JP7742726B2 - Facility control device, facility control system, facility control method and program - Google Patents

Description

The present disclosure relates to an equipment control device, an equipment control system , an equipment control method, and a program.

Technology is known for controlling equipment installed in a target area based on information about people present in that area. For example, Patent Document 1 discloses an air conditioner that uses a Doppler sensor to acquire biological information such as breathing, heart rate, and body movement, and controls the air conditioning unit based on the acquired biological information.

JP 2011-15887 A

In technology that controls equipment based on information about people such as those described above, there is a need to accurately control equipment according to the condition of people, thereby improving the concentration of people in the target area and providing an environment where they can focus more on their work.

This disclosure has been made to solve the above-mentioned problems, and aims to provide a facility control device and the like that can improve the concentration of people in a target area.

In order to achieve the above object, a first aspect of an equipment control device according to the present disclosure includes:
An equipment control device that controls equipment installed in a target area,
an acquisition means for acquiring biometric information of a person present in each of a plurality of individual areas within the target area;
an estimation means for estimating a concentration level of the person present in each of the plurality of individual areas from the biometric information acquired by the acquisition means, based on a trained model generated by learning the relationship between the biometric information of a learning target and the corresponding concentration level;
equipment control means for controlling equipment installed in each of the plurality of individual areas based on the concentration degree estimated by the estimation means;
and an output means for outputting the estimation result by the estimation means of the concentration of people in each of the plurality of individual areas and the control status of the equipment installed in each of the plurality of individual areas, in association with each individual area .
In order to achieve the above object, a second aspect of the equipment control device according to the present disclosure includes:
An equipment control device that controls equipment installed in a target area,
an acquisition means for acquiring biometric information of a person present in the target area;
an estimation means for estimating a concentration level of the person present in the target area from the biometric information acquired by the acquisition means, based on a trained model generated by learning the relationship between the biometric information of the learning target and the corresponding concentration level;
an equipment control unit that controls the equipment based on the concentration degree estimated by the estimation unit,
The equipment control means causes the equipment to perform a stimulating operation to stimulate the person present in the target area in each of the following cases: when the concentration level estimated by the estimation means changes from a first state to a second state in which the concentration level is lower than the first state; when the concentration level estimated by the estimation means remains in the second state for a period longer than a threshold time; and when the concentration level estimated by the estimation means remains in the first state for a period longer than a threshold time.

In this disclosure, an equipment control device estimates the concentration level of a person present in a target area from the person's biometric information based on a trained model generated by learning the relationship between the biometric information of the person being studied and the corresponding concentration level, and controls equipment installed in the target area based on the estimated concentration level. Therefore, this disclosure makes it possible to improve the concentration of people present in the target area.

FIG. 1 is a diagram showing the overall configuration of an equipment control system according to a first embodiment. FIG. 1 is a block diagram showing a hardware configuration of a learning device according to a first embodiment. FIG. 1 is a block diagram showing a functional configuration of a learning device according to a first embodiment. FIG. 1 is a diagram showing an example of learning data according to the first embodiment; FIG. 10 is a diagram showing the relationship between concentration level and biological information according to the first embodiment. FIG. 1 is a diagram schematically illustrating a neural network that is an example of supervised learning according to the first embodiment. A flowchart showing the flow of a trained model generation process executed by the learning device according to the first embodiment. FIG. 1 is a block diagram showing a hardware configuration of an equipment control device according to a first embodiment. FIG. 1 is a block diagram showing a functional configuration of an equipment control device according to a first embodiment. FIG. 10 is a diagram showing an example of transition of a concentration state in the first embodiment. 1 is a flowchart showing the flow of equipment control processing executed by the equipment control device according to the first embodiment. A flowchart showing the flow of the control content determination process in the equipment control process shown in FIG. 11 . FIG. 10 is a block diagram showing the functional configuration of an equipment control device according to a second embodiment. FIG. 10 is a diagram showing the configuration of a trained model according to a second embodiment. FIG. 10 is a diagram showing the overall configuration of an equipment control system according to a third embodiment. FIG. 10 is a diagram showing an example of an output of a concentration degree estimated by an equipment control device according to a third embodiment. FIG. 10 is a diagram showing the overall configuration of an equipment control system according to a modification of the third embodiment.

Embodiments will be described in detail below with reference to the drawings. Note that identical or equivalent parts in the drawings will be designated by the same reference numerals.

(Embodiment 1)
1 shows the overall configuration of an equipment control system 1 according to a first embodiment. The equipment control system 1 is a system that estimates the concentration of people in a target area 5 and controls equipment 40 based on the estimation results. The equipment control system 1 includes a learning device 10, a sensor device 20, an equipment control device 30, and equipment 40.

The target area 5 is a space where people work, do tasks, study, etc., such as a shared office, working space, study room, library, or study room. A sensor device 20 and facility equipment 40 are installed in the target area 5. The facility equipment 40 is equipment installed in the target area 5. The facility equipment 40 is equipment that affects the concentration of people present in the target area 5, such as an air conditioner indoor unit, ventilation equipment, lighting equipment, etc.

<Learning phase>
The learning device 10 is a device that learns the relationship between a person's biometric information and concentration level. The learning device 10 is realized by an information processing device such as a personal computer, a server, or a tablet. As shown in FIG. 2 , the learning device 10 includes a control unit 11, a storage unit 12, an operation reception unit 13, a display unit 14, and a communication unit 15.

The control unit 11 includes a CPU (Central Processing Unit), ROM (Read Only Memory), and RAM (Random Access Memory). The CPU is also called a central processing unit, central arithmetic unit, processor, microprocessor, or microcomputer, and functions as a central processing unit that executes processes and calculations related to the control of the learning device 10. In the control unit 11, the CPU reads programs and data stored in the ROM and uses the RAM as a work area to perform overall control of the learning device 10.

The storage unit 12 is equipped with non-volatile semiconductor memory such as flash memory, EPROM (Erasable Programmable ROM), or EEPROM (Electrically Erasable Programmable ROM), and serves as a so-called secondary storage device or auxiliary storage device. The storage unit 12 stores programs and data used by the control unit 11 to perform various processes. It also stores data generated or acquired by the control unit 11 as a result of performing various processes.

The operation reception unit 13 is equipped with input devices such as a keyboard, mouse, switches, touchpad, and touch panel, and receives operations from the user. The user can input various instructions to the learning device 10 by operating the operation reception unit 13. When the operation reception unit 13 receives an operation instruction input by the user, it transmits the received operation instruction to the control unit 11.

The display unit 14 includes a display device such as an LCD (Liquid Crystal Display) panel or an organic EL (Electro-Luminescence) panel. The display unit 14 is driven by a display drive circuit (not shown) and displays various images under the control of the control unit 11.

