Odor determination device, odor determination method, and odor determination system

Abstract

A smell decision device according to an embodiment is provided with: an acquisition unit which acquires a detection value obtained from each of a plurality of detection elements; a decision unit which decides whether the detection value exceeds a preset threshold; a smell decision unit which decides on the type of smell on the basis of smell patterns obtained by patterning a plurality of the detection values acquired by the acquisition unit, when the detection value of at least one of the plurality of detection elements exceeds the threshold value; and a determination unit which determines a correspondence process corresponding to the type of the smell decided on by the smell decision unit.

Description

[Technical field]

[0001] An embodiment of the present invention relates to an odor determination device, an odor determination method, and an odor determination system. [Background technology]

[0002] In order to determine odors, which are a collection of multiple gas components, odor sensors have been developed that array detection elements with adsorption films that adsorb odor-causing substances, and output odor patterns that are patterns of detection values ​​related to the odor-causing substances detected by the detection elements. Note that transducers that output the odor-causing substances adsorbed on the adsorption film as electrical signals include QCM (Quartz Crystal Microbalance), as well as piezoelectric resonator, semiconductor, and electrochemical transducers.

[0003] Then, the odor pattern is used to learn the odor pattern and determine the type of odor through machine learning. This makes it possible to determine the type of odor by associating the odor pattern with the type of odor without manual analysis of the odor pattern. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2002-298240 A [Patent Document 2] Japanese Patent Application Publication No. 05-223720 Summary of the Invention [Problem to be solved by the invention]

[0005] According to the above-mentioned technology, it is possible to obtain a result of determining the type of odor (e.g., orange odor, apple odor, flower odor), but it is also necessary to determine the process corresponding to the type of odor based on the result of the odor type determination. For example, if the odor type determination result is a bad odor, it is necessary to determine the process of spraying a deodorant, ventilating the room, stopping the device that emits the bad odor, sending an alarm to a mobile device, etc.

[0006] The present invention has been made in consideration of the above, and aims to provide an odor determination device, an odor determination method, and an odor determination system that can automatically execute processing corresponding to the type of odor based on the result of determining the type of odor. [Means for solving the problem]

[0007] The odor determination device according to the embodiment includes an acquisition unit that acquires detection values ​​obtained from each of a plurality of detection elements, a judgment unit that judges whether the detection values ​​exceed a preset threshold value, an odor determination unit that, when the detection value of at least one of the plurality of detection elements exceeds the threshold value, determines a type of odor based on an odor pattern obtained by patterning the plurality of detection values ​​acquired by the acquisition unit, and a decision unit that decides a response process corresponding to the type of odor determined by the odor determination unit, and the decision unit is configured to execute a plurality of the response processes simultaneously. And the smell gets worse case ,before At least one of the corresponding processes is discarded and the other is executed. Effect of the Invention

[0008] According to the present invention, for example, it is possible to automatically execute processing corresponding to the type of odor based on the result of determining the type of odor. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing an example of a schematic configuration of an odor determination system according to the first embodiment. [Diagram 2] FIG. 2 is a diagram showing an example of the configuration of the odor sensor included in the odor determination system according to the first embodiment. [Diagram 3] FIG. 3 is a diagram showing an example of a detection value output from the odor sensor of the odor determination system according to the first embodiment. [Figure 4] FIG. 4 is a diagram for explaining an example of a process for generating determination information in the cloud of the odor determination system according to the first embodiment. [Diagram 5] FIG. 5 is a diagram illustrating an example of a hardware configuration of an SCU included in the odor determination system according to the first embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of a hardware configuration of a cloud included in the odor determination system according to the first embodiment. [Figure 7A] FIG. 7A is a diagram illustrating an example of the functional configuration of a cloud included in the odor determination system according to the first embodiment. [Figure 7B] FIG. 7B is a diagram for explaining an example of an external device control module and an external device included in the odor determination system according to the first embodiment. [Figure 8] FIG. 8 is a flowchart showing an example of the flow of a response process decision process by the cloud in the odor determination system according to the first embodiment. [Figure 9] FIG. 9 is an explanatory diagram of an example in which the odor determination system according to the first embodiment is used in an automatic vacuum cleaner robot. [Figure 10] FIG. 10 is an explanatory diagram of an example in which the odor determination system according to the first embodiment is used as a deodorization system for a hotel room. [Figure 11] FIG. 11 is an explanatory diagram of an example in which the odor determination system according to the first embodiment is used in an exhaust system of a restaurant. [Figure 12] FIG. 12 is an explanatory diagram of an example in which the odor determination system according to the first embodiment is used to combat bad breath. [Figure 13]FIG. 13 is an explanatory diagram of an example in which the odor determination system according to the first embodiment is used for risk prediction in important facilities. [Figure 14] FIG. 14 is a flowchart showing another example of the flow of the response processing decision process by the cloud in the odor determination system according to the second embodiment. [Figure 15] FIG. 15 is a flowchart showing another example of the flow of the response processing decision process by the cloud in the odor determination system according to the third embodiment. [Figure 16] FIG. 16 is a diagram illustrating an example of the functional configuration of an SCU included in the odor determination system according to the fourth embodiment. As illustrated in FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an odor determination system to which the odor determination device, odor determination method, and odor determination system according to the present embodiment are applied will be described with reference to the accompanying drawings.

[0011] (First embodiment) FIG. 1 is a diagram showing an example of a schematic configuration of an odor determination system according to the first embodiment.

[0012] First, an example of a schematic configuration of an odor determination system 10 according to the present embodiment will be described with reference to FIG.

[0013] The odor determination system 10 according to this embodiment determines the type of odor. Here, the type of odor is, for example, the type of odor present in various environments, such as odor present in the air, odor present in gas, odor present inside a housing, etc. Next, the odor determination system 10 determines a process corresponding to the type of odor (hereinafter, referred to as a corresponding process) based on the determination result.

[0014] As shown in FIG. 1, the odor determination system 10 includes a plurality of odor sensors 11, an SCU 12, a cloud 13, an external device control module 14, a mobile terminal 15, and an external device X.

