. If the server control unit 14 determines that the user's location is not within a predetermined range centered on the room location, it does not output a control instruction to the air conditioner 20. In other words, if it determines that the user is not in the room or near the room, the server control unit 14 does not automatically start the air conditioner 20.

[0108] In one embodiment, even if the user has previously given consent for the automatic control function, the server control unit 14 attempts to obtain the user's consent for the control instruction before outputting the control instruction. The server control unit 14 displays the content of the control instruction, such as "Start the air conditioner," to the user via the associated application 32 of the information terminal 30, prompting the user for consent. After receiving consent from the user via the information terminal 30, the server control unit 14 outputs the control instruction to the air conditioner 20.

[0109] This completes the process of estimating the estimated last entry time and outputting control instructions based on the estimation result. The estimation device can automatically estimate the estimated last entry time based on the acquired illuminance information log, and can automatically control the operation of the air conditioner based on the estimation result. If the indoor environment of the room is not comfortable, the estimation device can automatically start the air conditioner a first predetermined time before the estimated last entry time. Therefore, the estimation device can appropriately control the air conditioner to make the room comfortable before the user finally enters the room and goes to sleep, thereby promoting a good night's sleep for the user.

[0110] Third Embodiment <Method for outputting control instructions based on estimated bedtime start time> In the third embodiment, similarly to the second embodiment, the estimation method and estimation device estimate a sleep-related time and output a control instruction to the air conditioner 20 to operate the air conditioner 20 based on the estimation result. The estimation method in the third embodiment does not perform a setting process or a determination process. Below, the estimation method and estimation device of the present disclosure will be described using an aspect in which the server 10 executes the estimation device, but the present disclosure is not limited to this.

[0111] The estimation method and estimation device in the third embodiment estimate the sleep-related time by executing steps S100 to S520 similar to those in the second embodiment, but the content of the control instructions to be output and the timing of output are different. As in the second embodiment, the estimation device automatically controls the air conditioner 20, so it is preferable to obtain the user's consent in advance.

[0112] The bedtime-related time in the third embodiment includes an estimated bedtime start time when the lights in the room are finally turned off before the user goes to sleep. In step S520, the server control unit 14 may output a control instruction to the air conditioner 20 to switch the operation of the air conditioner 20 to the sleep-related mode when a second predetermined time has elapsed since the estimated bedtime start time. The second predetermined time may be, for example, 0 hours, half an hour, 1 hour, 2 hours, or 3 hours.

[0113] The sleep-related mode is one of the operation modes of the air conditioner 20, which aims to create a comfortable sleeping environment and promote good sleep for the user. In the sleep-related mode, the air conditioning control unit 22 may control the air conditioning in accordance with the user's sleep activity. For example, in the sleep-related mode, the air conditioning control unit 22 may control the air conditioning to keep the room temperature between 26°C and 28°C so that the room temperature does not drop too much. In addition, in the sleep-related mode, the air conditioning control unit 22 may control the direction of the airflow blown out of the air conditioner 20 so that the airflow does not directly hit the user. The sleep-related mode is also called a good night's sleep mode or a good sleep mode.

[0114] In one embodiment, when a second predetermined time has elapsed since the estimated bedtime start time, the server control unit 14 queries the air conditioner 20 for its operation information. When it determines that the air conditioner 20 is operating based on the received operation information, it outputs a control instruction to the air conditioner 20 to switch the air conditioner 20 to a sleep-related mode.

[0115] In one embodiment, when a second predetermined time has elapsed since the estimated bedtime start time, the server control unit 14 determines whether to output a control instruction based on the room's on / off state. If the room is still lit even after the second predetermined time has elapsed since the estimated bedtime start time, the user may not actually be asleep yet. Therefore, when the server control unit 14 determines that the second predetermined time has elapsed since the estimated bedtime start time and that the room is lit, the server control unit 14 does not output a control instruction to switch to the sleep-related mode, or delays the output. If the output is delayed, the server control unit 14 determines the room state again after a certain time has elapsed. The server control unit 14 outputs a control instruction to switch to the sleep-related mode only when it determines that the room is lit.

[0116] In another embodiment, the air conditioner 20 includes a human presence sensor. The server control unit 14 can determine via the human presence sensor whether the user is in the room or asleep. If the server control unit 14 determines that a second predetermined time has elapsed since the estimated bedtime start time and that the user is not in the room, the server control unit 14 does not output, or delays output of, a control instruction to switch to a sleep-related mode.

[0117] To determine whether the room is lit or unlit, the server control unit 14 may acquire the latest illuminance value from the illuminance sensor 25 and an illuminance threshold for determining whether the room is lit or unlit. The illuminance threshold may include, for example, a first illuminance threshold for determining whether the room is unlit and / or a second illuminance threshold for determining whether the room is lit, both of which are set in advance in a setting process. If the latest acquired illuminance value is equal to or less than the first illuminance threshold or is not higher than the second illuminance value, the room is determined to be in an unlit state. If the latest acquired illuminance value is not lower than the first illuminance threshold or is equal to or greater than the second illuminance value, the room is determined to be in an lit state.

[0118] To determine the lit / unlit state of the room, the server control unit 14 may obtain switch-on / switch-off information of the lighting device 40. In this case, the server control unit 14 queries the lighting device 40 for its switch-on / switch-off information. If the lighting device 40 is switched on, the room is determined to be in a lit state, and if the lighting device 40 is switched off, the room is determined to be in an unlit state.

[0119] In this way, the operation of the air conditioner 20 can be automatically switched to the sleep-related mode at a more accurate timing.

[0120] In one embodiment, in step S520, the server control unit 14 outputs a control instruction to the air conditioner 20 to stop operation of the air conditioner 20, instead of a control instruction to switch the air conditioner 20 to a sleep-related mode. Automatically stopping operation helps to save energy in the air conditioner 20. Note that, since automatic control is performed on the air conditioner 20, it is preferable to obtain the user's consent in advance.

[0121] To execute the automatic operation shutdown function more safely, the server control unit 14 determines whether to output a control instruction based on the outside air temperature or external information related to the outside air temperature. The server control unit 14 may acquire the outside temperature of the room from an external temperature sensor of the air conditioner 20. The server control unit 14 may also acquire the current or future outside air temperature from an external information source (e.g., a weather information source) that can be connected to the server 10 via the Internet.

[0122] The server control unit 14 outputs a control instruction to the air conditioner 20 to stop operation of the air conditioner 20 only when it determines that the acquired outside air temperature is within a predetermined temperature range. In other words, the air conditioner 20 stops operation only when the outside air temperature is neither too hot nor too cold. On the other hand, when the server control unit 14 determines that a second predetermined time has passed since the estimated bedtime start time and that the outside air temperature is outside the predetermined temperature range, it outputs a control instruction to switch the air conditioner 20 to a sleep-related mode.

[0123] This completes the process of estimating the estimated bedtime start time and outputting control instructions based on the estimation result. The estimation device can automatically estimate the estimated last entry time based on the acquired illuminance information log, and can automatically control the operation of the air conditioner based on the estimation result. After the estimated bedtime start time, the estimation device can automatically switch the operation of the air conditioner to a sleep-related mode or stop it. Therefore, the estimation device can appropriately control the air conditioner and reduce power consumption while maintaining an indoor environment suitable for sleep.

