JP7072952B1 - Life support device and life support robot system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a life support device and a life support robot system which appropriately support a target person through dialogue with the target person while reducing the stress of the target person. SOLUTION: In a life support system 1, a life support robot 3 as a life support device has a load usage status estimation unit 300 for estimating the usage status of a plurality of loads used in the living environment of a target person, and a usage status. Based on the behavior analysis unit 301 that analyzes the implementation status of the recommended behavior, and based on the implementation status, the notification information that induces the behavior change of the target person is generated, and the notification information is output to the voice output unit 35 and the display output unit 36. The behavior change guidance unit 302 that outputs from at least one of the (notification units), and the behavior change determination unit 303 that determines whether or not the behavior change has been performed according to the change in the usage status after the notification information is output. A support processing unit 304 that performs support processing to support the implementation of the recommended behavior based on the determination result of behavior change is provided. [Selection diagram] Fig. 2

Description

Patent Law Article 30, Paragraph 2 Applicable Published through telecommunication line, October 23, 2nd year of Reiwa, https: // www. youtube. com / watch? v = xyTBNNrwWQw, UIST 2020 (33rd ACM Symposium on User Interface Software and Technology)

The present invention relates to a life support device that improves communication with a life support robot by coordinating with the operating status of household electric machinery and equipment, and a life support robot system.

Conventionally, a system has been proposed in which the target person to be watched is installed in the house where the target person lives and remotely monitors the life of the target person. For example, in the monitoring monitoring system disclosed in Patent Document 1, a plurality of surveillance cameras, microphones, speakers, etc. are installed in the subject's home, and the subject is imaged by the surveillance camera, or the subject is imaged by the microphone and the speaker. It is a system that enables remote family members to talk.

Japanese Unexamined Patent Publication No. 2018-18457

In the monitoring monitoring system disclosed in Patent Document 1, since the surveillance camera is installed in the house, a great sense of security can be obtained for the subject. On the other hand, it has been clarified from previous studies that the subject is uncomfortable and feels stressed by the existence of the surveillance camera because the surveillance camera captures the state of his / her life.

Further, in the watching monitoring system disclosed in Patent Document 1, a remote family member can talk with the subject through a microphone and a speaker. However, the behavior change of the subject is timely so that the remote family can always grasp the behavior of the subject with the surveillance camera and take the recommended behavior recommended by the subject regarding, for example, eating, bathing, sleeping, etc. It is practically difficult to guide.

The present invention has been made in view of such circumstances, and is a life support device capable of appropriately providing support for a subject through dialogue with the subject while reducing stress on the subject. , And the purpose is to provide a life support robot system.

The present invention solves the above-mentioned problems, and the life support device according to the embodiment of the present invention is
A load usage estimation unit that estimates the usage status of multiple loads used in the living environment of the target person,
Based on the usage status, the behavior analysis unit that analyzes the implementation status of the recommended behavior recommended for the subject,
A behavior change guidance unit that generates notification information that induces behavior change of the target person based on the implementation status and outputs the notification information from the notification unit.
A behavior change determination unit that determines whether or not the behavior change has been performed in response to a change in the usage status after the notification information is output, and a behavior change determination unit.
It is provided with a support processing unit that performs support processing for supporting the implementation of the recommended behavior based on the determination result of the behavior change.

According to the life support device according to the embodiment of the present invention, the behavior analysis unit analyzes the implementation status of the recommended behavior based on the usage status of the plurality of loads used by the subject. Therefore, it is possible to grasp the behavior of the subject while reducing the stress of the subject. In addition, the behavior change guidance unit outputs notification information that induces the behavior change of the subject based on the implementation status of the recommended behavior, and the behavior change determination unit responds to changes in the usage status after outputting the notification information. It is determined whether or not the behavior change has been performed, and the support processing unit performs support processing to support the implementation of the recommended behavior based on the judgment result of the behavior change. Therefore, it is possible to appropriately support the target person through dialogue with the target person.

Issues, configurations and effects other than the above will be clarified in the form for carrying out the invention described later.

It is an overall block diagram which shows an example of a life support system 1. It is a block diagram which shows an example of a life support robot 3 (life support device 10). It is a data structure diagram which shows an example of the 1st recommended action table 311A. It is a data structure diagram which shows an example of the 2nd recommended action table 311B. It is a flowchart which shows an example of the operation of the health management of the life support robot 3 (life support device 10). It is a flowchart which shows an example of the operation about the abnormality correspondence of the life support robot 3 (life support device 10).

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

FIG. 1 is an overall configuration diagram showing an example of a life support system 1 according to an embodiment of the present invention. The life support system 1 monitors the implementation status of whether or not the recommended behavior recommended by the target person U is actually performed as the daily behavior of the target person U living in the living environment 11. It is a system to support health management and abnormal response in daily life. In the present embodiment, the subject U is described as being an elderly person living in each private room of a facility for the elderly, but the subject U may be an elderly person living alone at home. Further, the age of the subject U is not particularly limited to the elderly.