[0015] The odor sensor 11 and the SCU 12 are communicatively connected to each other via a communication means L1 such as NFC (Near Field Communication), Bluetooth (registered trademark), etc. The SCU 12, the cloud 13, the external device control module 14, and the mobile terminal 15 are communicatively connected to each other via a communication means L2 such as a LAN (Local Area Network), a dedicated line such as a wireless LAN, or a public line such as the Internet.

[0016] The odor sensor 11 is a sensor that detects the mass, etc., of a substance that causes an odor.

[0017] The SCU (Sensor Control Unit) 12 is a single board computer, a personal computer, or the like, and controls the detection of the mass of odor-causing substances by the odor sensor 11. The SCU 12 also has a communication function between the odor sensor 11 and the cloud 13 (e.g., wireless communication using Wi-Fi (registered trademark), Bluetooth (registered trademark), Zigbee (registered trademark), or the like) and a communication function between the cloud 13 and a mobile terminal 15 (e.g., a 4G line or a LAN (Local Area Network)).

[0018] The cloud 13 acquires the detection results of the odor sensor 11, such as the mass of the odor-causing substance, via the SCU 12, determines the type of odor based on the detection results, and determines the response process corresponding to the determination results.

[0019] The external device control module 14 is a single board computer, an I / O expansion board, etc., and acquires the odor type determination results and the response processing decision results from the cloud 13, and transmits a control signal to the external device X instructing it to execute the acquired response processing.

[0020] The mobile terminal 15 executes operations such as logging in the user to the odor determination system 10 and acquiring the odor type determination results and response processing decision results from the cloud 13.

[0021] FIG. 2 is a diagram showing an example of the configuration of the odor sensor included in the odor determination system according to the first embodiment.

[0022] Next, an example of the configuration of the odor sensor 11 according to this embodiment will be described with reference to FIG.

[0023] The odor sensor 11 has a detection element such as a QCM (Quartz Crystal Microbalance) sensor, and is capable of detecting values ​​related to the odor-causing substances, such as the mass of the odor-causing substances (hereinafter referred to as detection values). The detection element of the odor sensor 11 is not limited to a QCM sensor, and may be a sensor of another type, such as a gas sensor using a semiconductor thin film, as long as it is a detection element capable of detecting detection values ​​related to the odor-causing substances.

[0024] 2, the odor sensor 11 includes a support portion 110, a plurality of detection elements F1 to F16, and a drive detection circuit 111. For example, 16 detection elements F are attached to the support portion 110. In the following description, when there is no need to distinguish between the detection elements F1 to F16, they will be referred to as detection elements F.

[0025] Each of the detection elements F1 to F16 detects the mass of a different type of gas causative substance. Specifically, the detection element F has a quartz crystal oscillator, two electrodes, and an adsorption film. In addition, a part of the side surface of the quartz crystal oscillator is held by a support 110 so as to be vibrable.

[0026] The two electrodes are provided on the flat surfaces on both sides of the quartz crystal unit, and an AC voltage is applied to the two electrodes from the drive detection circuit 111.

[0027] The adsorption film is provided on at least one of the flat surfaces of the quartz crystal oscillator. The adsorption film adsorbs a specific causative substance. Each of the detection elements F1 to F16 has an adsorption film that adsorbs causative substances having different chemical properties.

[0028] When an AC voltage of the resonant frequency is applied to the two electrodes of the detection element F, the quartz crystal oscillator vibrates due to the piezoelectric effect. The resonant frequency of the quartz crystal oscillator is determined by the mass and viscoelasticity of the causative substance attached to the adsorption film. Therefore, when a causative substance is adsorbed to the adsorption film and its mass changes, the resonant frequency of the detection element F changes in accordance with the change in the mass of the adsorbed causative substance.

[0029] The drive detection circuit 111 applies an AC voltage to at least one detection element F and detects the change per unit time in the resonant frequency of the detection element F. This allows the drive detection circuit 111 to detect the mass of the odor-causing substance for each detection element F.

[0030] The odor sensor 11 transmits the change in resonant frequency detected by the drive detection circuit 111 as a detection value to the SCU 12 via the communication means L1. In this embodiment, the odor sensor 11 transmits the change in resonant frequency detected by the drive detection circuit 111 as a detection value to the SCU 12, but this is not limited to the above as long as the detection value transmitted is a value related to the substance that causes the odor. For example, the odor sensor 11 may transmit a change in the resistance value of the detection element F as a detection value to the SCU 12.

[0031] FIG. 3 is a diagram showing an example of a detection value output from the odor sensor of the odor determination system according to the first embodiment.

[0032] Next, an example of a detection value output from the odor sensor 11 of the odor determination system 10 according to this embodiment will be described with reference to FIG.

[0033] 3, the odor sensor 11 may output data (hereinafter referred to as odor pattern) obtained by patterning the detection values ​​output from each of the detection elements F1 to F16. The odor pattern output by the odor sensor 11 becomes a wide variety of odor patterns according to the detection values ​​output from each of the detection elements F1 to F16.

[0034] In this embodiment, an example is shown in which the odor sensor 11 has 16 types of detection elements F, but this is not limited thereto, and the odor sensor 11 may have fewer or more than 16 types of detection elements F. When multiple types of odors are mixed, the odor sensor 11 can improve the accuracy of determining the type of odor by having more detection elements F.

[0035] The cloud 13 receives the detection value output from the odor sensor 11 via the SCU 12. The cloud 13 stores, in association with the detection value output from the odor sensor 11, determination information indicating the type of odor indicated by the detection value. The cloud 13 determines the type of odor corresponding to the detection value received from the odor sensor 11 based on the determination information stored in association with the received detection value. The cloud 13 also determines the strength of the odor based on the rate of change per unit time of the detection value output from the detection element F. The greater the rate of change in the detection value per unit time, the stronger the odor.

[0036] In an initial state in which no judgment information is stored, the cloud 13 generates judgment information indicating a type of odor indicated by a detection value based on the detection value received from the odor sensor 11. The cloud 13 performs multivariate analysis using, for example, a neural network, and generates judgment information indicating a type of odor indicated by the detection value.

[0037] FIG. 4 is a diagram for explaining an example of a process for generating determination information in the cloud of the odor determination system according to the first embodiment.

[0038] Next, an example of a process for generating determination information in the cloud 13 of the odor determination system 10 according to this embodiment will be described with reference to FIG.