[0124] Fourth Embodiment <Estimation of bedtime-related times> In the fourth embodiment, the setting process, determination process, and estimation process (FIG. 1B) for estimating the sleep-related time of a user in a room that is the target of air conditioning control will be described in detail. As described in the "Overall Technical Concept," the estimation method may include a learning phase and an execution phase. The learning phase may include a setting process and a determination process, and the learned determination results may be stored in a database. The execution phase may include an estimation process and an output process. The setting process, determination process, estimation process, and output process do not need to be executed consecutively, but can be executed independently of each other.

[0125] Here is a brief overview of each process:

[0126] In the setting process, the estimation device sets an illuminance threshold value for determining whether the room to be subject to air conditioning control is in a lit or unlit state, based on the illuminance information log of the room.

[0127] In the determination process, the estimation device determines the light-on time and the light-off time based on the log data of the illuminance information log. The estimation device further determines the last light-on time in the time period to be estimated as the last room entry determination time, and the last light-off time in the time period to be estimated as the sleep start determination time.

[0128] In the estimation process, the estimation device estimates the user's future estimated last entry time and estimated sleep start time based on the past last entry determination time and sleep start determination time.

[0129] In the output process, the estimation device outputs a notification or control instruction related to the operation of the air conditioner based on the estimated last entry determination time and / or sleep start determination time. That is, the sleep-related times estimated in the estimation process can be used in the output process to notify the user or for automatic control of the air conditioner 20, as in the above-mentioned first to third embodiments.

[0130] The setting process, the determination process, and the estimation process will be described in more detail below. These processes will be described using a mode in which the server 10 executes them as an estimation device, but the present invention is not limited to this.

[0131] <Setup process> In the setting process, the illuminance threshold for determining whether the light is on or off is set based on the illuminance information log including time-series log data of the illuminance value of the room. This allows an appropriate illuminance threshold to be set for the room, regardless of whether the room has a window or the type of lighting device 40.

[0132] If the illuminance information log includes log data of the lighting device 40 being switched on / off, the server control unit 14 of the server 10 can directly determine the on / off state of the room based on the log data of the lighting device 40. Therefore, in this case, it is not necessary to execute the setting process.

[0133] 7 is a flowchart showing an example of step S200 (setting process) in the fourth embodiment. The server control unit 14 sets an illuminance threshold based on the illuminance information log of the room over a first predetermined period acquired in step S100 (FIG. 1A). More specifically, the server control unit 14 sets a first illuminance threshold for determining whether the room is in a lights-out state based on the number of data items corresponding to each illuminance value over a first time period in the acquired illuminance information log (step S210). In a later determination process, if the illuminance value is equal to or less than the first illuminance threshold, it can be determined that the room is in a lights-out state at that time.

[0134] Here, the first time period is a time period related to sleep hours, i.e., a time period during which the lights in a room are generally considered to be off. For example, the first time period may be "00:00 to 06:00," "01:00 to 05:00," or "02:00 to 04:00." The first time period may be set depending on the region where the room is located or the season in which the illuminance value is detected.

[0135] The log data of illuminance values ​​is a collection of illuminance values ​​detected every certain time period (e.g., every 3 minutes, 5 minutes, 10 minutes). Therefore, the number of data items corresponding to each illuminance value over the first time period indicates how frequently the room is at the corresponding illuminance value during the first time period.

[0136] FIG. 8 is a schematic diagram showing an example of the first illuminance threshold in the fourth embodiment. By creating a histogram of illuminance values ​​as shown in FIG. 8 using time-series log data of illuminance values, the number of data items corresponding to each illuminance value can be determined. In general, during a first time slot when the lights in the room are likely to be off, there are many data items corresponding to relatively low illuminance values, while there are few data items corresponding to relatively high illuminance values. In one embodiment, in step S210, the server control unit 14 may set the illuminance value corresponding to the largest number of corresponding data items as the first illuminance threshold. In another embodiment, the server control unit 14 may set the average value of multiple (e.g., three or five) illuminance values ​​corresponding to the largest number of corresponding data items as the first illuminance threshold.

[0137] Furthermore, the server control unit 14 sets a second illuminance threshold for determining whether the room is in a lit state based on the illuminance values ​​over the second time period in the illuminance information log for the first predetermined period (step S220). In a later determination process, if the illuminance value is equal to or greater than the second illuminance threshold, it can be determined that the room is in a lit state at that time.

[0138] In one embodiment, the server control unit 14 sets the second illuminance threshold based on the illuminance value after an increase when the illuminance value increases over the second time period in the illuminance information log within the first predetermined period. For example, the server control unit 14 sets the illuminance value with the largest number of corresponding data items among the illuminance values ​​after an increase as the second illuminance threshold.

[0139] In another embodiment, the server control unit 14 sets the second illuminance threshold based on the illuminance value after the illuminance value rises from below the first illuminance threshold during the second time period. That is, in this embodiment, step S220 is performed after step S210.

[0140] The second time period is a time period associated with before going to bed, i.e., a time period during which a room is generally considered to be lit at night. In particular, a time period during which people frequently enter a room at night and turn on the lights is preferable. For example, the second time period may be "19:00 to 03:00," "21:00 to 03:00," or "23:00 to 04:00." The second time period may be set depending on the region where the room is located and the season in which the illuminance value is detected.

[0141] In one embodiment, the server control unit 14 excludes a portion of the illuminance values ​​with low values ​​after the increase (for example, the lowest 5%, 10%, or 15% of the data items), and sets the lowest illuminance value among the remaining illuminance values ​​as the second illuminance threshold. FIG. 9 is a schematic diagram showing an example of the second illuminance threshold in the fourth embodiment. The upper half of FIG. 9 shows the illuminance values ​​over a second time period. The dotted rectangles indicate the times within the second time period when the illuminance value increases from below the first illuminance threshold. The statistics of such illuminance values ​​and the corresponding data items are shown in the lower half of FIG. 9. As shown in the lower half of FIG. 9, the portion with the lowest 10% of the data items (the dotted block portion) is excluded, and then the lowest illuminance value is set as the second illuminance threshold.

[0142] In one embodiment, the server control unit 14 further sets an illuminance increase threshold for determining whether the room is in a lit state during the setting process. The server control unit 14 sets the illuminance information threshold based on the illuminance increase value when the illuminance value increases over the second time period (i.e., the increase in the illuminance value between successive preceding and following times in a time series). In the subsequent determination process, if the illuminance increase value is equal to or greater than the illuminance increase threshold, it may be determined that the room is in a lit state at the time of the increase.

[0143] 10 is a schematic diagram showing an example of an illuminance increase threshold in the fourth embodiment. Data with small illuminance increase values ​​is considered to be due to fluctuations or noise in the measurements by the illuminance sensor 25, so the server control unit 14 excludes some of the data with the lowest illuminance increase values. Then, of the remaining illuminance increase values, the server control unit 14 sets the lowest illuminance increase value in the data group containing the illuminance increase value corresponding to the largest number of data items as the illuminance increase threshold. The "data group" here refers to data in which there are one or more consecutive illuminance increase values, and is shown to the right of the dotted line in the lower half of FIG. 10.