The living environment 11 is mainly an indoor space, and in the living environment 11, a plurality of loads 13 that operate by receiving electric power from the distribution board 12 are used. The distribution board 12 is installed in a part of the power supply path for supplying electric power to the plurality of loads 13, and is connected to the main circuit breaker 120 connected to the commercial power source and the electric power from the commercial power source connected to the main circuit breaker 120. Is provided with a plurality of branch breakers 121 for branching and supplying the plurality of loads 13. The floor plan and size of the living environment 11 are not particularly limited.

The load 13 is various electric appliances (household electric appliances), for example, a TV, an air conditioner (air conditioner), a refrigerator, an IH cooker, a microwave oven, a rice cooker, an electric pot, a coffee maker, and an electric water heater. , Ventilation fan, washing machine, hot water washing toilet seat, electric reclining bed, lighting equipment, etc. The load 13 may be operated by being operated by the target person U via an operation unit (for example, a button provided on the load main body or a remote controller using infrared communication or the like), or may be automatically set by a timer or the like. It may operate according to the above. The number of loads 13 and the installation location are not particularly limited.

Recommended behaviors are predetermined for each behavior type that classifies daily behaviors. The recommended behavior is determined, for example, by the subject U himself / herself, his / her family, the attending physician, etc., mainly in consideration of the health management of the subject U and the response to abnormalities. Examples of behavior types include, but are not limited to, meals, medications, excretion, going out, sleeping, naps, television, air conditioning, bathing, and showers.

The life support system 1 is a support project that provides a power meter 2 installed on the distribution board 12, a life support robot 3 that operates as a life support device 10, and a support service to the target person U as its components. It is equipped with a life support management device 4 managed by a person. The power meter 2 and the life support robot 3 and the life support robot 3 and the life support management device 4 are connected by a communication network 5 capable of transmitting and receiving various data. The communication network 5 may adopt any communication method, and may be wired or wireless. The load 13 may be connected to the communication network 5.

The power meter 2 is configured by a current sensor that is clamped to the power line of the distribution board 12 (in this embodiment, the secondary side of the main breaker 120) and measures the current value flowing through the power line. The power meter 2 measures the current value at a predetermined sampling cycle, and obtains a power waveform indicating a time-series change in the current value via a communication network 5 (for example, Wi-Fi (registered trademark)). It is transmitted to the support robot 3.

The life support robot 3 is a stationary device that can be carried by the subject U, and is installed at any place in the living environment 11. The life support robot 3 has a dialogue function for interacting with the target person U by using voice or display (characters, icons, images, etc.), and incorporates, for example, a general-purpose or dedicated computer. The life support robot 3 receives a power waveform from the power meter 2, and acquires the target person's behavior performed by the target person U based on the power waveform. Then, the life support robot 3 interacts with the target person U according to the situation, controls the load 13 by transmitting the load control data to the load 13, and various information (report information, action record data, etc.). To the life support management device 4.

The life support management device 4 is composed of, for example, a general-purpose or dedicated computer that functions as a server or a cloud. The life support management device 4 receives the report information from the life support robot 3 and uses any reporting means such as e-mail or telephone to support the target person U (family members and support business operators (specifically, elderly people). Contact a certified care worker, home helper, facility manager, etc.)). The life support management device 4 receives various information (behavior record data, etc.) acquired by the life support robot 3 and stores it in a database.

Although the overall configuration of the life support system 1 is illustrated in FIG. 1, the life support robot system includes at least a life support robot 3 and a life support management device 4. Further, in FIG. 1, a case where one life support robot 3 is connected to the life support management device 4 is shown, but a plurality of cases used in each living environment 11 of a plurality of target persons U, respectively. The life support robot 3 of the stand may be connected. The life support management device 4 may be connected to a life support robot 3 installed in each of a plurality of facilities for the elderly, or a life support robot installed in the home of an elderly person (target person U) living alone. 3 may be connected.

(Structure of life support robot 3)
FIG. 2 is a block diagram showing an example of a life support robot 3 (life support device 10) according to an embodiment of the present invention.

The life support robot 3 is a network communication unit that is a communication interface between a control unit 30 composed of processors such as a CPU, MPU, and GPU, a storage unit 31 composed of HDDs, SDDs, memories, and the like, and a communication network 5. It is composed of an infrared communication unit 33 that performs infrared communication between the 32 and the load 13, an audio input unit 34 composed of a microphone or the like, an audio output unit 35 composed of a speaker or the like, a display, a touch panel, or the like. Display output unit 36, operation input unit 37 composed of buttons, switches, touch panels, etc., and various sensors (for example, temperature / humidity sensor, illuminance sensor, image sensor (camera), acceleration sensor, gyro sensor, etc.) It is provided with a configured sensor group 38. In addition, a part or all of the above-mentioned parts 32 to 37 may be omitted, or a part or all of the sensors constituting the sensor group 38 may be omitted.

The storage unit 31 stores a life support program 310, a recommended action table 311, power waveform feature data 312, voice recognition data 313, utterance data 314, and action record data 315. The various types of information stored in the storage unit 31 may be updated as appropriate by receiving update information from the life support management device 4.

The recommended behavior table 311 is composed of a first recommended behavior table 311A mainly used for health management and a second recommended behavior table 311B mainly used for abnormal response. In the present embodiment, the recommended action table 311 will be described as being divided into two tables, but it may be composed of one table or three or more tables.

