Automatic cooking device

Abstract

To provide an automatic cooking device that can easily prepare delicious cuisines automatically.SOLUTION: One aspect of the present invention is an automatic cooking device including: a main body that houses a pot part; a cooking utensil unit that is inserted into the pot part; a control unit that performs control about cooking; a main body side communication unit that is provided in the main body; and a terminal device that has an information input unit, an information output unit, and a terminal side communication unit. The terminal device transmits an image of a food material captured by the information input unit to the main body side communication unit via the terminal side communication unit. The control unit has: a food material calculation unit that calculates a type and a quantity of the food material from the image of the food material received by the main body side communication unit; a menu generation unit that uses artificial intelligence to generate a menu to be a cooking candidate from the type and the quantity of the food material based on preset teacher data; and a cooking control unit that controls an operation of the cooking utensil unit and a temperature of the pot part based on the menu generated by the menu generation unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an automatic cooking device that automatically cooks a desired menu by adding ingredients. [Background technology]

[0002] Patent Document 1 discloses a cooking assistance system that assists in cooking according to a recipe. This cooking assistance system has a control unit that accepts specification of information indicating a cooking procedure that includes a first term that describes the internal state of a cookware, receives data detected by a sensor that senses the internal state of the cookware, references a first memory area that stores one or more pairs of data detected by the sensor that senses the internal state of the cookware and terms that describe the internal state of the cookware, and, if the received data is considered to match the data associated with the first term in the first memory area, outputs both information indicating that the current internal state of the cookware corresponds to the first term and information indicating that it is time to proceed to the next step in the cooking procedure.

[0003] Patent Document 2 discloses a processing device that eliminates the need for human checks when processing an object by supplying energy and ensures uniformity in the state of the finished object. This processing device includes an energy supply means that supplies energy to the object, a motion detection means that detects the dynamic state of the object, and an energy supply control means that controls the amount of energy supplied to the object by the energy supply means. In this processing device, the energy supply control means controls the amount of energy supplied to the object by the energy supply means in accordance with the dynamic state of the object detected by the motion detection means. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2019-020981 [Patent Document 2] Japanese Patent Application Laid-Open No. 2015-068543 Summary of the Invention [Problem to be solved by the invention]

[0005] In an automatic cooking device that automatically prepares a desired menu by adding ingredients, it is desirable to automatically complete a desired dish with as little effort as possible.Furthermore, it is not only the ability to cook automatically, but also delicious food.

[0006] An object of the present invention is to provide an automatic cooking device that can easily and automatically prepare delicious dishes. [Means for solving the problem]

[0007] One aspect of the present invention is an automatic cooking device comprising a main body that houses a pot, a cooking utensil section that is inserted into the pot, a control section that controls cooking, a main body communication section provided in the main body, and a terminal device having an information input section, an information output section, and a terminal communication section, wherein the terminal device transmits images of ingredients captured by the information input section to the main body communication section via the terminal communication section, and the control section comprises an ingredient calculation section that calculates the type and amount of ingredients from the image of the ingredients received by the main body communication section, a menu generation section that generates menu candidates for cooking from the type and amount of ingredients using artificial intelligence based on preset teacher data, and a cooking control section that controls at least one of the operation of the cooking utensil section and the temperature of the pot based on the menu generated by the menu generation section.

[0008] With this configuration, the ingredient calculation unit calculates the types and amounts of ingredients from images of ingredients captured by the information input unit of the terminal device, and artificial intelligence generates a menu of cooking options based on the calculated types and amounts of ingredients. Based on this menu, the cooking control unit controls at least one of the operation of the cooking utensil unit and the temperature of the pot unit, thereby automatically cooking based on the menu derived from the ingredients.

[0009] In the automatic cooking device, the menu generator may extract supplementary ingredients based on the difference between the types and amounts of ingredients required to prepare the candidate menu and the types and amounts of ingredients calculated from the image by the ingredient calculator, and the main body communication unit may transmit information about the supplementary ingredients to the terminal device. This sends information about the shortage of added ingredients to the terminal device, and notifies the user of the ingredients that need to be replenished.

[0010] The automatic cooking device may further include an image input unit provided in the main body, and the main body communication unit may transmit the image of the pot during cooking input by the image input unit to the terminal device. This allows the image of the pot during cooking to be sent to the terminal device, allowing the user to understand the progress of cooking by referring to the image of the pot during cooking on the terminal device.

[0011] In the automatic cooking apparatus, the terminal device may transmit cooking correction information for the operation of the cookware unit and the temperature of the pot unit received from the user from the terminal communication unit to the main body communication unit, and the cooking control unit may correct the control of at least one of the operation of the cookware unit and the temperature of the pot unit based on the correction information received by the main body communication unit. In this way, the user can use the terminal device to correct at least one of the operation of the cookware unit and the temperature of the pot unit.

[0012] In the automatic cooking device, the information input unit of the terminal device may receive user comments about the dish cooked in the pot unit based on the menu generated by the menu generation unit, the terminal communication unit of the terminal device may transmit the comments to the main unit communication unit, and the menu generation unit of the control unit may reflect the comments received by the main unit communication unit in the training data. In this way, the user's comments about the dish are reflected in the training data, and the level of deliciousness when generating a menu using artificial intelligence is improved.