The communication unit 15 has a communication interface that enables the learning device 10 to communicate with external devices. For example, the communication unit 15 communicates with external devices in accordance with well-known communication standards such as LAN (Local Area Network), USB (Universal Serial Bus), and the Internet.

Next, the functional configuration of the learning device 10 will be described with reference to Figure 3. As shown in Figure 3, the learning device 10 functionally comprises a collection unit 110 and a model generation unit 120. Each of these functions is realized by software, firmware, or a combination of software and firmware. The software and firmware are written as programs and stored in the ROM or memory unit 12. The CPU then executes the programs stored in the ROM or memory unit 12 to realize each of these functions.

The learning device 10 also includes a learning database 180 and a model storage unit 190. These are constructed in appropriate storage areas of the storage unit 12.

The learning database 180 stores learning data 60. The learning data 60 is data for learning the relationship between the biometric information of the learning subject and the corresponding concentration level. Here, biometric information is information that indicates the biological state of the subject. Specifically, the biometric information includes respiratory information related to the subject's breathing, heart rate information related to the subject's heart rate, and body movement information related to the subject's body movements. Furthermore, the concentration level is information that indicates the subject's level of concentration, in other words, the degree to which the subject is paying attention to work, tasks, studies, etc.

As shown in Figure 4, the learning data 60 is data in which the biometric information of multiple people to be learned is linked to the corresponding concentration level. For ease of understanding, in Figure 4, the multiple sets of biometric information stored in the learning database 180 are represented by serial numbers such as 1, 2, 3, ... for each set of biometric information. The learning data 60 stores each of the multiple sets of biometric information, linking it to a concentration level of "high," "medium," or "low."

As shown in Figure 5, there is a correlation between a person's level of concentration at a given time and that person's biometric information at that time. Specifically, when a person's level of concentration is low, that person's body movements tend to increase, their heart rate fluctuations tend to increase, and their breathing rate tends to increase. Conversely, when a person's level of concentration is high, their body movements tend to decrease, their heart rate fluctuations tend to decrease, and their breathing rate tends to decrease. Therefore, focusing on this correlation, the learning device 10 uses machine learning techniques to generate a trained model 70 for estimating the level of concentration from biometric information. The learning data 60 defines training data that serves as the correct answer, allowing the learning device 10 to learn the relationship between biometric information and the level of concentration.

Returning to FIG. 3, the collection unit 110 collects biometric information and concentration data to generate learning data 60. First, the collection unit 110 communicates with the sensor device 20 via the communication unit 15 and collects biometric information data from the sensor device 20. As shown in FIG. 1, the sensor device 20 is installed in the target area 5 and measures the biometric information of people present in the target area 5.

The sensor device 20 is equipped with a Doppler sensor 21. The Doppler sensor 21 transmits electromagnetic waves, such as millimeter waves or microwaves, toward a target and receives the reflected waves of the transmitted electromagnetic waves from the target. The Doppler sensor 21 then measures the frequency and amplitude of the reflected waves. By comparing the frequency and amplitude of the transmitted electromagnetic waves with the received reflected waves, the Doppler sensor 21 can measure the distance from the Doppler sensor 21 to the target, the speed of the target, and the angle of the target. Furthermore, the Doppler sensor 21 can measure minute movements of the target based on phase changes in the reflected waves. The target is a person present within the target area 5.

Although not shown, the sensor device 20 is equipped with a control unit including a CPU, ROM, and RAM, a readable/writable non-volatile memory, and a communications interface. The control unit calculates biometric information of people present in the target area 5 based on measurement data of distance, speed, angle, and their fluctuations measured by the Doppler sensor 21. Here, the biometric information includes respiratory information related to the person's breathing, heart rate information related to the person's heart rate, and body movement information related to the person's body movements.

Respiratory information is information about movements caused by a person's breathing. Specifically, respiratory information includes the number of breaths per unit time and the amount of variation. The control unit of the sensor device 20 calculates respiratory information by analyzing the frequency of the reflected waves and extracting characteristic frequency components caused by movements of the person's thoracic and abdominal body surfaces.

Heartbeat information is information about the movement caused by a person's heartbeat. Specifically, heartbeat information includes the number of heartbeats per unit time and the amount of fluctuation. The control unit of the sensor device 20 calculates heartbeat information by analyzing the person's frequency and extracting characteristic frequency components caused by the movement of the person's heart.

Body movement information is information on the body movements of a target object that are larger than breathing and heartbeat, such as when a person changes posture, stands up, sits down, or moves. Specifically, body movement information includes the magnitude of body movement and its fluctuation amount. The control unit of the sensor device 20 calculates body movement information by analyzing the amplitude of the reflected wave and extracting components whose amplitude is equal to or greater than a threshold value.

The method for calculating such respiratory information, heart rate information, and body movement information from the measurement data of the Doppler sensor 21 can be a well-known method described in prior art documents such as Patent Document 1.

For ease of understanding, the following description will take as an example a case in which the collection unit 110 collects data on the respiratory rate per unit time as respiratory information, data on the heart rate per unit time as heart rate information, and data on the magnitude of body movement as body movement information.

The sensor device 20 acquires such biometric information from multiple people present in the target area 5. Furthermore, even if the biometric information is from the same person, the sensor device 20 acquires it as different biometric information if the biometric information is acquired at different times. The collection unit 110 collects from the sensor device 20 multiple sets of biometric information to be learned in this way.

Second, the collection unit 110 collects concentration level data. The collection unit 110 collects concentration level data from multiple people whose biometric information is the learning subject, based on reports from these people. Specifically, the multiple people whose biometric information is collected operate the operation reception unit 13 of the learning device 10 or an appropriate terminal capable of communicating with the communication unit 15 of the learning device 10 to input the concentration level at the time the biometric information was collected. For example, each person selects and inputs the concentration level at the time the biometric information was collected from multiple levels, such as "high," "medium," or "low." The collection unit 110 collects the concentration levels input in this way from multiple people as the concentration levels corresponding to the biometric information collected as the learning subject.

Note that the collection unit 110 is not limited to collecting biometric information data and corresponding concentration level data at the same time. In other words, the collection unit 110 may collect biometric information data and corresponding concentration level data at different times, as long as they can be associated with each other.

In this way, the collection unit 110 collects biometric information and concentration level data for multiple people who are the learning targets. Once the data is collected, the collection unit 110 performs a process for each of the multiple sets of biometric information to link the biometric information included in the collected data with the concentration level of the person from whom the biometric information was obtained. As a result, the collection unit 110 generates learning data 60 in which each of the multiple sets of biometric information is linked one-to-one with the corresponding concentration level, as shown in FIG. 4, for example.