FIG. 3 is a data structure diagram showing an example of the first recommended action table 311A according to the embodiment of the present invention. FIG. 4 is a data structure diagram showing an example of the second recommended action table 311B according to the embodiment of the present invention. The first and second recommended action tables 311A and 311B store information on recommended actions in each field for each action type corresponding to each record.

In the first recommended action table 311A, as shown in FIG. 3, the action content of the recommended action, the load 13 to be judged when determining the implementation status of the recommended action, and the recommended action are shown in FIG. The contents of the processing at the time of non-execution performed as the support processing when the implementation status is not implemented and the recorded contents recorded in the action record data 315 are stored.

In the second recommended action table 311B, as shown in FIG. 4, the action content of the recommended action, the load 13 to be judged when judging the implementation status of the recommended action, and the recommended action are shown in FIG. The abnormality judgment criteria for judging whether or not there is an abnormality in the implementation status and the contents of the abnormality processing performed as support processing when the implementation status of the recommended action meets the abnormality judgment criteria are stored.

As the load 13 to be determined, one or a plurality of loads 13 are set for one action type, and when a plurality of loads 13 are set, the order of the time series in which they are used is further set. May be done.

In the first recommended action table 311A exemplified in FIG. 3, 6 records, that is, 6 types of action types (meal, medication, excretion, going out, sleep, television) are registered, and are exemplified in FIG. In the second recommended action table 311B, four records, that is, four types of actions (meal, air conditioning, bathing, shower) are registered. In the first and second recommended action tables 311A and 311B, a common action type may be registered, or different action types may be registered, and the examples are shown in FIGS. 3 and 4. Action types other than may be registered. Further, the first and second recommended action tables 311A and 311B are configured to be editable by the target person U himself / herself, his / her family, the attending physician, etc. via the operation input unit 37, and records can be added or deleted. It may be performed as appropriate, or the content set in each field may be changed as appropriate.

The power waveform feature data 312 stores the features of the power waveform (size and slope at the time of current rise, size and slope at the time of current fall, etc.) for each load 13. The power waveform feature data 312 may be a learning model in which the correlation between the power waveform feature and the load 13 is machine-learned by a predetermined machine learning algorithm. In that case, the learning model as the power waveform feature data 312 outputs, for example, the usage status of the load 13 estimated to have been used when the power waveform was measured by inputting the power waveform.

The voice recognition data 313 stores data and a program for the life support robot 3 to recognize the utterance content of the target person U from the voice data of the target person U input via the voice input unit 34. The utterance data 314 stores data and programs for uttering utterance contents according to various situations via the voice output unit 35. The action record data 315 stores the target person's action performed by the target person U and the voice spoken by the target person U together with the action and the date and time when the voice was acquired. The action record data 315 may further store power waveform data.

By executing the life support program 310, the control unit 30 executes the load usage status estimation unit 300, the behavior analysis unit 301, the behavior change guidance unit 302, the behavior change determination unit 303, the support processing unit 304, and the first recommended behavior update. It functions as a unit 305A, a second recommended behavior updating unit 305B, and a third recommended behavior updating unit 305C.

The load usage status estimation unit 300 acquires a power waveform from the power meter 2 installed in the distribution board 12 that supplies power to the plurality of loads 13, and estimates the usage status of the load 13 from the power waveform. Specifically, the load usage status estimation unit 300 refers to the power waveform feature data 312, and determines which load 13 is being used from the time-series change of the power waveform measured by the power meter 2, for example, at what time point. Estimates the usage status of each load 13, such as when it was turned on and when it was turned off.

The behavior analysis unit 301 analyzes the implementation status of the recommended behavior based on the usage status of the load 13. Then, the behavior analysis unit 301 determines whether or not the implementation status of the recommended behavior has not been implemented, and whether or not the implementation behavior of the recommended behavior satisfies the abnormality determination criterion.

Specifically, the behavior analysis unit 301 acquires the target person's behavior performed by the target person U from the usage status of the load 13, and also refers to the recommended behavior table 311 and implements the recommended behavior based on the target person's behavior. Whether or not the situation has not been implemented, and whether or not the implementation status of the recommended action meets the abnormality judgment criteria based on the behavior of the subject is determined for each action type.

For example, when the action type is "meal", the behavior analysis unit 301 determines that the load 13 required for meals (for example, kitchen lighting, etc.) is used in the specified time zone as the usage status of the load 13. As a condition, it is judged that "meal" as a recommended action has been carried out. Further, when the action type is "going out", the behavior analysis unit 301 stops using the load 13 necessary for daily life (lighting of the entire room, etc.) from the state where the load 13 is used. It is determined that "going out" as a recommended action has been carried out on the condition that the state has been continued for a predetermined period. Further, when the behavior type is "sleep", the behavior analysis unit 301 turns on the bedroom lighting from a state where the load 13 necessary for daily life (lighting of the entire room, etc.) is not used, and further, the behavior analysis unit 301 is further turned on. It is determined that "sleep (wake up)" as a recommended action has been performed on condition that the living lighting and the kitchen lighting are turned on in this order. The order in which the plurality of loads 13 as described above are used may be set according to the floor plan of the living environment 11 and the lifestyle of the target person U.