[0144] In one embodiment, the portion to be excluded is determined as follows: The server control unit 14 sequentially sets the first target values ​​for searching for the exclusion threshold in the illuminance histogram, starting from the smallest illuminance increase value (illuminance increase value 1) in the direction of increasing illuminance increase values, and determines whether the currently set first target value satisfies a specific condition. That is, illuminance increase values ​​1, 2, 3, etc. are sequentially set as the first target values. Specific condition (1) is that the number of data items corresponding to an illuminance increase value "M+1" that is 1 greater than the current first target value "M" and the number of data items corresponding to an illuminance increase value "M+2" that is 2 greater than the first target value "M" are both 0. Specific condition (2) is that the number of data items corresponding to the illuminance increase value "M+1" is greater than the number of data items corresponding to the first target value "M" and the number of data items corresponding to the illuminance increase value "M+2" is greater than the number of data items corresponding to the illuminance increase value "M+1." That is, the specific condition (2) specifies that the number of corresponding data items must increase from the first target value two consecutive times.

[0145] If the server control unit 14 determines that the specific condition (1) or (2) is satisfied, the server control unit 14 sets the current first target value "M" as the exclusion threshold. Then, the server control unit 14 excludes data corresponding to an illuminance increase value equal to or less than the exclusion threshold. It should be noted that the initial value of the first target value does not have to be 1. Also, the number of consecutive 0s or consecutive increases in the number of data items specified in the specific condition does not have to be 2, and may be, for example, an integer equal to or greater than 3.

[0146] The illuminance increase threshold and the "data group" can also be determined in a similar manner. In one embodiment, the server control unit 14 sets a second target value for searching for the illuminance increase threshold and determines whether the currently set second target value satisfies a specific condition. The server control unit 14 sets the second target values ​​in a histogram of illuminance values ​​after exclusion processing using the exclusion threshold, sequentially from the illuminance increase value with the largest number of data items ("55" in FIG. 10) in a direction of decreasing illuminance increase values. That is, in the embodiment of FIG. 10, illuminance increase values ​​55, 54, 53, etc. are sequentially set as the second target values. The specific condition (3) is that the number of data items corresponding to an illuminance increase value "N-1" that is 1 less than the current second target value "N" and the number of data items corresponding to an illuminance increase value "N-2" that is 2 less than the second target value "N" are both 0. The specific condition (4) is that the number of data items corresponding to the illuminance increase value "N-1" is greater than the number of data items corresponding to the second target value "N," and the number of data items corresponding to the illuminance increase value "N-2" is greater than the number of data items corresponding to the illuminance increase value "N-1." In other words, the specific condition (4) specifies that the number of corresponding data items must increase from the second target value two consecutive times.

[0147] If the server control unit 14 determines that the specific condition (3) or (4) is satisfied, it sets the current second target value "N" as the illuminance increase threshold. The server control unit 14 then sets data corresponding to an illuminance increase value equal to or greater than the illuminance increase threshold as a "data group." It should be noted that the initial value of the second target value does not have to be the maximum value present in the histogram, but may be, for example, a value that is smaller than the maximum value by a predetermined amount. Furthermore, the number of consecutive zeros or consecutive increases in the number of data items specified in the specific condition does not have to be two, but may be, for example, an integer greater than or equal to three.

[0148] The setting of the first illuminance threshold, the second illuminance threshold, and the illuminance increase threshold described in the fourth embodiment may be performed in any order or in parallel. However, in an embodiment in which the second illuminance threshold is set based on the illuminance value when the illuminance value increases from below the first illuminance threshold, step S220 is performed after step S210.

[0149] The server control unit 14 stores the set first illuminance threshold, second illuminance threshold, and / or illuminance rise threshold in the server storage unit 12 or the like. This completes the setting process. In the setting process, the first illuminance threshold, second illuminance threshold, and / or illuminance rise threshold for determining whether the room is on or off are set based on time-series log data of the room's illuminance values. This allows an appropriate illuminance threshold to be set for the room, regardless of the presence or absence of windows in the room or the type of lighting device. For example, even when a bedside light other than the main lighting device 40 is used during sleep, the on / off state of the lighting device 40 can be accurately determined. Furthermore, for example, a case in which the illumination value rises slightly and then rises sharply to become a bright on state may not be detected using only the second illuminance threshold. However, by using the illuminance rise threshold in combination, it is possible to accurately determine that the lighting device 40 is on even in such a case.

[0150] This allows the setting process to set (learn) the first illuminance threshold, second illuminance threshold, and illuminance increase threshold for judgment. These thresholds are set based on the illuminance information log of a specific room, so they can be said to be thresholds optimized for that room. Using such thresholds allows for more accurate determination of the light-on time and sleep-related judgment times.

[0151] <Decision process> The determination process includes a first half for determining the light-on time and the light-off time based on the log data of the illuminance information log, and a second half for determining the last entry determination time and the sleep start determination time. The determination process of the present disclosure can accurately determine the light-on time, the light-off time, the last entry determination time, and the sleep start determination time. Therefore, the results of this determination can later be used to accurately estimate sleep-related times.

[0152] If the illuminance information log includes log data of the lighting device 40 being switched on / off, the server control unit 14 can directly determine the on / off state of the room based on the log data of the lighting device 40. In this case, the first half of the determination process (determining the on time and off time) does not need to be performed.

[0153] 11 is a flowchart showing an example of step S300 (determination process) in the fourth embodiment. First, the server control unit 14 acquires the first illuminance threshold and the second illuminance threshold set based on the illuminance information log within a first predetermined period (step S310). For example, the server control unit 14 reads the first illuminance threshold and the second illuminance threshold from the server storage unit 12. Furthermore, if an illuminance increase threshold is set, the illuminance increase threshold may also be read in step S310.

[0154] Next, the server control unit 14 determines at least one of the light-on time and the light-off time for the light-on information log for the third time slot within the past second predetermined period using the first illuminance threshold and the second illuminance threshold (step S320). In one embodiment, the server control unit 14 creates a light-on label (also referred to as an "ON label") for the determined light-on time, and creates an off label (also referred to as an "OFF label") for the determined light-off time.

[0155] Here, the second predetermined period is a period for defining the target of one processing in the determination process. For example, the second predetermined period may be one day, two days, three days, or seven days. Also, as an example, the second predetermined period refers to the second predetermined period immediately before the execution of the determination process, but is not limited to this.

[0156] The third time period is a time period associated with nighttime, i.e., a time period generally considered to be nighttime. For example, the third time period may be "19:00 to 05:00," "20:00 to 04:00," or "21:00 to 03:00." The third time period may be set depending on the region where the room is located and the season in which the illuminance value is detected.