When the collection unit 110 generates the training data 60, it stores the generated training data 60 in the training database 180. The collection unit 110 is realized by the control unit 11 working in cooperation with the communication unit 15. The collection unit 110 is an example of an acquisition means.

Returning to Figure 3, the model generation unit 120 learns the relationship between multiple sets of biometric information to be learned and the corresponding concentration levels based on the learning data 60 stored in the learning database 180. In this way, the model generation unit 120 generates a trained model 70. The trained model 70 is a model that outputs an estimation result of the concentration level of a person to be estimated in response to input of the biometric information of the person to be estimated.

The model generation unit 120 learns the relationship between multiple sets of biometric information to be learned and the corresponding concentration levels by performing machine learning according to a predetermined learning algorithm. The model generation unit 120 can use known algorithms such as supervised learning, unsupervised learning, and reinforcement learning as the learning algorithm. Supervised learning is a method of learning the relationship between input data and output data by providing combinations of input data and corresponding correct output data as training data. Unsupervised learning is a method of learning the characteristics of input data without providing training data.

Below, we will explain the application of a neural network, a type of supervised learning algorithm, as an example. A neural network is a method of representing the neural circuit network in the human brain using a mathematical model called an artificial neuron.

As shown in Figure 6, a neural network is a model that includes an input layer, an output layer, and at least one intermediate layer, and outputs output data from the output layer in response to input data input to the input layer. The intermediate layer is also called a hidden layer. Note that while Figure 6 shows one intermediate layer, there may be two or more. Each layer is made up of multiple nodes. The nodes are also called neurons.

Nodes X1-X3 in the input layer receive as input data the respiratory rate, heart rate, and magnitude of body movement contained in the biometric information. Nodes Z1-Z3 in the output layer correspond to three levels of concentration as output data: "high," "medium," and "low." The number of nodes X1-X3 in the input layer and the number of nodes Z1-Z3 in the output layer correspond to the number of input data and the number of output data, respectively. Therefore, as the amount of input data and output data increases, the number of nodes in the input and output layers also increases accordingly.

When input data is input to nodes X1-X3 in the input layer, the input value is multiplied by weights W11-W16 and input to nodes Y1-Y2 in the hidden layer, and this value is then further multiplied by weights W21-W26 and output from nodes Z1-Z3 in the output layer. The output result from the output layer changes depending on the values of weights W11-W16 and W21-W26.

Each node Z1 to Z3 in the output layer outputs a score that is a value between 0 and 1. The score is a value that indicates the likelihood that the input data input to the input layer corresponds to the concentration level associated with each node Z1 to Z3. The closer the score is to 1, the higher the probability that the input data input to the input layer corresponds to the concentration level associated with that node.

The model generation unit 120 uses the training data 60 stored in the training database 180 as training data to adjust the connection weights W11-W16 and W21-W26 of each layer in the neural network. Specifically, the model generation unit 120 inputs sets of respiratory rate, heart rate, and body movement magnitude contained in each piece of biometric information in the training data 60 as input data to the input layer. The model generation unit 120 then adjusts the connection weights W11-W16 and W21-W26 of each layer using backpropagation or the like so that, among the scores output from each node Z1-Z3 in the output layer, the score output from the correct node approaches 1 and the scores output from the other nodes approach 0. Here, the correct node is the node in the training data 60 that corresponds to the concentration level linked to the biometric information input to the input layer.

The model generation unit 120 performs this process for each of the multiple sets of biometric information and concentration level combinations included in the training data 60, thereby optimizing the connection weights W11-W16, W21-W26 of each layer in the neural network. In this way, the model generation unit 120 constructs a neural network that receives input biometric information and outputs a score indicating the likelihood that the biometric information corresponds to one of three levels of concentration: "high," "medium," or "low."

Once the model generation unit 120 constructs a neural network, it stores the constructed neural network in the model storage unit 390 as a trained model 70. The model generation unit 120 is realized by the control unit 11 working in cooperation with the storage unit 12. The model generation unit 120 functions as a model generation means.

Next, with reference to Figure 7, the process for generating the trained model 70 executed by the learning device 10 will be described. By executing the process for generating the trained model 70 shown in Figure 7, a method for generating the trained model 70 is realized. When the control unit 11 of the learning device 10 receives a user instruction via the operation reception unit 13, it starts the process for generating the trained model 70 shown in Figure 7.

When the process of generating the trained model 70 begins, the control unit 11 functions as the collection unit 110 and collects biometric information and concentration level data (step S11). Specifically, the control unit 11 collects, as biometric information data, breathing information, heart rate information, and body movement information of multiple people who are the learning subjects from the sensor device 20. The control unit 11 also collects concentration level data based on reports from the multiple people who are the learning subjects.

Once the data is collected, the control unit 11 associates the collected biometric information with the corresponding concentration level to generate learning data 60 (step S12). The control unit 11 generates learning data 60 such as that shown in FIG. 4, for example, and stores it in the learning database 180.

Once the training data 60 is generated, the control unit 11 functions as the model generation unit 120 and generates a trained model 70 by learning the relationship between biometric information and concentration level using the generated training data 60 (step S13). Specifically, the control unit 11 performs machine learning using multiple sets of combinations of biometric information and concentration level included in the training data 60 as training data, and generates a trained model 70 that outputs an estimated result of concentration level in response to input biometric information.

Once the trained model 70 has been generated, the control unit 11 stores the generated trained model 70 in the model storage unit 190 (step S14). This completes the trained model generation process shown in Figure 7.

<Utilization phase>
Next, a process for utilizing the trained model 70 generated by the training device 10 will be described.

The equipment control device 30 shown in FIG. 1 is a device that controls equipment 40 installed in the target area 5. The equipment control device 30 is realized by an information processing device such as a personal computer, server, or tablet. The equipment control device 30 estimates the concentration of people present in the target area 5 based on the learning results of the learning device 10, and controls the equipment 40. As shown in FIG. 8, the equipment control device 30 includes a control unit 31, a memory unit 32, an operation reception unit 33, a display unit 34, and a communication unit 35.

The control unit 31 includes a CPU, ROM, and RAM. The CPU is also called a central processing unit, central arithmetic unit, processor, microprocessor, microcomputer, etc., and functions as a central processing unit that executes processing and calculations related to the control of the equipment control device 30. In the control unit 31, the CPU reads programs and data stored in the ROM and uses the RAM as a work area to perform overall control of the equipment control device 30.