The behavior change guidance unit 302 generates notification information for each behavior type to induce the behavior change of the target person U based on the implementation status of the recommended behavior for each behavior type, and the notification information is output to the voice output unit 35 and the display. It is output from at least one of the units 36 (notification unit). The broadcast information may include content that induces behavior change. For example, the utterance content to be spoken (notified) by the voice output unit 35 as the notification information is stored in the utterance data 314, and the display content displayed by the display output unit 36 as the notification information is based on the utterance data 314. It is displayed in characters.

Whether or not the behavior change determination unit 303 has changed the behavior of the subject U according to the change in the usage status of the load 13 after the notification information is output from at least one of the voice output unit 35 and the display output unit 36. Is determined for each action type. For example, the behavior change determination unit 303 cooperates with the load usage status estimation unit 300, and when the load usage status estimation unit 300 detects (estimates) a change in the usage status of the load 13, the target person U Judge that behavior change has occurred. Further, the behavior change determination unit 303 recognizes, for example, the utterance content to the effect that the recommended action has been performed as the utterance of the target person U from the voice data of the target person U input via the voice input unit 34. Determines that the behavioral change of the subject U has been performed.

The support processing unit 304 performs support processing to support the implementation of the recommended action based on the determination result of the behavior change for each action type. Specifically, the support processing unit 304 outputs the action content of the recommended action from the notification unit as the support process when the behavior change is not performed and the implementation status of the recommended action is not implemented. Further, the support processing unit 304 recommends the load 13 of the judgment target to be used in the recommended behavior as the support processing when the behavior of the target person U is not changed and the implementation status of the recommended behavior satisfies the abnormality judgment criterion. At least one of control is performed according to the operation to be performed as an action, and notification information for notifying the implementation status of the recommended action is transmitted via the network communication unit 32. As a load control method in which the support processing unit 304 controls the load 13 to be determined, load control data (for example, data based on an arbitrary API) corresponding to the operation to be performed as the recommended action is used in the network communication unit 32 or the network communication unit 32 or. It is transmitted via the infrared communication unit 33.

At that time, the support processing unit 304 refers to the recommended action table 311 and performs the support processing set for each action type. For example, when the implementation status of the recommended action has not been implemented, the first one. The support process is performed according to the non-execution processing content of the recommended action table 311A, and when the implementation status of the recommended action meets the abnormality determination criteria, the support process is performed according to the abnormality processing content of the second recommended action table 311B.

Further, the support processing unit 304 may acquire the request information for the target person U at the timing of performing the support processing, and perform the support processing based on the request information. For example, the support processing unit 304 may acquire the request of the target person U himself / herself (things related to lifestyle, etc.) as request information via the voice input unit 34 or the operation input unit 37, or may be a family member or a doctor in charge (a family member or a doctor in charge). Requests from medical institutions (such as those related to disease treatment and prevention) or requests from facilities for the elderly (such as those related to living rules in facilities for the elderly and rules for using shared facilities) can be received from the life support management device 4 and It may be acquired from an external information providing device via the network communication unit 32, or when the requested information acquired as described above is stored in the storage unit 31, it is acquired by referring to the storage unit 31. You may. As a result, it is possible to perform timely and appropriate support processing and provide information that reflects various requests.

Further, the support processing unit 304 may acquire the advertisement information for the target person U at the timing of performing the support processing, and perform the support processing based on the advertisement information. For example, the support processing unit 304 may use advertisements from facilities for the elderly (such as those related to paid programs and paid events in the facilities for the elderly) or advertisements from companies, stores, and governments in neighboring areas as advertisement information. (Sales and events in the area, business trip services, delivery services, administrative programs, etc.) may be acquired from the life support management device 4 or an external information providing device via the network communication unit 32, or may be obtained as described above. When the acquired advertisement information is stored in the storage unit 31, it may be acquired by referring to the storage unit 31. Then, the support processing unit 304 generates an advertisement list based on the acquired advertisement information, and performs support processing from the advertisement list according to at least one of the usage status of the load 13 and the action type of the support target. Appropriate advertising information is selected for the target person U in the above, and support processing is performed based on the selected advertising information.

Specifically, when the action record data 315 records that the coffee maker is used on a daily basis as the usage status of the load 13, the support processing unit 304 performs support processing related to meals. At the timing, the advertisement information about the coffee beans and the equipment of the coffee maker may be selected, and at the timing of the support processing regarding going out, the advertisement information about the coffee shop or the sale of the coffee beans may be selected. In addition, when the request information from the attending physician (medical institution) is requested to incorporate exercise on a daily basis, the support processing unit 304 performs support processing related to going out in the facility for the elderly. You may select advertising information about paid programs such as exercises, or you may select advertising information about gymnastics programs at nearby sports gyms. As a result, it is possible to perform timely and appropriate support processing and provide information reflecting various advertisements.

The first recommended action update unit 305A acquires the request information for the target person U, and based on the request information, the action content of the recommended action (fields of the recommended actions of the first and second recommended action tables 311A and 311B). To update. The first recommended action update unit 305A acquires request information by, for example, a push method and a pull method, and the timing of acquiring the request information is different from the timing of the support processing unit 304 performing support processing. But it may be. Since the content of the request information is the same as the request information acquired by the support processing unit 304, the description thereof will be omitted.