[0157] Therefore, the illuminance information log for the "third time slot within the second predetermined period" that is the processing target of the determination process is, for example, the illuminance information log for the most recent night (20:00 to 04:00) from the time the determination process is executed. However, the illuminance information log for the "third time slot within the second predetermined period" may contain only the light-on time and the light-off time, or may contain neither. In such cases, the server control unit 14 determines only the light-on time or the light-off time that exists and creates the corresponding label. Furthermore, if the second predetermined period is one day or more, the illuminance information log for the third time slot for each day within the second predetermined period is processed in the determination process.

[0158] The server control unit 14 determines the light-on time and light-off time based on the chronologically consecutive illuminance values ​​of the processing object in accordance with various determination conditions. For example, the following first to third determination conditions can be used.

[0159] <First judgment condition> If it is determined that the previous illuminance value is equal to or less than the first illuminance threshold and the subsequent illuminance value is equal to or greater than the second illuminance threshold, the server control unit 14 sets the time corresponding to the subsequent illuminance value as the lighting time.

[0160] <Second judgment condition> If it is determined that the previous illuminance value is equal to or greater than the second illuminance threshold and the subsequent illuminance value is equal to or less than the first illuminance threshold, the server control unit 14 sets the time corresponding to the subsequent illuminance value as the light-off time.

[0161] <Third judgment condition> If it is determined that the subsequent illuminance value is greater than the illuminance increase threshold value compared to the previous illuminance value, the server control unit 14 sets the time corresponding to the subsequent illuminance value as the lighting time.

[0162] To further improve the accuracy of the determination, the server control unit 14 may exclude determination results (labels) that are unlikely to be related to lights-off immediately before going to sleep. For example, if the lights-off state lasts for a very short time, it is unlikely that the lights-off was an act intended for going to sleep. Therefore, the server control unit 14 may exclude determination results (labels) that satisfy the following fourth determination condition based on the duration of lights-off. The duration of lights-off refers to the length of time the lights-off state lasts, and is formed in the data by a series of times the lights are turned off and times the lights are turned on.

[0163] <Fourth judgment condition> If it is determined that there is a lights-out duration shorter than the lights-out duration threshold, the server control unit 14 excludes a series of lights-out times and lights-on times that form the lights-out duration. The lights-out duration threshold may be, for example, 3 minutes, 5 minutes, or 10 minutes.

[0164] These determination conditions can be used separately or in any combination. The server control unit 14 determines the light-on and light-off times according to these determination conditions and creates the corresponding labels.

[0165] If the illuminance information log includes log data of the lighting device 40 being switched on / off, the server control unit 14 may create a label according to the time of the lighting device 40 being switched on or off recorded in the log data. The server control unit 14 may further process the label created according to the time of the lighting device 40 being switched on or off in accordance with a fourth determination condition.

[0166] Next, the server control unit 14 determines at least one of the last entry determination time and the sleep start determination time of the processing target based on the light-on time or the light-off time (step S330). Note that the processing target is, as described above, the illuminance information log for the third time slot of each day within the second predetermined period. The light-on time or the light-off time used in step S330 may be the time determined in step S320, or may be the switch-on or switch-off time recorded in the log data of the lighting device 40.

[0167] 12 is a flowchart showing an example of step S330 in the fourth embodiment. The server control unit 14 creates a pair of consecutive light-on times and light-off times in chronological order (step S332). That is, the server control unit 14 pairs consecutive ON labels and OFF labels for one illuminance information log for the third time slot.

[0168] In one embodiment, if only an OFF label exists for a specific third time slot, in step S332, the server control unit 14 creates a virtual ON label corresponding to the start time of the third time slot. The server control unit 14 pairs the virtual ON label with an actually existing OFF label.

[0169] Then, the server control unit 14 determines the latest light-on time of the pair of pairs in the third time slot as the final entry determination time for the third time slot, and determines the latest light-off time of the pair as the sleep start determination time for the third time slot (step S334). That is, if the user enters the room and turns on and off lighting device 40 multiple times during the third time slot, and finally turns off lighting device 40, the latest light-on time is determined as the final entry determination time, and the latest light-off time is determined as the sleep start determination time.

[0170] When a virtual ON label is created, the server control unit 14 does not set the time corresponding to the virtual ON label as the final entry determination time. In this case, only the sleep start determination time is determined in step S334.

[0171] The server control unit 14 stores the determined light-on time, light-off time, last entry determination time, and sleep start determination time in the server storage unit 12 or the like. This completes the processing of the determination process. The server control unit 14 can determine the last entry determination time and / or the sleep start determination time for the third time slot of the second predetermined period. Furthermore, by creating a pair of consecutive earlier light-on times and later lights-off times, the determined sleep-related determination times also comply with the order of the earlier last entry determination time and later sleep start determination time.

[0172] Next, another example of the determination process will be introduced. Figures 13A and 13B are flowcharts showing an example of the determination process in embodiment 4. This determination process determines a sleep-related determination time, that is, determines at least one of the last entry determination time and the sleep start determination time.

[0173] In this example, the determination process includes steps S601 to S620. Of these, step S601 corresponds to step S310 in Fig. 11, steps S602 to S606 correspond to step S320 in Fig. 11, and steps S608 to S620 correspond to step S330 in Fig. 11. Steps S608 to S616 and S620 correspond to step S332 in Fig. 12, and step S618 corresponds to step S334 in Fig. 12.

[0174] Please refer to Figures 14A to 14D in addition to Figures 13A and 13B. Figure 14A is a schematic diagram showing an example of label creation in embodiment 4. Figure 14B is a schematic diagram showing an example of pairing in embodiment 4. Figure 14C is a schematic diagram showing an example of pairing in embodiment 4. Figure 14D is a schematic diagram showing an example of a sleep-related determination time in embodiment 4.

[0175] First, the server control unit 14 acquires the first illuminance threshold value and the second illuminance threshold value (step S601), and also acquires an illuminance information log for a third time slot spanning a second predetermined period (step S602). The server control unit 14 determines at least one of the light-on time and the light-off time for the acquired log data in accordance with the first to third determination conditions described above, and creates an ON label and an OFF label (step S604). As shown in FIG. 14A, the ON label and the OFF label are created for the illuminance information log for the third time slot in step S604.

[0176] Next, the server control unit 14 deletes labels before and after a light-off duration shorter than the first time threshold in accordance with the fourth judgment condition described above (step S606). The first time threshold here is the light-off duration threshold described above. That is, in step S606, the server control unit 14 deletes a series of OFF labels and ON labels that form a light-off duration shorter than the first time threshold. As shown in FIG. 14B, the light-off duration enclosed by the ellipse is shorter than the first time threshold, so the OFF labels and ON labels that form that light-off duration are deleted. The deletion result is shown in FIG. 14C.

[0177] Next, the server control unit 14 determines whether or not there is an OFF label in the third time slot (step S608). The absence of an OFF label indicates that there is no bedtime start time in the third time slot, and the determination process is completed. On the other hand, if there is an OFF label in the third time slot, the server control unit 14 further determines whether or not there is an ON label in the third time slot (step S610). If both an OFF label and an ON label exist in the third time slot, the server control unit 14 creates a pair of consecutive ON labels and OFF labels in chronological order (step S612). That is, the server control unit 14 pairs consecutive ON labels and OFF labels.