The storage unit 32 is equipped with non-volatile semiconductor memory such as flash memory, EPROM, or EEPROM, and serves as a so-called secondary storage device or auxiliary storage device. The storage unit 32 stores programs and data used by the control unit 31 to perform various processes. It also stores data generated or acquired by the control unit 31 as a result of performing various processes.

The operation reception unit 33 is equipped with input devices such as a keyboard, mouse, switches, touchpad, and touch panel, and receives operations from the user. By operating the operation reception unit 33, the user can input various instructions to the equipment control device 30. When the operation reception unit 33 receives an operation instruction input by the user, it transmits the received operation instruction to the control unit 31.

The display unit 34 includes a display device such as an LCD panel or organic EL. The display unit 34 is driven by a display drive circuit (not shown) and displays various images under the control of the control unit 31.

The communication unit 35 has a communication interface that enables the equipment control device 30 to communicate with external devices. For example, the communication unit 35 communicates with external devices in accordance with well-known communication standards such as LAN, USB, and the Internet.

Next, the functional configuration of the equipment control device 30 will be described with reference to Figure 9. As shown in Figure 9, the equipment control device 30 functionally comprises an acquisition unit 310, an estimation unit 320, an output unit 330, and an equipment control unit 340. Each of these functions is realized by software, firmware, or a combination of software and firmware. The software and firmware are written as programs and stored in the ROM or memory unit 32. The CPU then executes the programs stored in the ROM or memory unit 12 to realize each of these functions.

The equipment control device 30 also includes a model storage unit 390 that stores the trained model 70. The model storage unit 390 is constructed in an appropriate storage area of the storage unit 32. After the trained model 70 is generated in the learning device 10, it is stored in the model storage unit 390 via communication by the communication unit 35 or via an external recording medium.

The functions of each part of the equipment control device 30 other than the equipment control unit 340 can be collectively referred to as the estimation device 50.

The acquisition unit 310 acquires biometric information of people present in the target area 5. As described above, biometric information is information that indicates a person's biological state, and specifically includes respiratory information related to the person's breathing, heart rate information related to the person's heart rate, and body movement information related to the person's body movement. As in the learning phase, the sensor device 20 measures the biometric information of people present in the target area 5 using the Doppler sensor 21. The acquisition unit 310 communicates with the sensor device 20 via the communication unit 35 and acquires the biometric information from the sensor device 20. The acquisition unit 310 is realized by the control unit 31 cooperating with the communication unit 35. The acquisition unit 310 is an example of an acquisition means.

More specifically, if a person is present in the target area 5, the acquisition unit 310 acquires the person's biometric information. Here, whether or not a person is present in the target area 5 is determined using an appropriate method. For example, the acquisition unit 310 communicates with an information terminal installed in the target area 5 via the communication unit 35, and determines that a person is present in the target area 5 when the information terminal is operated. Alternatively, a human presence sensor may be installed in the target area 5, and the acquisition unit 310 may determine that a person is present in the target area 5 when a human presence sensor detects a person.

The estimation unit 320 estimates the concentration level of people present in the target area 5 from the biometric information acquired by the acquisition unit 310, based on the trained model 70. Specifically, the estimation unit 320 inputs the biometric information acquired by the acquisition unit 310 into the trained model 70. As described above, the trained model 70 outputs a score indicating the likelihood that the concentration level of the person to be estimated corresponds to one of three levels: "high," "medium," or "low," for the input biometric information.

Of these three levels, the level with the highest score output by the trained model 70 is most likely to correspond to the concentration level of the person being estimated. Therefore, the estimation unit 320 compares the scores for each of the "high," "medium," and "low" levels output by the trained model 70, and estimates that the level with the highest score corresponds to the concentration level of the person being estimated.

For example, if the "high" score is the highest among the scores for each stage output from the trained model 70 for input biometric information, the estimation unit 320 estimates that the person from whom the biometric information was obtained has a high level of concentration. On the other hand, if the "low" score is the highest among the scores for each stage output from the trained model 70 for input biometric information, the estimation unit 320 estimates that the person from whom the biometric information was obtained has a low level of concentration. The estimation unit 320 is realized by the control unit 31 working in cooperation with the memory unit 32. The estimation unit 320 is an example of an estimation means.

The output unit 330 outputs the estimation results estimated by the estimation unit 320. The output unit 330 generates an image showing the concentration level estimated by the estimation unit 320 and displays it on the display unit 34. By checking the display unit 34, the user can know the concentration level of people present in the target area 5.

Alternatively, the output unit 330 may output output information indicating the concentration level estimated by the estimation unit 320 to an external device via the communication unit 35 and display it on a display unit of the external device. The output unit 330 is realized by the control unit 31 working in cooperation with the display unit 34 or the communication unit 35. The output unit 330 is an example of an output means.

The equipment control unit 340 controls the equipment 40 based on the concentration level estimated by the estimation unit 320. First, the equipment control unit 340 determines the control content of the equipment 40 based on the concentration level estimated by the estimation unit 320.

To determine the control content, the equipment control unit 340 defines the concentration state of the person to be estimated as a first state, a high concentration state, and a second state, a low concentration state, which is lower in concentration than the first state, as shown in FIG. 10. When a person's concentration level is in a low concentration state, if the state in which the person's concentration level estimated by the estimation unit 320 is "high" continues for a predetermined period of time or longer, the person's concentration level changes from the low concentration state to the high concentration state. Conversely, when a person's concentration level is in a high concentration state, if the state in which the person's concentration level estimated by the estimation unit 320 is "low" continues for a predetermined period of time or longer, the person's concentration level changes from the high concentration state to the low concentration state.

The equipment control unit 340 measures the duration of the concentration level estimated by the estimation unit 320. Then, based on the duration of the concentration level, the equipment control unit 340 determines whether the current concentration level of the person being estimated is in a low concentration state or a high concentration state, and determines the control content based on the determination result.

Specifically, when the concentration level estimated by the estimation unit 320 changes from a high concentration state to a low concentration state, the equipment control unit 340 causes the equipment 40 to perform a stimulating operation that stimulates people present in the target area 5. Similarly, when the concentration level estimated by the estimation unit 320 remains in a low concentration state for longer than a predetermined threshold time, the equipment control unit 340 causes the equipment 40 to perform a stimulating operation. Here, stimulating operation refers to an operation that stimulates people present in the target area 5 to wake them up and refresh their minds.

As a first example, if the facility equipment 40 is an indoor unit of an air conditioner, the stimulating operation is an operation that changes the blowing direction of the conditioned air blown out from the indoor unit, or an operation that lowers the temperature of the conditioned air. An example of an operation that changes the blowing direction of the conditioned air is an operation that swings the airflow. This changes the feeling of wind blowing, which can stimulate people in the target area 5. Lowering the temperature of the conditioned air can also stimulate people, which can increase their concentration.