[0039] The cloud 13 uses a neural network having an input layer N1, an intermediate layer N2, and an output layer N3 to generate judgment information indicating one type of odor corresponding to a received detection value. The input layer N1 receives the detection value from the odor sensor 11. The intermediate layer N2 extracts features of the odor type from the detection value received from the input layer N1. Here, a feature is an amount that can distinguish a difference from other odor types. The output layer N3 outputs the type of odor indicated by the features extracted in the intermediate layer N2 as judgment information.

[0040] After generating the determination information, the cloud 13 updates the determination information to determination information whose accuracy has been improved by a learning process. The cloud 13 may execute the learning process multiple times to improve the accuracy of the determination information.

[0041] For example, the odor-causing substances detected by the detection element F of the odor sensor 11 may contain a mixture of multiple odor-causing substances. In this case, the type of odor determined by the cloud 13 may not necessarily match the odor that is actually sensed by the human sense of smell. This is often due to a problem with the feature amount of the odor type extracted in the intermediate layer N2. Therefore, the cloud 13 extracts a feature amount corresponding to the type of odor sensed by the human sense of smell by modifying the feature amount output from the intermediate layer N2.

[0042] In this embodiment, the cloud 13 uses a neural network to determine the type of odor corresponding to the detection value received from the odor sensor 11, but this is not limited to this, and for example, the cloud 13 may use pattern matching or SVN to determine the type of odor corresponding to the detection value received from the odor sensor 11.

[0043] FIG. 5 is a diagram illustrating an example of a hardware configuration of an SCU included in the odor determination system according to the first embodiment.

[0044] Next, an example of the hardware configuration of the SCU 12 included in the odor determination system 10 according to this embodiment will be described with reference to FIG.

[0045] As shown in FIG. 5, the SCU 12 includes a central processing unit (CPU) 121, a read only memory (ROM) 122, a random access memory (RAM) 123, and a memory unit 124.

[0046] The CPU 121 controls the entire SCU 12. The ROM 122 and the memory unit 124 store various programs. The RAM 123 temporarily stores various data. The CPU 121, the ROM 122, the RAM 123, and the memory unit 124 are connected to each other via a data bus 125.

[0047] The CPU 121 loads various programs stored in the ROM 122 and the memory unit 124 into the RAM 123 and executes the loaded programs, thereby controlling the entire SCU 12 .

[0048] 5, the CPU 121 is connected to an operation unit 127 and a display unit 128 via a data bus 125 and a controller 126. The operation unit 127 is a keyboard operated by an operator to input data. The display unit 128 is configured, for example, with a liquid crystal display, and displays information to the operator of the SCU 12.

[0049] 5, the CPU 121 is connected to the communication I / F 129 via the data bus 125. The communication I / F 129 is connected to communication means L1 and communication means L2. The communication I / F 129 can communicate with the odor sensor 11 via the communication means L1, and receives various information such as detection values ​​from the odor sensor 11. The communication I / F 129 can communicate with the cloud 13 via the communication means L2, and receives various information from the cloud 13 such as teacher data and corrected determination information. Here, the communication I / F 129 also transmits various information such as the detection value of the detection element F to the cloud 13 via the communication means L2. The communication I / F 129 can also communicate with the mobile terminal 15.

[0050] FIG. 6 is a diagram illustrating an example of a hardware configuration of a cloud included in the odor determination system according to the first embodiment.

[0051] Next, an example of the hardware configuration of the cloud 13 included in the odor determination system 10 according to this embodiment will be described with reference to FIG.

[0052] As shown in FIG. 6, the cloud 13 includes a CPU 131, a ROM 132, a RAM 133, and a memory unit 134.

[0053] The CPU 131 controls the entire cloud 13. The ROM 132 and the memory unit 134 store various programs. The RAM 133 temporarily stores various data. The CPU 131, the ROM 132, the RAM 133, and the memory unit 134 are connected to each other via a data bus 135.

[0054] The CPU 131 loads various programs stored in the ROM 132 and the memory unit 134 into the RAM 133 and executes the various loaded programs, thereby controlling the entire cloud 13 .

[0055] 6, the CPU 131 is connected to an operation unit 137 and a display unit 138 via a data bus 135 and a controller 136. The operation unit 137 is a keyboard operated by an operator to input data. The display unit 138 is configured, for example, with a liquid crystal display, and displays information to the operator of the cloud 13.

[0056] 6, the CPU 131 is connected to a communication I / F 139 via a data bus 135. The communication I / F 139 is connected to a communication means L2. The communication I / F 139 can communicate with the SCU 12 via the communication means L2, and receives information (e.g., a detection value of a detection element F) from the SCU 12. The communication I / F 139 also transmits information to the SCU 12 via the communication means L2.

[0057] FIG. 7A is a diagram illustrating an example of the functional configuration of a cloud included in the odor determination system according to the first embodiment.

[0058] Next, an example of the functional configuration of the cloud 13 included in the odor determination system 10 according to this embodiment will be described with reference to FIG. 7A.

[0059] In this embodiment, the CPU 131 loads various programs stored in the ROM 132 or the memory unit 134 into the RAM 133 and executes the loaded various programs to function as an acquisition unit 701, a judgment unit 702, an odor determination unit 703, and a decision unit 704. In this embodiment, the CPU 131 is an example of an odor determination device.

[0060] In this embodiment, the acquisition unit 701, judgment unit 702, odor judgment unit 703, and decision unit 704 are realized in the cloud 13, but it is also possible to realize some or all of the acquisition unit 701, judgment unit 702, odor judgment unit 703, and decision unit 704 in a device external to the cloud 13 (e.g., SCU 12).

[0061] The acquiring unit 701 acquires, from the plurality of odor sensors 11, detection values ​​obtained from each of the plurality of detection elements F. In this embodiment, the acquiring unit 701 acquires, from the plurality of odor sensors 11, odor patterns of the detection values ​​of the plurality of detection elements F that each of the plurality of odor sensors 11 has, via the SCU 12.

[0062] The judgment unit 702 judges whether the detection value acquired by the acquisition unit 701 exceeds a predetermined threshold. Here, the predetermined threshold is a threshold that is set in advance. Specifically, the predetermined threshold is a detection value threshold at which it is determined that a substance causing an odor is present.