[0178] To further improve the accuracy of the determination, the server control unit 14 may further delete certain pairs. If the lighting state lasts for a short period of time, for example, a user may temporarily turn on the lighting device 40 for a short time to check their smartphone or go to the bathroom. Because such turning on and off may not be an act of going to bed, it may be possible to delete pairs with relatively short lighting durations. However, if only such pairs exist in the third time slot, deleting the pairs would make it impossible to determine whether or not the lighting state is occurring. Here, the server control unit 14 deletes pairs with relatively short lighting durations only if there are pairs with relatively long lighting durations. The lighting duration refers to the duration of the lighting state, and is formed in the data by a series of lighting times and lighting-off times.

[0179] The server control unit 14 first determines whether there is a pair whose lighting duration is equal to or greater than the second time threshold (step S614). That is, the server control unit 14 determines whether there is a pair whose lighting duration is relatively long. If there is a pair whose lighting duration is equal to or greater than the second time threshold, the server control unit 14 deletes pairs whose lighting duration is shorter than the second time threshold (step S616). On the other hand, if there is no pair whose lighting duration is equal to or greater than the second time threshold, even pairs whose lighting duration is relatively short are used for the determination.

[0180] For example, in FIG. 14C, there are three pairs, pair (1) to pair (3). Suppose the lighting durations of pair (1) and pair (2) are equal to or greater than the second time threshold, and the lighting duration of pair (3) is shorter than the second time threshold. In this case, since there is a pair whose lighting duration is equal to or greater than the second time threshold, the server control unit 14 deletes pair (3). Note that the second time threshold may be, for example, 5 minutes, 10 minutes, 15 minutes, or 30 minutes.

[0181] When the pairing process is completed, the server control unit 14 determines at least one of the last entry determination time and the sleep start determination time using the remaining pairs (step S618). Here, the server control unit 14 may determine the last entry determination time or the sleep start determination time in a manner similar to that of step S334 described above. That is, the server control unit 14 determines the light-on time of the latest pair among the remaining pairs in the third time slot as the last entry determination time for the third time slot, and the light-off time of the latest pair as the sleep start determination time for the third time slot.

[0182] 14D were created in step S612, but pair (3) was deleted in step S616 because its lighting duration was shorter than the second time threshold. Consequently, only pair (1) and pair (2) remain in the third time slot. Because pair (2) is the latest pair remaining in the third time slot, server control unit 14 determines the lighting time of pair (2) as the final entry determination time for the third time slot, and the lighting time of pair (2) as the sleep start determination time for the third time slot.

[0183] If it is determined in steps S608 and S610 that only an OFF label exists in the third time slot, the server control unit 14 executes step S618. That is, the server control unit 14 extracts the latest OFF label in the third time slot (step S620) and determines the sleep start determination time (step S618). More specifically, similar to step S332 described above, the server control unit 14 creates a virtual ON label and pairs the virtual ON label with an actual OFF label to create a virtual pair. Since no other pairs exist in the third time slot, the server control unit 14 determines the sleep-related determination time using the virtual pair. The server control unit 14 extracts the OFF label of the virtual pair in step S620 and determines the lights-out time corresponding to the OFF label extracted in step S618 as the sleep start determination time.

[0184] This completes the determination process. The server control unit 14 can more accurately determine the sleep-related determination time within the third time slot of the second predetermined period. Furthermore, by pairing the ON label with the OFF label, the determined sleep-related determination time does not have a reversed order of occurrence.

[0185] <Estimation process> In the estimation process, the estimation device estimates the user's future estimated last entry time and / or estimated bedtime start time based on the past last entry determination time and past bedtime start determination time. Note that the estimated last entry time and the estimated bedtime start time can be calculated independently.

[0186] As described above, the sleep-related determination time includes the last entry determination time and the sleep start determination time, and the sleep-related time includes the estimated last entry time and the estimated sleep start time. The server control unit 14 may estimate the last entry determination time and the sleep start determination time using a similar method. Hereinafter, the estimation process will be described using the broader concepts of "sleep-related determination time" and "sleep-related time" with reference to FIGS. 15 and 16. However, the "sleep-related determination time" in the description can be replaced with the "last entry determination time," "sleep start determination time," or "last entry determination time and sleep start determination time." Similarly, the "sleep-related determination time" in the description can be replaced with the "last entry estimation time," "estimated sleep start time," or "estimated last entry time and sleep start time."

[0187] 15 is a flowchart showing an example of step S400 (estimation process) in the fourth embodiment. The server control unit 14 acquires sleep-related determination times (step S410). For example, the server control unit 14 reads out the light-on time, the light-off time, the last entry determination time, and the sleep start determination time, which are determined based on the illuminance information log and stored in the server storage unit 12. Note that the server control unit 14 may acquire only the portion used for estimation, i.e., only the portion corresponding to the third predetermined period in the past.

[0188] The third predetermined period may be determined based on the amount of data required to accurately estimate the sleep-related time. Hereinafter, the third predetermined period will be described as 35 days, similar to the first predetermined period, but is not limited to this period. Also, for example, the third predetermined period refers to the third predetermined period immediately preceding the execution of the estimation method, but is not limited to this.

[0189] The server control unit 14 calculates the median of the acquired sleep-related determination times (step S420) and sets the median of the sleep-related determination times as the sleep-related time (step S430). Note that depending on the contents of the illuminance information log, there may be days when the sleep-related determination time (last entry determination time and / or sleep start determination time) cannot be determined. The server control unit 14 may calculate the median by ignoring the sleep-related determination time that cannot be determined.

[0190] For example, the server control unit 14 acquires the last entry determination time for the most recent 35 days in step S410, and calculates the median of the last entry determination time for those 35 days in step S420. If there are no last entry determination times for two days in the most recent 35 days, the server control unit 14 calculates the median of the last entry determination times for the remaining 33 days, and sets this median as the estimated last entry time.

[0191] In one embodiment, to obtain a more accurate estimate, the server control unit 14 calculates and estimates the overall median and the median for each day of the week.

[0192] In this embodiment, in step S420, the server control unit 14 calculates the overall median and the median for each day of the week of the acquired sleep-related determination times. For example, the server control unit 14 calculates the overall median of the last estimated entry time using all of the acquired and existing last estimated entry times. The server control unit 14 also calculates the median for Monday using the sleep-related determination time corresponding to Monday, and similarly calculates the median for each day of the week using the sleep-related determination time corresponding to each day of the week.

[0193] Fig. 16 is a flowchart showing an example of step S430 in this embodiment. The server control unit 14 calculates the error between the overall median of the sleep-related determination times and the actually determined sleep-related determination times. Then, the server control unit 14 calculates the error between the median of the sleep-related times by day of the week and the actually determined sleep-related determination times (step S432). The method of error calculation will be described later with reference to Figs. 18A to 19B.