As a second example, if the equipment 40 is a ventilation device, the stimulation operation is an operation that increases the ventilation intensity of the ventilation device. By increasing the ventilation intensity, the carbon dioxide concentration in the target area 5 can be reduced, thereby refreshing the mood of people present in the target area 5 and increasing their concentration.

As a third example, if the facility equipment 40 is a lighting device, the stimulating operation is an operation that changes the color of the lighting provided by the lighting device. Changing the color of the lighting can change a person's mood, which is expected to have the effect of increasing concentration. An example of an operation that changes the color of the lighting is an operation that changes the color of the lighting to a warm color. Warm colors tend to stimulate people's autonomic nervous system, so changing the color of the lighting to a warm color can more effectively increase concentration.

The equipment control unit 340 determines the control content so that the equipment 40 continues this stimulating operation for a certain period of time, and then returns to its original state. In this way, by temporarily changing the environment to one that makes it easier for people in the target area 5 to change their mood, and then returning to the original control state, it is possible to encourage subsequent concentration.

Furthermore, the equipment control unit 340 also causes the equipment 40 to perform a stimulating operation if the concentration level estimated by the estimation unit 320 remains in a high concentration state for a period of time longer than a predetermined threshold time. Even when concentration is high, the stimulating operation can be used to temporarily refresh the user, making it easier to maintain a high concentration state. Note that the content of the stimulating operation may be different when the concentration level changes from a high concentration state to a low concentration state and when the concentration level remains in a high concentration state for a period of time longer than the threshold time.

These control switching conditions, control details in low-concentration states, and control details in high-concentration states can be freely set. For example, the user may be able to select and set these, and the equipment control unit 340 may determine the control details according to the switching conditions or control details set by the user.

Once the control content has been determined, the equipment control unit 340 operates the equipment 40 according to the determined control content. Specifically, the equipment control unit 340 communicates with the equipment 40 via the communication unit 35 and sends a command to the equipment 40 to execute the determined control content. As a result, the equipment control unit 340 causes the equipment 40 to operate in accordance with the concentration level estimated by the estimation unit 320. The equipment control unit 340 is realized by the control unit 31 working in cooperation with the communication unit 35. The equipment control unit 340 is an example of equipment control means.

Next, the equipment control process executed by the equipment control device 30 will be described with reference to Figure 11. The control unit 31 of the equipment control device 30 executes the equipment control process shown in Figure 11 when the trained model 70 generated by the learning device 10 is stored in the model storage unit 390.

When the equipment control process starts, the control unit 31 determines whether or not a person is present in the target area 5 (step S31). Specifically, the control unit 31 determines whether or not a person is present in the target area 5 based on whether or not an information terminal installed in the target area 5 has been operated, or whether or not a person has been detected by a human presence sensor installed in the target area 5.

If no person is present in the target area 5 (step S31; NO), the control unit 31 remains in step S31 and waits until a person enters the target area 5.

On the other hand, if a person is present in the target area 5 (step S31; YES), the control unit 31 functions as the acquisition unit 310 and acquires biometric information of the person present in the target area 5 (step S32). Specifically, the control unit 31 acquires, as biometric information, breathing information, heart rate information, and body movement information of the person present in the target area 5 from the sensor device 20.

Once the biometric information is acquired, the control unit 31 functions as the estimation unit 320 and estimates the concentration level from the acquired biometric information based on the trained model 70 (step S33). Specifically, the control unit 31 inputs the biometric information acquired in step S31 to the trained model 70. The control unit 31 then compares the multiple levels of concentration level scores output by the trained model 70 in response to the input biometric information, and estimates the concentration level with the highest score to be the concentration level of the person being estimated.

Once the concentration level has been estimated, the control unit 31 functions as the output unit 330 and outputs the estimation result (step S34). The control unit 31 notifies the user by displaying an image indicating the estimated concentration level on the display unit 34.

After outputting the estimation result, the control unit 31 determines the control content of the equipment 40 based on the estimated concentration level (step S35). Details of the control content determination process in step S35 will be described with reference to Figure 12.

When the control content determination process starts, the control unit 31 determines the state of concentration estimated in step S33 (step S351). Specifically, the control unit 31 measures the duration of the estimated concentration and determines whether the current concentration state is a high concentration state or a low concentration state as shown in Figure 10.

After determining the state of concentration, the control unit 31 determines whether the estimated concentration level has changed from a high concentration state to a low concentration state, whether the low concentration state has continued for longer than the threshold time, and whether the high concentration state has continued for longer than the threshold time (steps S352 to S354). If the estimated concentration level has not changed from a high concentration state to a low concentration state, the low concentration state has not continued for longer than the threshold time, and the high concentration state has not continued for longer than the threshold time (steps S352 to S354; NO in all cases), the control unit 31 determines the control content to be normal operation (step S355).

On the other hand, if the estimated concentration level changes from a high concentration state to a low concentration state (step S352; YES), if the low concentration state continues for longer than the threshold time (step S353; YES), or if the high concentration state continues for longer than the threshold time (step S354; YES), the control unit 31 determines the control content to be stimulation operation (step S356). This concludes the control content determination process shown in Figure 12.

Returning to FIG. 11, once the control content has been determined, the control unit 31 functions as the equipment control unit 340 and controls the equipment 40 according to the determined control content (step S36). As a result, the control unit 31 causes the equipment 40 to operate in accordance with the concentration level estimated by the estimation unit 320.

The control unit 31 then returns the process to step S31 and continues to determine whether or not a person is present in the target area 5. If a person is present, the control unit 31 executes the processes of steps S32 to S36 again. In this way, while a person is present in the target area 5, the control unit 31 repeats the process of estimating the concentration level from the person's biometric information and controlling the equipment 40 based on the estimated concentration level. This completes the equipment control process shown in FIG. 11.

As described above, in the equipment control system 1 according to embodiment 1, the learning device 10 generates a trained model 70 by learning the relationship between biometric information and the corresponding concentration level. The equipment control device 30 then estimates the concentration level of a person present in the target area 5 from the person's biometric information based on the trained model 70, and controls the equipment 40 based on the estimated concentration level. This makes it possible to control the equipment 40 according to the state of the person present in the target area 5, thereby improving the concentration of the person present in the target area 5. As a result, users of the equipment control system 1 can maintain a high level of concentration while working, performing tasks, studying, etc. in the target area 5.

In addition, in the equipment control system 1 according to embodiment 1, biometric information of people present in the target area 5 is acquired using a Doppler sensor 21. By using the Doppler sensor 21, biometric information can be acquired contactlessly, and people do not need to wear the sensor, so the data necessary for stimulation can be acquired without disrupting people's concentration.