[0063] When the judgment unit 702 judges that at least one of the multiple detection values ​​exceeds a predetermined threshold, the odor judgment unit 703 judges the type of odor based on an odor pattern obtained by patterning the multiple detection values. In this embodiment, the odor judgment unit 703 judges the type of odor based on the odor pattern, but this is not limited to this as long as the odor type is judged based on the detection values ​​of multiple detection elements F. For example, the odor judgment unit 703 extracts a feature amount of the type of odor from the multiple detection values ​​acquired by the acquisition unit 701, and judges the type of odor based on the feature amount.

[0064] The determination unit 704 determines a response process corresponding to the type of odor determined by the odor determination unit 703. In this way, when it is determined that an odor is present, the response process corresponding to the type of odor can be automatically executed based on the result of the odor type determination.

[0065] For example, the determination unit 704 determines the type of odor to be dealt with by the odor determination unit 703, such as spraying a deodorant, ventilation, generating ozone, or operating an air cleaner.

[0066] Furthermore, the determination unit 704 makes the odor type determination result and the response process determination result public to the external device control module 14 and the mobile terminal 15 via a Web API (Application Programming Interface) or the like.

[0067] This allows the external device control module 14 and the mobile terminal 15 to acquire the odor type determination result and the response process decision result. For example, the external device control module 14 acquires the response process decision result published by the decision unit 704 at a preset cycle.

[0068] Then, the external device control module 14 functions as an example of a control unit that outputs control information instructing the external device X to execute the acquired corresponding process. For example, the external device control module 14 executes a control program to turn on a relay that can instruct the external device X to execute the determined corresponding process, among relays of multiple channels (e.g., four channels) that the relay output board has.

[0069] FIG. 7B is a diagram for explaining an example of an external device control module and an external device included in the odor determination system according to the first embodiment.

[0070] Next, an example of the external device control module 14 and the external device X included in the odor determination system 10 according to this embodiment will be described with reference to FIG. 7B.

[0071] In this embodiment, the external device control module 14 has a single board computer, a relay output board, etc. Then, the external device control module 14 acquires the result of the response process determined by the determination unit 704 of the cloud 13 via the SCU 12, as shown in Fig. 7B. Next, based on the acquired result of the response process, the external device control module 14 turns on a relay that can instruct the external device X to execute the determined response process, among the relays of multiple channels that the relay output board has. As shown in Fig. 7B, the external device X executes a response process (e.g., spraying a deodorant) corresponding to the relay turned on by the external device control module 14.

[0072] FIG. 8 is a flowchart showing an example of the flow of a response process decision process by the cloud in the odor determination system according to the first embodiment.

[0073] Next, an example of the flow of a response process decision process by the cloud 13 will be described with reference to FIG.

[0074] First, the acquiring unit 701 acquires the detection values ​​of the multiple detection elements F from the multiple odor sensors 11 via the SCU 12 (step S801). In this embodiment, the acquiring unit 701 acquires the detection values ​​of the multiple detection elements F from the multiple odor sensors 11 at a preset cycle.

[0075] Next, the determination unit 702 determines whether or not the detection value of the detection element F exceeds a predetermined threshold value each time the acquisition unit 701 acquires a detection value (step S802).

[0076] If none of the detection values ​​of the multiple detection elements F exceeds the predetermined threshold (step S802: No), the determination unit 702 determines that no odor is present, and ends the response process decision process.

[0077] On the other hand, if the detection value of at least one of the multiple detection elements F exceeds a predetermined threshold (step S802: Yes), the odor determination unit 703 determines the type of odor based on the detection values ​​of the multiple detection elements F (step S803).

[0078] Next, the determination unit 704 determines a response process corresponding to the determined type of odor based on the result of the odor type determination by the odor determination unit 703 (step S804).

[0079] Next, with reference to Figs. 9 to 13, examples of various situations to which the odor determination system 10 according to this embodiment can be applied will be described. Fig. 9 is an explanatory diagram of an example in which the odor determination system according to the first embodiment is used in an automatic vacuum cleaner robot. Fig. 10 is an explanatory diagram of an example in which the odor determination system according to the first embodiment is used in a deodorizing system for a hotel room. Fig. 11 is an explanatory diagram of an example in which the odor determination system according to the first embodiment is used in an exhaust system for a restaurant. Fig. 12 is an explanatory diagram of an example in which the odor determination system according to the first embodiment is used to combat bad breath. Fig. 13 is an explanatory diagram of an example in which the odor determination system according to the first embodiment is used to predict danger in an important facility.

[0080] First, an example in which the odor determination system 10 according to the present embodiment is applied to deodorization and cleaning will be described.

[0081] For example, when the odor determination system 10 according to the present embodiment is used for deodorizing odors emitted from a factory, the odor sensor 11 is installed in the exhaust duct of the factory. Then, when the detection value of the detection element F of the odor sensor 11 exceeds a predetermined threshold, the odor determination unit 703 of the cloud 13 determines the type of odor based on the multiple detection values. Next, the determination unit 704 of the cloud 13 determines the corresponding process to be a process of spraying a deodorant corresponding to the determined type of odor.

[0082] For example, when the odor determination system 10 according to the present embodiment is used to control an automatic vacuum cleaner robot 900, as shown in FIG. 9, the automatic vacuum cleaner robot 900 is equipped with an odor sensor 11. Then, when the detection value of the detection element F of the odor sensor 11 exceeds a predetermined threshold while the automatic vacuum cleaner robot 900 is cleaning, the odor determination unit 703 of the cloud 13 determines the type of odor based on the multiple detection values. Next, the determination unit 704 of the cloud 13 determines the corresponding process to spray a deodorant corresponding to the determined type of odor, and causes the automatic vacuum cleaner robot 900 to execute the corresponding process. When the detection element F of the odor sensor 11 mounted on the automatic vacuum cleaner robot 900 can detect the detection value of the odor-causing substance throughout the room, the determination unit 704 of the cloud 13 can also determine the corresponding process to be ventilation that creates an air flow by blowing air.