[0194] Next, the server control unit 14 determines the sleep-related time as the one with the smaller error between the overall median of the sleep-related determination times and the median of the sleep-related times by day of the week (step S434). For example, the server control unit 14 determines the last entry estimated time as the one with the smaller error between the overall median of the last entry determination times and the median by day of the week as the last entry estimated time, and determines the last entry start estimated time as the one with the smaller error between the overall median of the sleep start determination times and the median by day of the week as the last entry start estimated time.

[0195] Next, another example of the estimation process will be introduced. Fig. 17 is a flowchart showing an example of the estimation process in embodiment 4. This estimation process can obtain an estimation result that respects the order of the first estimated last entry time and the second estimated bedtime start time.

[0196] In this example, the estimation process includes steps S702 to S712, in which step S702 is substantially similar to step S410 in Fig. 15, and step S704 corresponds to steps S420 and S430 in Fig. 15.

[0197] In the estimation process of Fig. 17, the server control unit 14 first acquires the sleep-related determination times over the third predetermined period (step S702). Then, as in steps S420 and S430 described above, the server control unit 14 calculates the median of the sleep-related determination times to provisionally estimate the sleep-related times (step S704). In this estimation process, the sleep-related times obtained in step S704 are treated as provisional estimation results. The provisional estimation results are confirmed after confirmation in steps S706 and S708 and are output as final estimation results.

[0198] The server control unit 14 determines whether the provisional estimation result includes both the estimated last entry time and the estimated sleep start time in the third time slot (step S706). If only one of the estimated last entry time and the estimated sleep start time is found, the provisionally estimated sleep-related time is determined as the final estimation result (step S710).

[0199] On the other hand, if both the estimated last entry time and the estimated sleep start time exist, the server control unit 14 further determines whether the estimated last entry time is later than the estimated sleep start time (step S708). If the estimated last entry time is not later than the estimated sleep start time, that is, if no time reversal has occurred in the provisional estimation result, the provisionally estimated sleep-related time is determined as the final estimation result (step S710).

[0200] If the provisional estimation result shows a time reversal in which the estimated last entry time is later than the estimated bedtime start time, the server control unit 14 may provisionally re-estimate the bedtime-related time. For example, the server control unit 14 extracts the bedtime-related determination times for days on which both the last entry determination time and the bedtime start determination time are present over the third predetermined period (step S712) and provisionally re-estimates the bedtime-related time using the extracted data (step S704). In other words, the bedtime-related determination times for days on which only the last entry determination time and only the bedtime start determination time are present are not used in the provisional re-estimate.

[0201] As described above, if the last entry determination time and the sleep start determination time are also determined by pairing, the two times will not be reversed. Therefore, if the last entry determination time and the sleep start determination time are estimated using only data from a day on which both times are also determined, it is possible to ensure that the estimated results do not have any time reversals.

[0202] In one embodiment, in step S708, it is determined whether there is any combination among the following combinations 1 to 4 in which the estimated final entry time is not later than the estimated bedtime start time.

[0203] (Combination 1) Combination of estimated last entry time (overall median) and estimated bedtime start time (overall median) (Combination 2) Combination of estimated last entry time (overall median) and estimated bedtime start time (median by day of the week) (Combination 3) Combination of estimated last entry time (median by day of the week) and estimated bedtime start time (overall median) (Combination 4) Combination of estimated last entry time (median by day of the week) and estimated bedtime start time (median by day of the week)

[0204] For example, for combination 1, the server control unit 14 determines whether the overall median of the estimated last entry times is later than the overall median of the estimated bedtime start times. The same logic is used for combinations 2 to 4.

[0205] If any of combinations 1 to 4 is successful, the server control unit 14 confirms the provisionally estimated bedtime-related time for that combination. For example, when judging combination 2, the overall median of the estimated final entry time, "23:00," is not later than the median of the estimated bedtime start time by day, "01:00." In this case, "23:00" is finally confirmed as the estimated final entry time, and "01:00" is finally confirmed as the estimated bedtime start time.

[0206] Hereinafter, a method for estimating a bedtime-related time by calculating an error between the median of bedtime-related determination times and an actually determined bedtime-related determination time will be described with reference to Figs. 18A to 19B. Fig. 18A is a flowchart showing an example of estimation using an overall median in the fourth embodiment. Fig. 18B is a flowchart showing an example of estimation using a day-of-week median in the fourth embodiment. Fig. 19A is a schematic diagram showing an example of error calculation using an overall median in the fourth embodiment. Fig. 19B is a schematic diagram showing an example of error calculation using a day-of-week median in the fourth embodiment.

[0207] The server control unit 14 may calculate the error for each of the last entry determination time and the sleep start determination time using a similar method. In the explanation of Figures 18A to 19B, the estimation process is explained using the higher-level concept of "sleeping-related determination time," but the "sleeping-related determination time" in the explanation can be replaced with the "last entry determination time" or the "sleeping start determination time."

[0208] In this example, steps S802 and S822 correspond to step S420 in Fig. 15, and steps S804 to S810 and steps S824 to S830 correspond to step S430 in Fig. 15. Note that steps S802 to S810 and steps S822 to S830 may all be included in step S704 in Fig. 17.

[0209] First, the estimation based on the overall median shown in Fig. 18A will be described. The server control unit 14 calculates the overall median of the sleep-related determination times over the third predetermined period (step S802). Suppose the server control unit 14 calculates the overall median "23:00" based on the data sequence (1) (sleep-related determination times) shown in Fig. 19A.

[0210] Next, the server control unit 14 sets a provisional estimated time for a fourth predetermined period based on the overall median (step S804). The fourth predetermined period can be set depending on the frequency of execution of the estimation method. For example, if the estimation device executes the estimation method weekly to estimate bedtime-related times for seven days in the following week, the fourth predetermined period is set to seven days or more. Basically, the fourth predetermined period is set to be shorter than the third predetermined period. In the example of FIGS. 18A to 19B, the fourth predetermined period is set to seven days.

[0211] In step S804, the server control unit 14 sets all the tentative estimated times for the fourth predetermined period (one week) to the overall median. That is, all the tentative estimated times for the fourth predetermined period are the same value. As shown in data string (2) in FIG. 19A, all the tentative estimated times for the fourth predetermined period based on the overall median "23:00" are set to "23:00."

[0212] The server control unit 14 calculates and stores the error between the tentatively estimated time and the actually determined sleep-related determination time (step S806). In one embodiment, for each day of the fourth predetermined period, the error is calculated as the time difference obtained by subtracting the corresponding sleep-related determination time from the tentatively estimated time. As shown in data sequence (3) in FIG. 19A, if the sleep-related determination time is earlier than the tentatively estimated time, the error is negative, and the unit of error is minutes. For example, the error between the tentatively estimated time "23:00" on Monday and the corresponding sleep-related determination time "23:30" on the same day is "-30 (minutes)."

[0213] In one embodiment, after calculating the error of the overall median for the most recent fourth predetermined period, the server control unit 14 stores the error (i.e., data string (3) in FIG. 19A) in the server storage unit 12. In this way, every time the estimation process is executed, a new error of the overall median for the fourth predetermined period is calculated and stored in the server storage unit 12.