(Embodiment 2)
Next, a second embodiment of the present disclosure will be described. Descriptions of the same configurations and functions as those of the first embodiment will be omitted as appropriate.

As shown in FIG. 13, the equipment control device 30a according to the second embodiment functionally comprises an acquisition unit 310, an estimation unit 320, an output unit 330, an equipment control unit 340, and an identification unit 350. The functions other than those of the identification unit 350 are the same as those of the first embodiment.

The identification unit 350 identifies people present in the target area 5. When a user starts using the equipment control system 1, the user operates the operation reception unit 33 of the equipment control device 30a or an information terminal capable of communicating with the communication unit 35 of the equipment control device 30a to input user information such as the user's name, account name, and ID. The identification unit 350 accepts the user information input by the user in this manner, thereby identifying people present in the target area 5. The identification unit 350 is realized by the control unit 31 working in cooperation with the operation reception unit 33 or communication unit 35. The identification unit 350 is an example of an identification means.

In embodiment 2, the trained model 70 includes a plurality of individual models 71, as shown in FIG. 14. Each of the multiple individual models 71 is a model generated by learning the relationship between the biometric information of different people and the corresponding concentration level, and is an independent model that outputs an estimated concentration level in response to input biometric information.

In the learning device 10 according to embodiment 2, the model generation unit 120 learns the relationship between the biometric information collected by the collection unit 110 and the concentration level, separately for each person for whom the biometric information and concentration level were collected. As a result, the model generation unit 120 generates multiple individual models 71, and stores the multiple generated individual models 71 together as a trained model 70 in the model storage unit 190.

The estimation unit 320 estimates the concentration level from the biometric information acquired by the acquisition unit 310, based on the individual model 71 of the multiple individual models 71 that corresponds to the person identified by the identification unit 350. Specifically, the estimation unit 320 inputs the biometric information acquired by the acquisition unit 310 into the individual model 71 of the multiple individual models 71 that corresponds to the person identified by the identification unit 350. The estimation unit 320 then compares the scores of each stage output by the individual model 71, and estimates that the stage with the highest score corresponds to the concentration level of the person present in the target area 5 identified by the identification unit 350.

In this way, the model used to estimate the concentration level is switched depending on the person, improving the accuracy of the concentration level estimation. As a result, the concentration level is improved to suit the person, allowing the facility equipment 40 to be controlled accurately.

In addition, each of the multiple individual models 71 may be updated based on biometric information data acquired while the corresponding person is using the equipment control system 1, as well as reported data such as the level of concentration during that time and satisfaction with control.

The identification unit 350 may also identify people present in the target area 5 based on the biometric information acquired by the acquisition unit 310. For example, if the target area 5 is an area used only by a small number of specific people, such as a home study or study room, the identification unit 350 accumulates data on the biometric information acquired in the target area 5, extracts similar patterns from the biometric information, and stores them in the memory unit 32. When operation of the equipment control system 1 begins, the identification unit 350 determines, from multiple patterns of biometric information pre-stored in the memory unit 32, the pattern that is most similar to the pattern of biometric information acquired by the acquisition unit 310. The identification unit 350 then identifies the user corresponding to the determined pattern as a person present in the target area 5. By identifying people from biometric information in this way, input work is not required when operation begins.

(Embodiment 3)
Next, a third embodiment of the present disclosure will be described. Descriptions of the same configurations and functions as those of the first and second embodiments will be omitted as appropriate.

As shown in FIG. 15, the equipment control system 1a according to the third embodiment includes a learning device 10, multiple sensor devices 20, an equipment control device 30, and multiple equipment devices 40. In the third embodiment, the target area 5 is divided into multiple individual areas 5a-5c. A sensor device 20 and equipment devices 40 are installed in each of the multiple individual areas 5a-5c. Each of the multiple individual areas 5a-5c corresponds to the seat of each user in, for example, a shared office, a study room, etc.

Each of the multiple sensor devices 20 is similar to the sensor device 20 described in embodiment 1. Furthermore, each of the multiple facility devices 40 is similar to the facility devices 40 described in embodiment 1. In other words, in embodiment 3, the sensor devices 20 and facility devices 40 are installed in each of the multiple individual areas 5a to 5c.

The acquisition unit 310 acquires biometric information of people present in each of the multiple individual areas 5a-5c within the target area 5 from the sensor devices 20 installed in each of the multiple individual areas 5a-5c. The estimation unit 320 estimates the concentration level of people present in each of the multiple individual areas 5a-5c from the biometric information acquired by the acquisition unit 310 based on the trained model 70. In other words, the estimation unit 320 executes a process for each of the multiple individual areas 5a-5c to estimate the concentration level of people present in an individual area from the biometric information acquired from the sensor devices 20 installed in that individual area. The equipment control unit 340 controls the equipment devices 40 installed in each of the multiple individual areas 5a-5c based on the concentration level estimated by the estimation unit 320. In other words, the equipment control unit 340 executes a process for each of the multiple individual areas 5a-5c to control the equipment devices 40 installed in an individual area based on the concentration level of people present in that individual area. The processing by the acquisition unit 310, estimation unit 320, and equipment control unit 340 in each individual area is the same as in embodiment 1.

The output unit 330 outputs the estimation result by the estimation unit 320 of the concentration level of people present in each of the multiple individual areas 5a-5c, and the control status of the equipment 40 installed in each of the multiple individual areas 5a-5c. Specifically, as shown in FIG. 16, the output unit 330 displays an image showing the concentration level estimation result and the control status of the equipment 40 on the display unit 34. More specifically, the output unit 330 displays the concentration level of people present in each of the multiple individual areas 5a-5c estimated by the estimation unit 320 in three levels: "high," "medium," and "low." The output unit 330 also displays whether the current control status of the equipment 40 installed in each of the multiple individual areas 5a-5c is "normal operation" or "stimulation operation," as described in embodiment 2.

By displaying the concentration levels and control status of multiple areas in this way, seat management and seat selection can be carried out efficiently in shared offices, study rooms, etc. For example, when another user selects a seat, they can refer to the current concentration level and control status of each area. Also, when allocating a seat to a new user, it becomes possible to prioritize allocation of a seat near a seat in a stimulating driving state over a seat with a high concentration level. As a result, user convenience is improved.

The output unit 330 may output the display image shown in FIG. 16 to an external device via the communication unit 35. The external device may be, for example, an information terminal such as a smartphone, tablet terminal, or personal computer.

As described above, in the equipment control system 1a according to the third embodiment, the concentration levels and equipment devices 40 in multiple individual areas 5a-5c are managed collectively by a single equipment control device 30. This allows the equipment control system 1a to be constructed efficiently.