The second recommended action update unit 305B acquires advertisement information for the target person U, and from the acquired advertisement information, an appropriate advertisement for the target person U according to at least one of the usage status of the load 13 and the action type of the recommended action. Information is selected, and the action content of the recommended action (the field of the recommended action in the first and second recommended action tables 311A and 311B) is updated based on the selected advertisement information. The second recommended action update unit 305B receives advertisement information by, for example, a push method and a pull method, and the timing of receiving the advertisement information may be different from the timing of the support processing unit 304 performing the support processing. .. Since the content of the advertisement information is the same as the advertisement information acquired by the support processing unit 304, the description thereof will be omitted.

The third recommended action update unit 305C calculates the activity amount of the subject U based on the usage status of the load 13, and the action content of the recommended action (first and second recommended action tables 311A) based on the activity amount. , 311B recommended action field) is updated. For example, the third recommended behavior update unit 305C estimates the movement locus of the target person U in the living environment 11 from the usage status of the load 13, and calculates the activity amount of the target person U from the movement distance based on the movement locus. Alternatively, the activity amount of the target person U may be calculated from the viewing time of the television, or the activity amount of the target person U may be calculated from the outing frequency and the outing time of the target person U. Then, when the activity amount of the target person U is less than the predetermined reference value, the third recommended action update unit 305C encourages the implementation of the recommended action and increases the activity amount of the target person U, for example. When the action type is "TV", the specified time for the action content of the recommended action is changed to a shorter time, and when the action type is "going out", the frequency of going out for the action content of the recommended action is changed a lot. This makes it possible to prevent mental illness such as dementia and depression in the subject U.

(Operation of life support system 1)
Next, the operation of the life support robot 3 (life support device 10) having the above configuration will be described.

FIG. 5 is a flowchart showing an example of an operation related to health management of the life support robot 3 (life support device 10) according to the embodiment of the present invention. The series of processes shown in FIG. 5 is executed for each action type set in the first recommended action table 311A (FIG. 3), and when a plurality of action types are set, the series of processes is executed, for example. A series of processes shown in FIG. 5 are executed in parallel. Here, the behavior for the action type of "going out" will be mainly described.

First, in step S100, the control unit 30 of the life support robot 3 refers to the first recommended action table 311A, and the information set for a specific action type (“going out”) (action content of the recommended action). , The load 13 to be judged, the processing content when not executed, and the recorded content) are acquired.

Next, in step S110, the load usage status estimation unit 300 receives the power waveform from the power meter 2 at a predetermined cycle, and the usage status of the load 13 (“lighting of the entire room”) to be determined from the power waveform. To estimate.

Next, in step S120, the behavior analysis unit 301 analyzes the implementation status of the recommended action based on the usage status of the load 13 (“lighting of the entire room”) of the judgment target estimated by the load usage status estimation unit 300. By making a (judgment), the target person's behavior performed by the target person U is acquired.

Next, in step S130, the behavior analysis unit 301 determines whether or not the behavior content of the recommended behavior has not been implemented based on the behavior of the target person. For example, the behavior analysis unit 301 determines that the action content of the recommended action has been executed in response to turning off the "lighting of the entire room" as the target person's action for the action type of "going out".

Then, when the behavior analysis unit 301 determines that the action content of the recommended action has not been implemented (step S130: Yes), in step S131, the behavior change guidance unit 302 performs the behavior change of the subject U. The notification information to be guided is generated, and the notification information is output from the notification unit. For example, the notification information for the action type of "going out" includes information for inducing the person to go out. At that time, the behavior change guidance unit 302 generates notification information for inducing a preliminary action before performing the recommended action (for example, when the action type is "going out", the lighting of the entire room is turned off, etc.). The notification information may be output from the notification unit.

Next, in step S140, whether or not the behavior change determination unit 303 has changed the behavior of the subject U according to the change in the usage status of the load 13 after the voice output unit 35 outputs the notification information. To judge. For example, the behavior change determination unit 303 cooperates with the load usage status estimation unit 300, and when the load usage status estimation unit 300 detects (estimates) a change in the usage status of the load 13, the target person U Judge that behavior change has occurred. Further, the behavior change determination unit 303 recognizes, for example, the utterance content to the effect that the recommended action has been performed as the utterance of the target person U from the voice data of the target person U input via the voice input unit 34. Determines that the behavioral change of the subject U has been performed.

Then, when the behavior change determination unit 303 determines that the behavior change of the target person U has not been performed (step S140: No), in step S150, the support processing unit 304 supports according to the processing content at the time of non-implementation. Perform processing. For example, when the processing content at the time of non-execution for the action type of "going out" is set to "recommended action presentation and confirmation", the support processing unit 304 presents the action content of the recommended action and the action. Generates notification information for confirming whether the content is planned (or has been implemented), and outputs the notification information from the notification unit. The notification information here may include the same contents as the notification information output in step S131.