[0083] For example, when the odor determination system 10 according to the present embodiment is used as a deodorization system for a hotel room, an odor sensor 11 is installed in the hotel guest room as shown in FIG. 10. Then, after a guest checks out of the hotel, if the detection value of the detection element F of the odor sensor 11 exceeds a predetermined threshold, the odor determination unit 703 of the cloud 13 determines the type of odor based on the multiple detection values. Next, the determination unit 704 of the cloud 13 determines the corresponding process to be spraying a deodorant corresponding to the determined type of odor or generating ozone. This eliminates the need for hotel staff to check the odor in the hotel guest room and execute a process corresponding to the detected odor, thereby reducing the number of hotel staff.

[0084] Next, an example in which the odor determination system 10 according to this embodiment is applied to food-related products will be described.

[0085] 11, for example, building 1100 containing multiple restaurants 1101-1109 often exhausts exhaust air from each of the restaurants 1101-1109 to the outside all at once. Therefore, if the filter in the exhaust equipment of one restaurant (e.g., restaurant 1101) deteriorates or breaks down and the restaurant's exhaust air is exhausted to the outside, it is difficult to identify which restaurant's exhaust equipment has deteriorated or broken down, and it takes time to improve the situation.

[0086] Therefore, when the odor determination system 10 according to the present embodiment is used to deal with odors exhausted from a building 1100 containing multiple restaurants 1101-1109, as shown in FIG. 11, an odor sensor 11 is installed in a duct 1110 that exhausts the odor to the outside of the building 1100. Then, when the detection value of the detection element F of the odor sensor 11 exceeds a predetermined threshold, the odor determination unit 703 of the cloud 13 determines the type of odor based on the multiple detection values. Next, the determination unit 704 of the cloud 13 determines the type of odor based on the determined type of odor, and determines the process of identifying the restaurant (e.g., restaurant 1101) that is emitting the odor and notifying the restaurant 1101 of a request to improve the odor as the response process. This allows the restaurant 1101 to immediately repair the exhaust equipment, thereby shortening the time required to improve the odor exhausted from the building 1100. At this time, the determination unit 704 of the cloud 13 may obtain information on the smart meters of the restaurants 1101-1109 and identify the restaurant that is emitting the odor from among the restaurants that are in operation.

[0087] For example, when the odor determination system 10 according to the present embodiment is used for process management in a food factory that manufactures food, an odor sensor 11 is installed on the production line of the food factory. Then, when the detection value of the detection element F of the odor sensor 11 exceeds a predetermined threshold, an odor determination unit 703 of the cloud 13 determines the type of odor based on the multiple detection values. Next, a determination unit 704 of the cloud 13 determines a response process corresponding to the determined type of odor.

[0088] Specifically, when the determination unit 704 of the cloud 13 determines that the type of odor is a burnt odor generated during cooking of food, it determines that the response process is to stop the device in the food factory that is generating the burnt odor. Also, when the determination unit 704 of the cloud 13 determines that the type of odor is an odor generated by spoiled food, it determines that the response process is to issue an alert notifying that an odor is generated by spoiled food. Also, when the determination unit 704 of the cloud 13 determines that the type of odor is an animal odor, it determines that the response process is to issue an alert notifying that an animal has appeared.

[0089] Next, an example in which the odor determination system 10 according to the present embodiment is applied to etiquette will be described.

[0090] For example, when the odor determination system 10 according to the present embodiment is used for dealing with bad breath, an odor sensor 11 is installed in a hospitality office, hotel, or the like, as shown in FIG. 12. Then, when the detection value of the detection element F of the odor sensor 11 when breath is blown onto the odor sensor 11 exceeds a predetermined threshold, the odor determination unit 703 of the cloud 13 determines the type of odor based on the multiple detection values. At this time, the odor determination unit 703 may input information such as what the subject ate, whether or not they consumed alcohol, the subject's body odor, and a photo of the subject's mouth, and may determine the type of odor by adding the input information. Next, the determination unit 704 of the cloud 13 determines the response process corresponding to the determined type of odor.

[0091] Specifically, when the odor of dumplings is determined as the type of odor, the determination unit 704 of the cloud 13 determines, as the response process, a notification process that encourages the use of a garlic mouthwash. When the odor of periodontal disease is determined as the type of odor, the determination unit 704 of the cloud 13 determines, as the response process, a notification process that encourages the use of a mouthwash 1203 that treats periodontal disease. When the odor of indigestion is determined as the type of odor, the determination unit 704 of the cloud 13 determines, as the response process, a notification process that encourages the use of a stomach medicine or an intestinal regulator. When the odor of indigestion is determined as the type of odor, the determination unit 704 of the cloud 13 determines, as the response process, a process that displays on the display device 1202 the degree of annoyance that the determined type of odor causes to the other party, as shown in FIG. 12. In addition, if the type of odor determined is one that cannot be deodorized by mouthwash 1203 or prescribed medicine, the decision unit 704 of cloud 13 may decide to perform the corresponding process of displaying on the display device 1202 the distance at which the bad breath will not affect the other person, or the time at which the effects of mouthwash 1203 or prescribed medicine will wear off.

[0092] For example, when the odor determination system 10 according to the present embodiment is also used to deal with body odor, the odor sensor 11 is installed in an enclosed space (e.g., a room) installed in a hospitality office, a hotel, a running station, etc. Then, if the detection value of the detection element F of the odor sensor 11 when a subject enters the enclosed space exceeds a predetermined threshold, the odor determination unit 703 of the cloud 13 determines the type of odor based on the multiple detection values. Next, the determination unit 704 of the cloud 13 determines the response process corresponding to the determined type of odor.

[0093] Specifically, when the determination unit 704 of the cloud 13 determines that the type of odor is sweat odor, it determines that the corresponding process is to spray a deodorant spray. When the determination unit 704 of the cloud 13 determines that the type of odor is odor caused by damp clothes, it determines that the corresponding process is to spray an antibacterial spray. When the determination unit 704 of the cloud 13 determines that the type of odor is armpit odor, it determines that the corresponding process is to spray a deodorant spray mainly on the armpits. When the determination unit 704 of the cloud 13 determines that the type of odor is foot odor, it determines that the corresponding process is to spray a deodorant spray mainly on the feet. When the determination unit 704 of the cloud 13 determines that the type of odor is scalp odor, it determines that the corresponding process is to spray a hair fragrance.