Note that in embodiment 3, the sensor device 20 is not limited to being installed in each of the multiple individual areas 5a-5c. For example, as in the equipment control system 1b shown in FIG. 17, a single sensor device 20 may be used to acquire biometric information of people present in each of the multiple individual areas 5a-5c. In particular, the Doppler sensor 21 can acquire measurement data from multiple objects without contact. Therefore, even if a sensor device 20 is not installed in each area, the measurement range of a single sensor device 20 can include multiple individual areas 5a-5c. In this way, by managing multiple individual areas 5a-5c with a single sensor device 20, the configuration of the equipment control system 1b can be made more efficient.

(Modification)
Although the embodiments have been described above, it is possible to combine the embodiments, or to modify or omit the embodiments as appropriate.

For example, in the above embodiment, the learning device 10 and the equipment control device 30 used information on respiratory rate, heart rate, and magnitude of body movement as biometric information. However, the learning device 10 and the equipment control device 30 may use information other than these as biometric information. For example, instead of or in addition to respiratory rate, heart rate, and magnitude of body movement, the amount of fluctuation of these over time may be used as biometric information. Normally, heart rate, respiratory rate, body movement, and the amount of fluctuation of these over time differ from person to person, but by using data on the amount of fluctuation over time, the influence of such individual differences can be reduced, thereby further improving the accuracy of estimating the level of concentration.

Alternatively, the model generation unit 120 may generate the trained model 70 by employing a learning algorithm that can learn including time-series changes, such as an RNN (Recurrent Neural Network) or an LSTM (Long-Short Term Model). These learning algorithms can learn including time fluctuations, and therefore have the effect of reducing the influence of individual differences, similar to inputting data on the amount of time fluctuations into the input data.

Furthermore, the low frequency (LF) or high frequency (HF) components of the heart rate may be used as heart rate information. Generally, the low frequency components of the heart rate are components in the frequency band of 0.004 to 0.15 Hz and reflect the activity of both the sympathetic and parasympathetic nervous systems. The high frequency components of the heart rate are components in the frequency band of 0.15 to 0.4 Hz and reflect the activity of the parasympathetic nervous system. In this way, heart rate variability is related to the autonomic nervous system and correlates with a person's level of alertness. Therefore, using the low frequency and high frequency components of the heart rate as heart rate information can further improve the accuracy of estimating the level of concentration.

In the above embodiment, the sensor device 20 acquired biometric information based on measurement data measured by the Doppler sensor 21. However, at least a portion of the biometric information may be acquired by means other than the Doppler sensor 21. For example, heart rate information or body movement information may be acquired by a wristwatch-type heart rate sensor or body movement sensor. When using the Doppler sensor 21, biometric information can be acquired contactlessly, but because it uses the reflection of electromagnetic waves in space, false detections may occur due to the influence of people around the Doppler sensor 21. When using a wristwatch-type sensor, the subject must wear the sensor, but biometric information from the subject can be reliably acquired.

In the above embodiment, the biometric information collected by the collection unit 110 of the learning device 10 and the biometric information acquired by the acquisition unit 310 of the equipment control device 30 were acquired from the same sensor device 20 installed in the same target area 5. However, the collection unit 110 may collect biometric information of multiple people in an area different from the target area 5 from a device other than the sensor device 20 as multiple sets of biometric information data to be learned. By collecting biometric information and concentration data to be learned from various people in various target areas, not just the target area 5 where the equipment control device 30 is installed, it becomes possible to generate a trained model 70 that can estimate concentration levels from biometric information with high accuracy.

A trained model 70 generated by learning on one learning device 10 may be used by multiple different equipment control devices 30. Furthermore, after the equipment control device 30 acquires the trained model 70 from the learning device 10 once, the learning device 10 may perform training again to update the trained model 70, and the updated trained model 70 may be used by the equipment control device 30.

In the above embodiment, the collection unit 110 collected concentration level data based on reports from multiple people whose biometric information was collected as the learning target. However, the collection unit 110 may collect concentration level data using other methods. For example, the sensor device 20 may be equipped with an eye-gaze measurement device or a visible camera, and while collecting the biometric information of the person being studied, the eye-gaze measurement device may measure the person's eye movement, or the visible camera may measure the person's body movement. In this case, the collection unit 110 collects data on the amount of eye-gaze movement or body movement from the sensor device 20 as concentration level data. In the collected data, if the amount of eye-gaze movement or body movement is large, the concentration level is determined to be low, and if the amount of eye-gaze movement or body movement is small, the concentration level is determined to be high.

In the above embodiment, the model generation unit 120 learned the relationship between biometric information and concentration level according to a neural network model, which is a supervised learning algorithm, as a learning algorithm, and generated the trained model 70. However, the model generation unit 120 may also use a supervised learning algorithm other than a neural network. For example, the model generation unit 120 may learn the relationship between biometric information and concentration level according to a method such as genetic programming or a support vector machine, and generate the trained model 70.

Alternatively, the model generation unit 120 may use an unsupervised learning or reinforcement learning algorithm. For example, the model generation unit 120 may use a clustering technique as an unsupervised learning algorithm. In this case, the model generation unit 120 classifies the biometric information into multiple groups according to the characteristics of the respiratory information, heart rate information, and body movement information contained in the biometric information, and associates each group with a level of concentration, such as "high," "medium," or "low." In this way, the model generation unit 120 learns the relationship between the biometric information and the level of concentration. The model generation unit 120 may also use deep learning, a method of learning to extract the features themselves.

In the above embodiment, the learning device 10, sensor device 20, equipment control device 30, and equipment 40 existed as independent devices. However, any of the devices may include the functions of at least one of the other devices. For example, the functions of the sensor device 20 and equipment control device 30 may be included in the equipment 40, and the equipment 40 may estimate the concentration level based on data acquired by the sensor device 20 and control the equipment 40 based on the estimated concentration level. Alternatively, the functions of the equipment control device 30 or learning device 10 may be included in the sensor device 20, or the functions of the learning device 10 may be included in the equipment control device 30.

The functions of the learning device 10 and the equipment control device 30 may also be implemented by a cloud server. Here, the cloud server is a server that provides cloud computing resources. In this case, the sensor device 20 transmits biometric information acquired by the Doppler sensor 21 to the cloud server via a communication network. The cloud server performs machine learning based on the biometric information received from the sensor device 20 and generates a trained model 70. The cloud server also estimates the concentration level from the biometric information received from the sensor device 20, determines the control content for the equipment 40, and transmits it to the equipment 40.