[0214] Next, the server control unit 14 acquires the error of the overall median over the third predetermined period (step S808). For example, if the third predetermined period is set to 35 days, the server control unit 14 reads the error of the overall median for the most recent 35 days from the server storage unit 12. Because the fourth predetermined period is set to be shorter than the third predetermined period, the acquired error includes the error calculated in step S806. An example of the error of the overall median over the third predetermined period is shown in data column (4) in FIG. 19A.

[0215] Next, the server control unit 14 calculates the mean absolute error (MAE) for each day of the week based on the acquired errors (step S810). For example, the server control unit 14 calculates the MAE for Monday using the errors corresponding to Monday over the 35 days. The calculated MAE for each day of the week is shown in data column (5) in FIG. 19A. The server control unit 14 uses the MAE for each day of the week as an index for determining the magnitude of the error.

[0216] Next, we will explain the estimation using the day-of-week median shown in Figure 18B. The difference between the process shown in Figure 18B and the process shown in Figure 18A is the median value used. In other processing techniques, the process shown in Figure 18B and the process shown in Figure 18A are the same.

[0217] In the process shown in FIG. 18B, the server control unit 14 calculates the median value of the sleep-related determination time by day of the week over a third predetermined period based on the sleep-related determination time (data sequence (6) in FIG. 19B) (step S822). Next, the server control unit 14 sets the tentative estimated time (data sequence (7) in FIG. 19B) for a fourth predetermined period based on the day-of-week median (step S824). The server control unit 14 calculates and stores the error (data sequence (8) in FIG. 19B) between the tentative estimated time based on the day-of-week median and the actually determined sleep-related determination time (step S826). Then, the server control unit 14 acquires the error (data sequence (9) in FIG. 19B) of the median value by day of the week over the third predetermined period from the server storage unit 12 (step S828). Based on the acquired error, the server control unit 14 calculates the MAE (data sequence (10) in FIG. 19B) for each day of the week (step S830).

[0218] 18A and 18B, the server control unit 14 acquires the "last estimated entry time (overall median)," "last estimated entry time (median by day of week)," "estimated bedtime start time (overall median)," and "estimated bedtime start time (median by day of week)" in steps S804 and S824. Furthermore, the server control unit 14 acquires the "MAE for each day of week based on the overall median of the last entry determination time," "MAE for each day of week based on the median by day of the last entry determination time," "MAE for each day of week based on the overall median of the bedtime start determination time," and "MAE for each day of week based on the median by day of the bedtime start determination time" in steps S810 and S830.

[0219] Thereafter, the server control unit 14 can execute steps S706 to S712 in FIG. 17 using these estimated times to produce a final estimation result.

[0220] In one embodiment, if it is determined in step S708 that there are multiple combinations in which the estimated final entry time is not later than the estimated bedtime start time, the server control unit 14 determines the bedtime-related time based on the sum of the MAEs in step S710. More specifically, the provisional estimation result corresponding to the combination with the smallest sum of the MAE of the estimated final entry time and the MAE of the estimated bedtime start time is determined as the final estimation result.

[0221] For example, in "Combination 1: combination of estimated final entry time (overall median) and estimated bedtime start time (overall median)" and "Combination 2: combination of estimated final entry time (overall median) and estimated bedtime start time (median by day of week)," the estimated final entry time is not later than the estimated bedtime start time. The server control unit 14 determines the sum of the MAE corresponding to the estimated final entry time (overall median) and the MAE corresponding to the estimated bedtime start time (overall median) as the error value for combination 1. Similarly, the server control unit 14 determines the sum of the MAE corresponding to the estimated final entry time (overall median) and the MAE corresponding to the estimated bedtime start time (median by day of week) as the error value for combination 2. If the error value of combination 1 is smaller than the error value of combination 2, the server control unit 14 determines the last estimated entry time (overall median) as the final estimated result of the last estimated entry time, and determines the estimated bedtime start time (overall median) as the final estimated result of the bedtime start time.

[0222] This completes the estimation process. The server control unit 14 calculates the overall median and the median for each day of the week of the sleep-related determination times, and then calculates the error for each. The estimated time with the smallest error is set as the sleep-related time. Therefore, the estimation process can accurately estimate the sleep-related time.

[0223] As explained in the first to third embodiments, the estimated bedtime-related time can be used in the output process for notifications or automatic control related to the operation of the air conditioner.

[0224] The setting process, determination process, estimation process, and output process described above do not need to be executed consecutively but can be executed independently. Each process has multiple technical features, but these technical features can also be executed independently. These technical features can be arbitrarily combined. The estimation device performs the estimation method by executing any combination of techniques based on the illuminance information log. For example, the estimation device may determine the light-on time and light-off time based only on the first and second determination conditions and output a notification to an information terminal based on the estimated sleep-related time. For example, the estimation device may estimate the sleep-related time using the determination process shown in Figures 13A and 13B and the estimation process shown in Figure 17 and control an air conditioner based on the estimated sleep-related time.

[0225] The above are merely specific embodiments of the present disclosure, and the scope of protection of the present disclosure is not limited thereto. The present disclosure includes the contents described above in the drawings and the specific embodiments described above, but the present disclosure is not limited thereto. Various disclosed embodiments or examples can be combined without departing from the scope or spirit of the present disclosure. Modifications that do not depart from the functional and structural principles of the present disclosure are within the scope of the claims. [Explanation of symbols]

[0226] 10 Servers 12 Server storage unit 14 Server control unit 16 Server Communication Unit 20 Air conditioner 21 Air conditioning memory unit 22 Air conditioning control unit 23 Air Conditioning Communication Department 24 Temperature Sensor 25 Illuminance sensor 30 Information terminal 32 Related Applications 40 Lighting equipment SC1~SC3 screen R1~R7 display area

Claims

1. A method for estimating a sleep-related time of a user of an air conditioner, comprising: acquiring an illuminance information log of a room that is a target for air conditioning control by the air conditioner over a past first predetermined period; estimating a sleep-related time associated with the user falling asleep in the room based on the illuminance information log; outputting notification information relating to the sleep-related times and the operation of the air conditioner to an information terminal; Including, the bedtime-related time includes an estimated final entry time of the user when the user finally enters the room before going to bed; In the output step, outputting the estimated last entry time and the notification information to the information terminal a first predetermined time before the estimated last entry time; A method for estimating bedtime-related times.

2. A method for estimating a sleep-related time of a user of an air conditioner, comprising: acquiring an illuminance information log of a room that is a target for air conditioning control by the air conditioner over a past first predetermined period; estimating a sleep-related time associated with the user falling asleep in the room based on the illuminance information log; outputting a control instruction to the air conditioner to operate the air conditioner based on the sleep-related time; Including, the bedtime-related time includes an estimated final entry time of the user when the user finally enters the room before going to bed; In the output step, outputting the control instruction to the air conditioner to start the air conditioner a first predetermined time before the estimated last room entry time; A method for estimating bedtime-related times.

3. the bedtime-related time includes an estimated bedtime start time at which the lights in the room are finally turned off before the user goes to bed; In the output step, When it is determined that the air conditioner is operating when a second predetermined time has elapsed from the estimated bedtime start time, the control instruction to switch the air conditioner to a sleep-related mode is output to the air conditioner. The method for estimating bedtime-related times according to claim 2 .