[0094] Next, an example in which the odor determination system 10 according to this embodiment is applied to risk prediction in important facilities (for example, data centers, shrines and temples, power plants, and substations) will be described.

[0095] For example, when the odor determination system 10 according to the present embodiment is used for risk prediction in a data center 1301, a power plant, or a substation 1302, an odor sensor 11 is installed in an information processing device 1301a installed in the data center 1301, and in a distribution board 1302a, a distribution board 1302b, and a control board 1302c of the power plant or substation 1302, as shown in Fig. 13. Then, when the detection value of the detection element F of the odor sensor 11 exceeds a predetermined threshold, an odor determination unit 703 of the cloud 13 determines the type of odor based on the multiple detection values. Next, when the odor of burning wiring is determined as the type of odor, a determination unit 704 of the cloud 13 determines the response process to be an alert notifying that a fire may occur.

[0096] Conventionally, when detecting smoke using a smoke detector installed in a data center 1301, a person needs to enter the data center 1301 in order to identify the location within the data center 1301 from which the smoke is coming. On the other hand, according to the odor determination system 10 of this embodiment, by identifying the location of the detection element F that detected the detection value of the determined type of odor, the location within the data center 1301 from which the smoke is coming can be easily identified.

[0097] Also, since smoke is unlikely to come out of the wiring in the power plant or substation 1302, some smoke detectors may not be able to detect smoke from the wiring until the damage becomes significant. On the other hand, according to the odor determination system 10 of this embodiment, if an odor sensor 11 is installed in the power plant or substation 1302, it becomes possible to execute a response process to the burnt wiring before the damage becomes significant. Also, by detecting the generation of smoke using a dust sensor that measures particles in the air with light in addition to the detection value of the detection element F of the odor sensor 11, it is possible to improve the accuracy of the response process to the burnt wiring and shorten the time until the response process is executed.

[0098] In this way, according to the odor determination system 10 of the first embodiment, it is possible to automatically execute a response process corresponding to the type of odor based on the result of the odor type determination.

[0099] Second Embodiment This embodiment is an example in which a response process is determined based on the similarity between an odor pattern obtained by patterning multiple detection values ​​and a reference pattern, which is a standard odor pattern set for each candidate odor type. In the following description, the same configuration as the first embodiment will not be described.

[0100] In this embodiment, the determination unit 704 of the cloud 13 determines the similarity between an odor pattern obtained by patterning a plurality of detection values ​​and a reference pattern. Here, the reference pattern is a pattern that is set in advance for each candidate odor type to be determined and serves as a reference for the odor pattern.

[0101] Then, the determination unit 704 determines the response process when the calculated similarity is equal to or greater than a predetermined similarity. Here, the predetermined similarity is a similarity that is set in advance, and is the lower limit of the similarity for executing the response process (e.g., 50%). This makes it possible to prevent an incorrect response process from being determined due to fluctuations such as measurement errors in the detection value by the detection element F. As a result, it is possible to prevent an incorrect response process from being executed, which would cause the odor to worsen.

[0102] On the other hand, when the obtained similarity is less than the predetermined similarity, the determination unit 704 does not determine the response process. For example, when the response process differs depending on the cause of the fire that is generating the odor, if an incorrect response process is executed, the fire may worsen. Therefore, as described above, the determination unit 704 determines the response process when the obtained similarity is equal to or greater than the predetermined similarity.

[0103] For example, if the similarity between an odor pattern and reference pattern A is 34%, the similarity between the odor pattern and reference pattern B is 33%, and the similarity between the odor pattern and reference pattern C is 33%, then the type of odor is likely to be the type of odor of reference pattern A, but is also close to the types of odor of reference patterns B and C.

[0104] Therefore, in this case, the determination unit 704 does not determine a response process because the similarity between the odor pattern and each of the reference patterns A to C is less than a predetermined similarity. Then, the above-mentioned acquisition unit 701 reacquires the detection values ​​of the multiple detection elements F. Furthermore, if the similarity is less than a predetermined similarity, the odor determination unit 703 re-determines the type of odor based on the multiple newly acquired detection values. As a result, if the similarity between the odor pattern and the reference patterns A to C is low and the reliability of the odor type determination result is low, the odor type is re-determined, and response processes can be determined based on the highly reliable odor type determination result. As a result, response processes that are more suitable for the odor type determination result can be executed.

[0105] The odor determination unit 703 can also determine, as the type of odor, a candidate for the type of odor corresponding to a reference pattern having the highest similarity for a preset number of consecutive times among the multiple reference patterns A to C. That is, the candidate for the type of odor corresponding to a reference pattern having the highest similarity for a preset number of consecutive times among the multiple reference patterns A to C is the odor type determination result by the odor determination unit 703. For example, if the similarity between the odor pattern and reference pattern A is the highest for a preset number of consecutive times among the similarities between the odor pattern and each of the reference patterns A to C, the odor determination unit 703 determines the candidate for the type of odor corresponding to reference pattern A as the type of odor. This makes it possible to determine a response process based on a more reliable odor type determination result based on the similarity between the odor pattern and each of the reference patterns A to C. As a result, a response process that is more suitable for the odor type determination result can be executed. In this embodiment, when the similarity between the odor pattern and each of the reference patterns A to C is less than a predetermined similarity, the odor determination unit 703 determines that the candidate odor type corresponding to the reference pattern among the multiple reference patterns A to C that has the highest similarity a predetermined number of times in succession is the odor type.

[0106] In this embodiment, the determination unit 704 can also change the response process depending on the obtained similarity. For example, if the response process is spraying a deodorant, the determination unit 704 changes the response process to one that sprays a larger amount of deodorant as the obtained similarity increases. If the response process is generating ozone, the determination unit 704 changes the response process to one that generates a larger amount of ozone as the obtained similarity increases. If the response process is operating an air purifier, the determination unit 704 changes the response process to one that extends the operating time of the air purifier as the obtained similarity increases.