In the above embodiment, the control unit 11 of the learning device 10 functions as each unit shown in FIG. 3 by having the CPU execute a program stored in the ROM or memory unit 12. Furthermore, the control unit 31 of the equipment control device 30 functions as each unit shown in FIG. 9 by having the CPU execute a program stored in the ROM or memory unit 32. However, the control units 11 and 31 may be dedicated hardware. Dedicated hardware includes, for example, a single circuit, a composite circuit, a programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination of these. When the control units 11 and 31 are dedicated hardware, the functions of each unit may be realized by individual hardware, or the functions of each unit may be realized together by a single piece of hardware.

Furthermore, some of the functions of each unit may be realized by dedicated hardware, and other functions may be realized by software or firmware. In this way, the control units 11 and 31 can realize each of the above-mentioned functions by hardware, software, firmware, or a combination of these.

By applying the programs that define the operation of the learning device 10 and the equipment control device 30 to an existing computer such as a personal computer or information terminal device, it is possible to make the computer function as both the learning device 10 and the equipment control device 30.

Furthermore, such programs may be distributed in any manner; for example, they may be stored on a computer-readable recording medium such as a CD-ROM (Compact Disk ROM), DVD (Digital Versatile Disk), MO (Magneto Optical Disk), or memory card, or they may be distributed via a communications network such as the Internet.

This disclosure allows for various embodiments and modifications without departing from the broad spirit and scope of the disclosure. Furthermore, the above-described embodiments are intended to illustrate the disclosure and do not limit the scope of the disclosure. In other words, the scope of the disclosure is defined by the claims, not the embodiments. Various modifications made within the scope of the claims and the meaning of equivalent disclosures are deemed to be within the scope of this disclosure.

1, 1a, 1b Equipment control system, 5 Target area, 5a-5c Individual areas, 10 Learning device, 11, 31 Control unit, 12, 32 Memory unit, 13, 33 Operation reception unit, 14, 34 Display unit, 15, 35 Communication unit, 20 Sensor device, 21 Doppler sensor, 30, 30a Equipment control device, 40 Equipment equipment, 50 Estimation device, 60 Training data, 70 Trained model, 72 Individual model, 110 Collection unit, 120 Model generation unit, 180 Training database, 190 Model storage unit, 310 Acquisition unit, 320 Estimation unit, 330 Output unit, 340 Equipment control unit, 350 Recognition unit, 390 Model storage unit

Claims (12)

An equipment control device that controls equipment installed in a target area,
an acquisition means for acquiring biometric information of a person present in each of a plurality of individual areas within the target area;
an estimation means for estimating a concentration level of the person present in each of the plurality of individual areas from the biometric information acquired by the acquisition means, based on a trained model generated by learning the relationship between the biometric information of a learning target and the corresponding concentration level;
equipment control means for controlling equipment installed in each of the plurality of individual areas based on the concentration degree estimated by the estimation means;
and an output means for outputting the estimation result by the estimation means of the concentration of people present in each of the plurality of individual areas and the control status of the facility equipment installed in each of the plurality of individual areas in association with each individual area .
Equipment control device. An equipment control device that controls equipment installed in a target area,
an acquisition means for acquiring biometric information of a person present in the target area;
an estimation means for estimating a concentration level of the person present in the target area from the biometric information acquired by the acquisition means, based on a trained model generated by learning the relationship between the biometric information of the learning target and the corresponding concentration level;
an equipment control unit that controls the equipment based on the concentration degree estimated by the estimation unit,
the facility control means causes the facility equipment to perform a stimulating operation to stimulate the person present in the target area in each of the following cases: when the concentration level estimated by the estimation means changes from a first state to a second state in which the concentration level is lower than the first state; when the time period during which the concentration level estimated by the estimation means is in the second state continues longer than a threshold time; and when the time period during which the concentration level estimated by the estimation means is in the first state continues longer than a threshold time.
Equipment control device. The facility equipment is an indoor unit of an air conditioner that air-conditions the target area,
The stimulation operation is an operation that changes the blowing direction of the conditioned air blown out from the indoor unit or an operation that lowers the temperature of the conditioned air.
The equipment control device according to claim 2 . The facility equipment is a ventilation device,
The stimulation operation is an operation of increasing the intensity of ventilation by the ventilation device.
The equipment control device according to claim 2 . the facility equipment is lighting equipment,
The stimulation operation is an operation of changing the color of the lighting by the lighting device.
The equipment control device according to claim 2 . the acquiring means acquires the biological information based on measurement data measured by a Doppler sensor.
The equipment control device according to any one of claims 1 to 5 . The biological information includes respiratory information related to the person's breathing, heartbeat information related to the person's heartbeat, and body movement information related to the person's body movement.
The equipment control device according to any one of claims 1 to 6 . The trained model includes a plurality of individual models generated by learning the relationship between different people's biometric information and corresponding concentration levels,
further comprising an identification means for identifying the person present in the target area;
the estimation means estimates a concentration level of the person from the biometric information acquired by the acquisition means, based on a personal model corresponding to the person identified by the identification means, among the plurality of personal models;
The equipment control device according to any one of claims 1 to 7. the identification means identifies the person present in the target area based on the biometric information acquired by the acquisition means.
The equipment control device according to claim 8. An equipment control system comprising the equipment control device according to any one of claims 1 to 9 and a learning device,
The learning device
a model generation means for generating the trained model by learning the relationship;
In the equipment control device,
The estimation means estimates a concentration level of the people present in the target area from the biometric information acquired by the acquisition means, based on the trained model generated by the model generation means.
Equipment control systems. Acquire biometric information of people present in each of a plurality of individual areas within the target area;
estimating a concentration level of the person present in each of the plurality of individual areas from the acquired biometric information based on a trained model generated by learning the relationship between the biometric information of the learning target and the corresponding concentration level;
controlling equipment installed in each of the plurality of individual areas based on the estimated concentration degree;
outputting, for each individual area, an estimation result of the concentration of people present in each of the plurality of individual areas and a control status of the facility equipment installed in each of the plurality of individual areas in association with each other;
Equipment control methods. Computer,
an acquisition means for acquiring biometric information of a person present in each of a plurality of individual areas within the target area;
an estimation means for estimating a concentration level of the person present in each of the plurality of individual areas from the biometric information acquired by the acquisition means, based on a trained model generated by learning the relationship between the biometric information of a learning target and the corresponding concentration level;
equipment control means for controlling equipment installed in each of the plurality of individual areas based on the concentration degree estimated by the estimation means;
and an output means for outputting, for each individual area, an estimation result by the estimation means of the concentration of people present in each of the plurality of individual areas and a control status of equipment installed in each of the plurality of individual areas, in association with each other .
program.