When the support processing unit 304 acquires, for example, request information for requesting exercise on a daily basis from the life support management device 4 at the timing of performing support processing, it generates notification information including that fact. , The notification information may be output from the notification unit. Further, when the support processing unit 304 acquires advertisement information regarding a paid program such as exercise performed in a facility for the elderly from the life support management device 4 at the timing of performing the support processing, a notification including that fact is included. Information may be generated and the notification information may be output from the notification unit.

Next, in step S160, the support processing unit 304 registers the recorded contents such as the implementation status of the recommended action (in the case of the action type of “going out”, the presence / absence and time of going out) in the action record data 315. When the behavior analysis unit 301 determines in step S130 that the action content of the recommended action has been executed (S130: No), and in step S140, the behavior change determination unit 303 determines the behavior of the target person U. Even when it is determined that the transformation has been performed (step S140: Yes), step S160 is also performed.

As described above, the series of processes shown in FIG. 5 is completed. When the behavior analysis unit 301 determines that the recommended action has been performed (step S130: No), or when the behavior change determination unit 303 determines that the behavior change of the subject U has been performed (step S140: Yes). ), The behavior change induction unit 302 performs post-behavior after performing the recommended behavior (for example, hydration when the behavior type is "bathing", medication when the behavior type is "meal", etc.). The notification information may be generated and the notification information may be output from the notification unit.

FIG. 6 is a flowchart showing an example of an operation related to an abnormality response of the life support robot 3 (life support device 10) according to the embodiment of the present invention. The series of processes shown in FIG. 6 is executed for each action type set in the second recommended action table 311B (FIG. 4), and when a plurality of action types are set, the series of processes is executed, for example. A series of processes shown in FIG. 6 are executed in parallel. Here, the operation for the action type of "air conditioning" will be mainly described.

First, in step S200, the control unit 30 of the life support robot 3 refers to the second recommended action table 311B, and the information set for a specific action type (“air conditioning”) (action content of the recommended action). , The load 13 to be judged, the abnormality judgment standard, and the processing content at the time of abnormality) are acquired.

Next, in step S210, the load usage status estimation unit 300 estimates the usage status of the load 13 (“air conditioner”) to be determined, as in step S110. Next, in step S220, the behavior analysis unit 301 analyzes (determines) the implementation status of the recommended action based on the usage status of the load 13 (“air conditioner”) to be determined, as in step S120. Acquire the subject action performed by the subject U. Next, in step S230, the behavior analysis unit 301 determines whether or not the action content of the recommended action has not been implemented based on the target person's behavior, as in step S130. For example, the behavior analysis unit 301 determines that the action content of the recommended action is executed according to the operation of the "air conditioner" as the target person action for the action type of the "air conditioner".

Then, when the behavior analysis unit 301 determines that the action content of the recommended action has not been implemented (step S230: Yes), in step S231, the behavior change guidance unit 302 is the target as in step S131. The notification information that induces the behavior change of the person U is generated, and the notification information is output from the notification unit. For example, the notification information for the action type of "air conditioning" includes information for inducing the operation of the air conditioner.

Next, in step S240, the behavior change determination unit 303 changes the behavior of the subject U according to the change in the usage status of the load 13 after the broadcast information is output from the voice output unit 35, as in step S140. Is determined whether or not has been performed.

Then, when the behavior change determination unit 303 determines that the behavior change of the target person U has not been performed (step S240: No), in step S250, the behavior analysis unit 301 determines the target person's behavior and abnormality determination criteria. Based on, it is determined whether or not the subject's behavior meets the abnormality judgment criteria. Specifically, the behavior analysis unit 301 does not operate the "air conditioner" as the target person's behavior for the behavior type of the "air conditioner", and the room temperature is equal to or higher than the specified temperature or lower than the specified temperature. , Judge that the abnormality judgment criteria are satisfied.

Then, when the behavior analysis unit 301 determines that the abnormality determination criterion is satisfied (step S250: Yes), in step S260, the support processing unit 304 performs support processing according to the content of the abnormality processing. For example, when the unimplemented processing content for the action type of "air conditioning" is set to "automatic control + report to administrator", the support processing unit 304 determines the implementation status of the recommended action. The load control data (air conditioner on operation) according to the operation to be performed as the recommended action is transmitted to the load 13 (“air conditioner”) to be judged, and the supporter of the target person U is notified. Send information. The report information may be directly transmitted to the supporter of the target person U without going through the life support management device 4.

As described above, the series of processes shown in FIG. 6 is completed. Although FIGS. 5 and 6 have described the case where the flowcharts are separate, life support is provided as long as the function related to health management (FIG. 5) and the operation related to abnormal response (FIG. 6) are realized. The robot 3 (life support device 10) is not limited to the examples of FIGS. 5 and 6, and may be one that operates according to one flowchart. Further, the life support robot 3 (life support device 10) may perform either an operation related to health management (FIG. 5) or an operation related to abnormal response (FIG. 6).

Therefore, according to the life support system 1 according to the present embodiment, the behavior analysis unit 301 analyzes the implementation status of the recommended behavior based on the usage status of the plurality of loads 13 used by the subject U. Therefore, it is possible to grasp the behavior of the target person U while reducing the stress of the target person U. Further, after the behavior change guidance unit 302 outputs the notification information for inducing the behavior change of the target person U based on the implementation status of the recommended behavior from the notification unit, and the behavior change determination unit 303 outputs the notification information. It is determined whether or not the behavior change has been performed according to the change in the usage situation, and the support processing unit 304 performs the support process to support the implementation of the recommended action based on the determination result of the behavior change. Therefore, the support of the target person U can be appropriately provided through the dialogue with the target person U.