[0107] Furthermore, in this embodiment, when the odor determination unit 703 determines multiple odor types and determines a corresponding response process for each of the multiple odor types, the determination unit 704 can change the determined response process based on the relationship between the multiple determined response processes. This makes it possible to prevent multiple response processes that would worsen the odor, etc., from being executed simultaneously when the odor determination unit 703 determines multiple odor types and determines a corresponding response process for each of the multiple odor types.

[0108] For example, if the odor, etc. worsens when multiple determined response processes are executed simultaneously, the determination unit 704 executes one of the multiple determined response processes. In this case, the determination unit 704 selects, from the multiple determined response processes, the response process that corresponds to the type of odor of the reference pattern that has the highest similarity to the odor pattern. Alternatively, if the odor, etc. worsens when multiple determined response processes are executed simultaneously, the determination unit 704 can discard the response process determination result.

[0109] Furthermore, in this embodiment, the determining unit 704 can also change the determined response process based on at least one of the temperature and humidity at the installation location of the detection element F. For example, the determining unit 704 acquires the temperature and humidity at the installation location of the detection element F from a sensor installed near the detection element F. Then, when the acquired temperature and humidity fall within a predetermined temperature or humidity range, the determining unit 704 changes the response process (for example, discards the response process). Here, the predetermined temperature or humidity range is a temperature or humidity range that is set in advance.

[0110] FIG. 14 is a flowchart showing another example of the flow of the response processing decision process by the cloud in the odor determination system according to the second embodiment.

[0111] Next, another example of the flow of the response process decision process by the cloud 13 will be described with reference to Fig. 14. Steps S801 to S803 shown in Fig. 14 are the same as those in Fig. 8, and therefore description thereof will be omitted.

[0112] The determining unit 704 determines the similarity between the odor pattern and each of a plurality of reference patterns (e.g., three reference patterns A to C) (step S1401). Here, the determining unit 704 determines the similarity between the odor pattern and each of the three reference patterns A to C, but the determination unit 704 is not limited to this as long as it determines the similarity between a plurality of reference patterns and an odor pattern.

[0113] Next, the determination unit 704 determines whether the calculated similarity is equal to or greater than a predetermined similarity for each type of odor corresponding to each of the multiple reference patterns (step S1402).Then, for an odor type whose calculated similarity is determined to be equal to or greater than the predetermined similarity (step S1402: Yes), the determination unit 704 determines a corresponding process corresponding to the odor type (step S1403).On the other hand, for an odor type whose calculated similarity is determined to be less than the predetermined similarity (step S1402: No), the determination unit 704 does not determine a corresponding process corresponding to the odor type.

[0114] In this way, according to the odor determination system 10 of the second embodiment, it is possible to prevent erroneous response processing from being determined due to fluctuations such as measurement errors in the detection value by the detection element F, thereby preventing erroneous response processing from being executed and causing the odor to worsen.

[0115] (Third embodiment) This embodiment is an example in which a response process is determined based on the number of times the same odor type is determined. In the following description, a description of the same configuration as in the first embodiment will be omitted.

[0116] In this embodiment, the decision unit 704 of the cloud 13 decides on a response process when the odor determination unit 703 judges the same type of odor a predetermined number of times n (n is an integer equal to or greater than 1, for example, n=3) or more consecutive times. Here, the predetermined number of times is a number that is set in advance, and is the number of times at which the odor determination unit 703 judges that the odor type determination result is highly reliable.

[0117] This makes it possible to prevent a response process from being determined based on an incorrect odor type determination result. As a result, it is possible to increase the likelihood of executing an appropriate response process for the odor. Furthermore, since a response process is not executed for an odor that occurs temporarily, and a response process (e.g., spraying a deodorant) is executed when an odor continues to occur for a certain period of time, it is possible to reduce the load on the external device X (e.g., a deodorant sprayer) caused by executing a response process for an odor that occurs temporarily.

[0118] In this embodiment, the determination unit 704 determines the response process based on the number of times the same odor type has been determined, but this is not limited to the above. For example, the determination unit 704 may determine the response process based on the time that the same odor type has been continuously determined.

[0119] Specifically, when the same type of odor continues to be determined for a predetermined consecutive time, the determination unit 704 determines a response process corresponding to the type of odor. On the other hand, when the time during which the same type of odor continues to be determined is less than the predetermined time, the determination unit 704 does not determine a response process. As a result, the response process is not performed for odors that occur temporarily, but is performed when an odor has been occurring continuously for a certain period of time, thereby reducing the load on the external device X caused by the response process being performed for odors that occur temporarily.

[0120] FIG. 15 is a flowchart showing another example of the flow of the response processing decision process by the cloud in the odor determination system according to the third embodiment.

[0121] Next, another example of the flow of the response process decision process by the cloud 13 will be described with reference to Fig. 15. Steps S801 to S803 shown in Fig. 15 are the same as those in Fig. 8, and therefore description thereof will be omitted.

[0122] The determination unit 704 determines, for each type of odor determined by the odor determination unit 703, whether or not the same type of odor has been determined a predetermined number of times n (eg, 3 times) or more in succession (step S1501).

[0123] If it is determined that the odor determination unit 703 has determined the same type of odor a predetermined number of times or more in succession (step S1501: Yes), the determination unit 704 determines a response process corresponding to the type of odor determined by the odor determination unit 703 (step S1502).

[0124] On the other hand, if it is determined that the odor determination unit 703 has not determined the same type of odor for a predetermined number of consecutive times or more (step S1501: No), the determination unit 704 does not determine a response process corresponding to the type of odor determined by the odor determination unit 703.

[0125] In this way, the odor determination system 10 according to the third embodiment can prevent a response process from being determined based on an incorrect odor type determination result, thereby increasing the likelihood of executing an appropriate response process for the odor. Also, since a response process is not executed for a temporary odor, but is executed when the odor has been generated continuously for a certain period of time, the load on the external device X caused by executing a response process for a temporary odor can be reduced.

[0126] (Fourth embodiment) This embodiment is an example in which a response process corresponding to the type of odor is determined in an SCU. In the following description, the same parts as in the first to third embodiments will not be described.

[0127] FIG. 16 is a diagram illustrating an example of the functional configuration of an SCU included in the odor determination system according to the fourth embodiment. As illustrated in FIG.