4. the illuminance information log includes time-series log data of the illuminance values ​​of the room, The method for estimating a sleep-related time includes: setting a first illuminance threshold for determining whether the room is in a lights-out state based on the number of data items corresponding to each illuminance value over a first time slot related to sleeping hours in the illuminance information log within the first predetermined period; setting a second illuminance threshold for determining whether the room is in a lighted state based on an illuminance value after an increase in the illuminance value over a second time period related to before going to bed in the illuminance information log within the first predetermined period; further comprising: The method for estimating a bedtime-related time according to any one of claims 1 to 3.

5. The method for estimating a sleep-related time includes: acquiring a first illuminance threshold value and a second illuminance threshold value set based on the illuminance information log within the first predetermined period; determining at least one of a light-on time and a light-off time for the illuminance information log of a third time slot associated with nighttime within a second predetermined period of time using the first illuminance threshold value and the second illuminance threshold value; determining at least one of a final entry determination time and a sleep start determination time for each of the third time slots within the second predetermined period based on the determined light-on time or light-off time; further comprising In the step of estimating a bedtime-related time, estimating the sleep-related time based on the last entry determination time or the sleep start determination time; The method for estimating a bedtime-related time according to any one of claims 1 to 4.

6. The step of determining at least one of the last entry determination time and the sleep start determination time includes: creating a pair of the determined turn-on times and the determined turn-off times that are consecutive in time series; a step of setting a lighting time of the latest pair of the pairs in the third time slot as the final entry determination time of the third time slot, and setting a lighting time of the latest pair as the sleep start determination time of the third time slot; Including, The method for estimating bedtime-related times according to claim 5 .

7. The step of estimating a bedtime-related time includes: acquiring a portion of the sleep-related determination time determined based on the illuminance information log, the portion corresponding to a third predetermined period in the past; Calculating a median of the acquired sleep-related determination times; determining a median value of the sleep-related determination times as the sleep-related time; Including, The method for estimating a bedtime-related time according to any one of claims 1 to 6.

8. In the step of calculating the median of the acquired sleep-related determination times, Calculating the overall median and the median for each day of the week of the acquired sleep-related determination time; The step of setting the median value of the sleep-related determination time as the sleep-related time includes: calculating an error between the overall median of the sleep-related determination times and the determined sleep-related determination times, and an error between the median of the sleep-related times by day of the week and the determined sleep-related determination times; determining the sleep-related time as the overall median of the sleep-related determination times or the median of the sleep-related times by day of the week, whichever has a smaller error; Including, The method for estimating bedtime-related times according to claim 7 .

9. A device for estimating a sleep-related time of a user of an air conditioner, acquiring an illuminance information log of a room that is the target of air conditioning control by the air conditioner over a past first predetermined period; estimating a sleep-related time related to the user's sleep in the room based on the illuminance information log; The notification information regarding the sleep-related time and the operation of the air conditioner is output to an information terminal. It is structured as follows: the bedtime-related time includes an estimated final entry time of the user when the user finally enters the room before going to bed; The device for estimating a sleep-related time, when outputting, The estimated last entry time and the notification information are output to the information terminal a first predetermined time before the estimated last entry time. further configured as follows: A device for estimating bedtime-related times.

10. A device for estimating a sleep-related time of a user of an air conditioner, acquiring an illuminance information log of a room that is the target of air conditioning control by the air conditioner over a past first predetermined period; estimating a sleep-related time related to the user's sleep in the room based on the illuminance information log; A control instruction for operating the air conditioner based on the sleep-related time is output to the air conditioner. It is structured as follows: the bedtime-related time includes an estimated final entry time of the user when the user finally enters the room before going to bed; The device for estimating a bedtime-related time comprises: outputting the control instruction to the air conditioner to start the air conditioner a first predetermined time before the estimated last room entry time; further configured as follows: A device for estimating bedtime-related times.

11. the bedtime-related time includes an estimated bedtime start time at which the lights in the room are finally turned off before the user goes to bed; The device for estimating a bedtime-related time comprises: When it is determined that the air conditioner is operating when a second predetermined time has elapsed from the estimated bedtime start time, the control instruction to switch the air conditioner to a sleep-related mode is output to the air conditioner. further configured as follows: The apparatus for estimating bedtime-related times according to claim 10.

12. the illuminance information log includes time-series log data of the illuminance values ​​of the room, The device for estimating a bedtime-related time comprises: setting a first illuminance threshold for determining whether the room is in a lights-out state based on the number of data items corresponding to each illuminance value over a first time slot related to a sleeping time in the illuminance information log within the first predetermined period; A second illuminance threshold is set based on the illuminance value after the illuminance value increases over a second time period related to before going to bed in the illuminance information log within the first predetermined period, for determining whether the room is in a lit state. further configured as follows: The device for estimating a bedtime-related time according to any one of claims 9 to 11.

13. The device for estimating a bedtime-related time comprises: acquiring a first illuminance threshold value and a second illuminance threshold value set based on the illuminance information log within the first predetermined period; determining at least one of a light-on time and a light-off time for the illuminance information log for a third time slot associated with nighttime within a second predetermined period of time using the first illuminance threshold value and the second illuminance threshold value; Based on the determined light-on time or light-off time, at least one of a final entry determination time and a sleep start determination time is determined for each of the third time slots within the second predetermined period. It is further structured as follows: The device for estimating a bedtime-related time, when estimating the bedtime-related time, The sleep-related time is estimated based on the last entry determination time or the sleep start determination time. further configured as follows: The device for estimating a bedtime-related time according to any one of claims 9 to 12.

14. When determining at least one of the last entry determination time and the sleep start determination time, the device for estimating sleep-related times: creating a pair of the determined light-on time and the determined light-off time that are consecutive in time series; The lighting time of the latest pair of the pairs in the third time slot is set as the final entry determination time of the third time slot, and the lighting time of the latest pair is set as the sleep start determination time of the third time slot. further configured as follows: The apparatus for estimating bedtime-related times according to claim 13.

15. The device for estimating a bedtime-related time, when estimating the bedtime-related time, A portion of the sleep-related determination time determined based on the illuminance information log, which corresponds to a third predetermined period in the past, is acquired; Calculating the median of the acquired sleep-related determination times; The median value of the sleep-related determination times is set as the sleep-related time. further configured as follows: The device for estimating a bedtime-related time according to any one of claims 9 to 14.

16. When calculating the median of the acquired sleep-related determination times, the device for estimating the sleep-related time Calculate the overall median and the median by day of the week of the sleep-related judgment time obtained above. It is further structured as follows: When the median of the sleep-related determination times is set as the sleep-related time, the device for estimating the sleep-related time calculating an error between the overall median of the sleep-related determination times and the determined sleep-related determination times, and an error between the median of the sleep-related times by day of the week and the determined sleep-related determination times; The sleep-related time is determined to be the one with the smaller error between the overall median of the sleep-related determination time and the median of the sleep-related time by day of the week. further configured as follows:

16. The apparatus for estimating bedtime-related times according to claim 15.

Images