(Other embodiments)
Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment and can be appropriately modified without departing from the technical idea of the present invention.

For example, in the above embodiment, the case where the life support robot 3 operates as the life support device 10 has been described, but the life support device 10 may be configured as a device separate from the life support robot 3. In that case, the life support device 10 may support the life of the target person U by transmitting and receiving various information to and from the life support robot 3. Further, the life support management device 4 may be operated as the life support device 10, and in that case, the life support device 10 transmits and receives various information to and from the plurality of life support robots 3. Then, the life of a plurality of target persons U may be supported.

Further, in the above embodiment, the case where the life support robot 3 is composed of a stationary device that can be carried by the target person U and is installed in the living environment 11 has been described, but the form of the life support robot 3 is appropriate. You may change it. For example, the life support robot 3 may be composed of an autonomously moving device that includes an actuator that runs on wheels, walks, and the like, and autonomously moves in the living environment 11. Further, when the life support robot 3 is provided with a finger-shaped or arm-shaped actuator, the life support robot 3 may realize a load control method by the support processing unit 304, for example, from the support processing unit 304. Based on the load control data, the actuator may be used to operate the operation unit of the load 13 (for example, a button provided on the load body, a remote controller using infrared communication, or the like). Further, the life support robot 3 is realized by an agent-type user interface in which a robot (character) is displayed on the screen of an arbitrary device (may be a plurality of devices) such as a television, a smartphone, and a tablet. It may be integrated with the room by being embedded in the wall or the ceiling. Further, the life support robot 3 may be installed in various places in the room so that a plurality of life support robots 3 cooperate with each other.

Further, in the above embodiment, the case where the power meter 2 is installed on the secondary side of the main breaker 120 (centralized installation type) has been described, but the power meter 2 is in the power supply path for supplying power to the load 13. It suffices if it is installed, and a plurality of power meters 2 may be installed. For example, the power meter 2 may be installed on the secondary side of each branch breaker 121, or may be installed between the outlet and the outlet plug of the load 13 like a smart plug. In the case of such an individually installed type, the load usage status estimation unit 300 may estimate the usage status of each load 13 by acquiring a power waveform from each power meter 2. Further, when the power meter 2 includes a changeover switch capable of switching between an energized state and a cutoff state, the power meter 2 may realize a load control method by the support processing unit 304, for example, the support processing unit. The changeover switch may be switched based on the load control data (energization or cutoff) from 304.

Further, in the above embodiment, the case where the load usage status estimation unit 300 estimates the usage status of each load 13 based on the power waveform acquired by the power meter 2 has been described, but the load usage status estimation unit 300 describes the case. The usage status (ON / OFF, operation setting, power consumption, etc.) of each load 13 may be acquired by transmitting / receiving data based on an arbitrary API to / from each load 13.

Further, in the above embodiment, the case where the behavior analysis unit 301 analyzes the implementation status of the recommended behavior based on the usage status of the load 13 has been described, but the behavior analysis unit 301 uses the information acquired by the sensor group 38. Further reference may be made to analyze the implementation status of the recommended behavior. Further, the behavior change determination unit 303 may detect (estimate) a change in the usage status of the load 13 by further referring to the information acquired by the sensor group 38.

Further, in the above embodiment, the case where the action content of the recommended action in the recommended action table 311 is determined by, for example, the subject U himself / herself, his / her family, the attending physician, etc. has been described. By referring to the behavior record data 315 which is the past subject behavior and analyzing the behavior record data 315, at least one of the behavior content and the abnormality judgment criterion of the recommended behavior may be changed. For example, the behavior analysis unit 301 performs statistical processing on the behavior record data 315 accumulated in the past fixed period (for example, 3 months, 1 year, etc.), and performs statistical processing on the meal time zone, bathing, sleeping, and television. Statistical indicators (mean, variance, etc.) such as such time and frequency of excretion may be calculated, and at least one of the behavioral content of the recommended behavior and the abnormality judgment criteria may be changed based on the calculation result of the statistical index. Specifically, for example, when the average meal time zone is 8 o'clock, 13 o'clock, and 19 o'clock, the behavior analysis unit 301 sets the specified time zone in the action content of the recommended action to 8. It may be changed to the hour, 13:00, and 19:00. As a result, the support of the target person U can be provided according to the daily behavior pattern of the target person U.

Further, in the above embodiment, the behavior change guidance unit 302 generates notification information for inducing a pre-action before performing the recommended behavior and notification information for inducing post-action after performing the recommended behavior, and the notification thereof. The case where information is output from the notification unit has been described, but the recommended action table 311 is configured so that the content of the pre-action, the content of the notification information for the pre-action, the content of the post-action, and the content of the notification information for the post-action can be registered. You may be. In that case, the behavior change guidance unit 302 may refer to the recommended behavior table 311 and output the notification information for inducing the pre-action or the post-action.