[0128] In this embodiment, the CPU 121 loads various programs stored in the ROM 122 or the memory unit 124 into the RAM 123 and executes the loaded various programs to function as an acquisition unit 701, a judgment unit 702, an odor determination unit 703, and a decision unit 704. In this embodiment, the CPU 121 is an example of an odor determination device.

[0129] This makes it possible for SCU12 to determine the type of odor and execute response processing corresponding to the determined type of odor even in situations where it is not connected to a network and cannot communicate with cloud 13 (for example, when SCU12 is installed in a car or boat).

[0130] In this embodiment, the odor determination unit 703 downloads in advance from the cloud 13 determination information generated by a learning process in the cloud 13. Then, the odor determination unit 703 uses the downloaded determination information to determine the type of odor based on the detection values ​​of the multiple detection elements F.

[0131] In this embodiment as well, the determination unit 704 discloses the result of the odor type determination and the result of the response process determination to the external device control module 14 and the mobile terminal 15 via a Web API or the like.

[0132] In this embodiment, the SCU 12 and the external device control module 14 are realized by separate devices, but when the SCU 12 is connected to an I / O expansion board or a relay output board that is connected to the external device X, it is possible to realize the functions of the external device control module 14 in the SCU 12.

[0133] However, if the SCU 12 is mounted in a dedicated housing, such as a commercially available gateway, and it is difficult to connect to an I / O expansion board or a relay output board, the SCU 12 and the external device control module 14 must be realized as separate devices.

[0134] In this way, according to the odor determination system 10 of the fourth embodiment, even in a situation where the SCU 12 is not connected to a network and cannot communicate with the cloud 13 (for example, when the SCU 12 is installed in a car or a boat), it is possible to determine the type of odor and execute response processing corresponding to the determined type of odor.

[0135] As described above, according to the first to fourth embodiments, it is possible to automatically execute a response process corresponding to the type of odor based on the result of the odor type determination.

[0136] The program executed by the odor determination system 10 of this embodiment is provided in a state where it is pre-installed in a ROM or the like. The program executed by the odor determination system 10 of this embodiment may be provided by being recorded in a computer-readable recording medium such as a CD-ROM, a flexible disk (FD), a CD-R, or a digital versatile disk (DVD) in the form of an installable or executable file.

[0137] Furthermore, the program executed by the odor determination system 10 of this embodiment may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Also, the program executed by the odor determination system 10 of this embodiment may be provided or distributed via a network such as the Internet.

[0138] The program executed by the odor determination system 10 of this embodiment has a modular structure including the above-mentioned units (acquisition unit 701, judgment unit 702, odor determination unit 703, and determination unit 704), and in terms of actual hardware, the CPU reads the program from the ROM and executes it, loading the above-mentioned units onto the main memory, and the acquisition unit 701, judgment unit 702, odor determination unit 703, and determination unit 704 are generated on the main memory.

[0139] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0140] 10. Odor Detection System 11 Odor Sensor 12 SCU 13. Cloud 14 External device control module 15 Mobile Devices 121,131 CPU 122,132 ROM 123,133 RAM 124,134 Memory section 125,135 Data Bus 126,136 Controllers 127,137 Operation section 128,138 Display section 129,139 Communication I / F 701 Acquisition Department 702 Judgment Department 703 Odor Judgment Department 704 Decision Section F Detector element X External device

Claims

1. an acquisition unit that acquires detection values ​​obtained from each of the plurality of detection elements; A determination unit that determines whether the detection value exceeds a preset threshold value; an odor determination unit that, when the detection value of at least one of the plurality of detection elements exceeds the threshold value, determines the type of odor based on an odor pattern obtained by patterning the plurality of detection values ​​acquired by the acquisition unit; a decision unit that decides a response process corresponding to the type of odor determined by the odor determination unit; Equipped with The odor determining device, wherein the decision unit, if executing a plurality of the response processes simultaneously would cause the odor to worsen, discards at least one of the response processes and executes the other.

2. The odor determination device according to claim 1 , wherein the odor determination unit extracts features of an odor type from the plurality of detection values ​​acquired by the acquisition unit, and determines the type of odor based on the features.

3. The odor determination device of claim 1 or 2, wherein the determination unit determines the response processing when the similarity between the odor pattern and a reference pattern set for each candidate odor type is greater than or equal to a predetermined similarity.

4. The odor determination device according to claim 3 , wherein the odor determination unit re-determines the type of odor when the degree of similarity is less than the predetermined degree of similarity.

5. An odor determination device as described in claim 3 or 4, wherein the candidate odor type corresponding to the reference pattern among the multiple reference patterns having the highest similarity a predetermined number of consecutive times is the odor type determination result by the odor determination unit.

6. The odor determination device according to claim 1 , wherein the determination unit determines the corresponding process when the same type of odor is determined a predetermined number of times in succession.

7. The odor determination device according to claim 3 , wherein the determination unit further changes the corresponding process depending on the degree of similarity.

8. 8. The odor determination device according to claim 1, wherein the decision unit changes the response process based on at least one of a temperature and a humidity at a location where the detection element is installed.

9. Obtaining detection values ​​obtained from each of the plurality of detection elements; determining whether the detected value exceeds a preset threshold value; When the detection value of at least one of the plurality of detection elements exceeds the threshold value, a type of odor is determined based on an odor pattern obtained by patterning the plurality of acquired detection values; determining a response process corresponding to the determined type of odor; If the odor worsens when a plurality of the countermeasures are executed simultaneously, at least one of the countermeasures is discarded and the other is executed. A method for determining an odor comprising the steps of:

10. An odor determination system having a plurality of detection elements and an external device, an acquisition unit that acquires detection values ​​obtained from each of the plurality of detection elements; A determination unit that determines whether the detection value exceeds a preset threshold value; an odor determination unit that, when the detection value of at least one of the plurality of detection elements exceeds the threshold value, determines the type of odor based on an odor pattern obtained by patterning the plurality of detection values ​​acquired by the acquisition unit; a decision unit that decides a response process corresponding to the type of odor determined by the odor determination unit; a control unit that outputs control information instructing the external device to execute the response process determined by the determination unit; Equipped with In an odor determination system, when the odor worsens when a plurality of the response processes are executed simultaneously, the decision unit discards at least one of the response processes and executes the other.

Images