Further, in the above embodiment, the case where the support processing unit 304 performs the support processing according to the processing contents at the time of non-implementation and the processing contents at the time of abnormality has been described, but the support processing unit 304 performs the support processing according to the other processing contents. Alternatively, the support process may be performed by referring to external information via the network communication unit 32. As external information, for example, neighborhood information (store location, business hours, friend's house location, etc.), support system information of the support company (family or family house location, supporter in the support company (nursing care welfare)) Positions, working hours, etc. of technicians, home helpers, facility managers, etc.).

Further, in the above embodiment, the case where the life support device 10 monitors the implementation status of "going out" as one of the action types of the recommended behavior has been described. However, in the life support device 10, the subject U is at midnight. You may try to keep an eye on the belt so that you do not go out. For example, the behavior analysis unit 301 determines that the subject U is going out, determines whether or not there is an abnormality in the implementation status of the outing behavior according to whether or not the time zone is midnight, and induces behavior change. When it is determined that there is an abnormality in the outing behavior, the unit 302 may generate guidance information for inducing the target person U not to go out, and output the notification information from the notification unit. .. Further, when the target person U has already gone out and it is determined that there is an abnormality in the outing behavior, the support processing unit 304 may transmit the report information to the life support management device 4. At that time, the report information includes the time when it is determined to go out, information on nearby areas, information on candidates for stopping by and information on the search range, and the like.

Further, in the above embodiment, the life support program 310 has been described as being stored in the storage unit 31, but it is a file in an installable format or an executable format and can be read by a computer such as a CD-ROM or a DVD. It may be recorded on a recording medium and provided. Further, the life support program 310 may be provided by storing it on a computer connected to a network such as the Internet and downloading it via the network.

1 ... life support system, 2 ... power meter, 3 ... life support robot, 4 ... life support management device,
5 ... Communication network, 10 ... Life support device, 11 ... Living environment, 12 ... Distribution board, 13 ... Load,
30 ... Control unit, 31 ... Storage unit, 32 ... Network communication unit, 33 ... Infrared communication unit,
34 ... Audio input unit, 35 ... Audio output unit, 36 ... Display output unit, 37 ... Operation input unit,
38 ... Sensor group, 120 ... Main breaker, 121 ... Branch breaker,
300 ... Load usage status estimation unit, 301 ... Behavior analysis unit, 302 ... Behavior change guidance unit,
303 ... Behavior change judgment unit, 304 ... Support processing unit, 305A ... First recommended behavior update unit,
305B ... 2nd recommended behavior update department, 305C ... 3rd recommended behavior update department,
310 ... Life support program,
311 ... Recommended behavior table, 311A ... First recommended behavior table,
311B ... Second recommended action table, 312 ... Power waveform feature data,
313 ... voice recognition data, 314 ... utterance data, 315 ... action record data

Claims (4)

A load usage estimation unit that estimates the usage status of multiple loads used in the living environment of the target person,
Based on the usage status, the behavior analysis unit that analyzes the implementation status of the recommended behavior recommended for the subject,
A behavior change guidance unit that generates notification information that induces behavior change of the target person based on the implementation status and outputs the notification information from the notification unit.
A behavior change determination unit that determines whether or not the behavior change has been performed in response to a change in the usage status after the notification information is output, and a behavior change determination unit.
A support processing unit that performs support processing to support the implementation of the recommended behavior based on the determination result of the behavior change is provided .
The support processing unit
Advertisement information for the target person is acquired at the timing of performing the support process, and the advertisement information is selected from the acquired advertisement information according to at least one of the usage status and the action type of the recommended behavior, and the advertisement information is selected. Perform the support process based on the advertisement information.
Life support device. A load usage estimation unit that estimates the usage status of multiple loads used in the living environment of the target person,
Based on the usage status, the behavior analysis unit that analyzes the implementation status of the recommended behavior recommended for the subject,
A behavior change guidance unit that generates notification information that induces behavior change of the target person based on the implementation status and outputs the notification information from the notification unit.
A behavior change determination unit that determines whether or not the behavior change has been performed in response to a change in the usage status after the notification information is output, and a behavior change determination unit.
A support processing unit that performs support processing to support the implementation of the recommended behavior based on the determination result of the behavior change, and a support processing unit.
The advertisement information for the target person is acquired, the advertisement information is selected from the acquired advertisement information according to at least one of the usage situation and the action type of the recommended behavior, and the advertisement information is selected based on the selected advertisement information. It has a second recommended action update unit that updates the action content of the recommended action.
Life support device. The behavior analysis unit
Based on the usage status, the implementation status of the recommended behavior is analyzed for each behavior type, and
The behavior change induction unit is
Based on the implementation status for each action type, the notification information is generated for each action type, and the notification information for each action type is output from the notification unit.
The behavior change determination unit is
Whether or not the behavior change has been performed is determined for each behavior type, and the behavior is determined.
The support processing unit
Based on the determination result of the behavior change for each behavior type, the support process is performed for each behavior type.
The life support device according to claim 1 or 2 . The life support device according to any one of claims 1 to 3 , and the life support device.
A life support robot having the notification unit is provided.
Life support robot system.

Images