[0013] In the automatic cooking device, the cooking control unit may control at least one of the temperature of the pot and the operation of the cooking utensil unit according to color components based on the Maillard reaction obtained from an image of the ingredients in the pot during cooking. This allows the cooking state of the ingredients to be determined according to the color components based on the Maillard reaction, and controls the heat level, etc.

[0014] In the automatic cooking device, the cooking implement unit may have a stirring part rotatably mounted within the pot, a rotatable lid part positioned above the stirring part within the pot, and a clutch provided between the lid part and the shaft part to interrupt the transmission of the rotational force of the shaft part to the lid part. This allows for variation in the stirring action on the ingredients to ensure effective stirring.

[0015] In the automatic cooking device, the main unit or the terminal device may be connected via a network to a database server that stores type and amount data that indicates associations between images of ingredients and the types and amounts of ingredients, training data, and menu data, and the ingredient calculation unit may calculate the types and amounts of ingredients from the images of ingredients using the type and amount data stored in the database server, and the menu generation unit may generate menus using the menu data stored in the database server. This allows a wide variety of menus to be generated using the vast amount of data stored in the database server connected via the network.

[0016] In the automatic cooking device, the menu data may include cooking procedure data for cooking based on the menu, and the cooking control unit may control at least one of the operation of the cooking utensil unit and the temperature of the pot unit using the cooking procedure data stored in the database server. This eliminates the need to store large amounts of data in the main unit or terminal device, since cooking is performed using the cooking procedure data stored in the database server.

[0017] The above-mentioned automatic cooking device may further include an umami component detection unit that detects the amount of umami components, which are at least one of glutamic acid, guanylic acid, and inosinic acid, in the ingredients being cooked, and the database server stores ingredient umami component data that indicates the correspondence between the types and amounts of ingredients before cooking and the umami components, and menu umami data that indicates the correspondence between menus and the umami components, and the cooking control unit may control at least one of the operation of the cooking utensil unit and the temperature of the pot unit so that the amount of umami components detected by the umami component detection unit approaches the amount of umami components corresponding to the menu using the ingredient umami data and menu umami data stored in the database server.

[0018] In addition, the cooking control unit may control at least one of the operation of the cooking utensil unit and the temperature of the pot unit so that the amount of umami components detected by the umami component detection unit is greater or less than the amount of umami components corresponding to the menu using the ingredient umami data and menu umami data stored in the database server.

[0019] In addition, the cooking control unit may control at least one of the operation of the cooking utensil unit and the temperature of the pot unit so that the amount of umami components detected by the umami component detection unit becomes the type and amount of umami components determined by artificial intelligence based on preset teacher data for umami components.

[0020] This allows the amount of umami components in the ingredients being cooked to be detected by the umami component detection unit, and then, based on the detected amount of umami components, the operation of the cooking utensil unit and / or the temperature of the pot unit are controlled using the ingredient umami data and menu umami data stored in the database server to approach the amount of umami components corresponding to the menu, or to increase or decrease the amount of umami components, or to achieve the type and amount of umami components determined by artificial intelligence, thereby automatically preparing a delicious dish based on the amount of umami components.

[0021] Here, glutamic acid refers to L-glutamic acid, and is referred to simply as "glutamic acid" in this embodiment. Guanylic acid refers to guanosine monophosphate, GMP, or 5'-guanylic acid, and is referred to simply as "guanylic acid" in this embodiment. Inosinic acid refers to inosine 5'-monophosphate, or IMP, and is referred to simply as "inosinic acid" in this embodiment. [Effects of the Invention]

[0022] According to the present invention, it is possible to provide an automatic cooking device that can automatically prepare delicious dishes easily. [Brief explanation of the drawings]

[0023] [Figure 1] 1 is a schematic diagram illustrating the configuration of an automatic cooking device according to an embodiment of the present invention; [Figure 2] 1 is a functional block diagram illustrating an automatic cooking device according to an embodiment of the present invention. [Figure 3] 10 is a flowchart illustrating a cooking method using the automatic cooking device according to the present embodiment. [Figure 4] 10 is a flowchart illustrating a cooking process. [Figure 5] FIG. 1 is a schematic diagram illustrating an example of a rotary tool. [Figure 6] FIG. 1 is a schematic diagram illustrating an example of a rotary tool. [Figure 7] FIG. 1 is a schematic diagram illustrating an example of a rotary tool. [Figure 8] 10(a) to 10(c) are schematic plan views illustrating clutch operation. [Figure 9] 5A and 5B are schematic diagrams illustrating an example of temperature detection by a temperature detection unit. [Figure 10] (a) to (c) are schematic diagrams showing an example of bread baking. [Figure 11] FIG. 10 is a diagram showing an example of an interactive cooking check between a user and an AI regarding the degree of browning of bread. [Figure 12] FIG. 10 is a diagram showing an example of input of a user review. DETAILED DESCRIPTION OF THE INVENTION

[0024] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the following description, the same components are designated by the same reference numerals, and the description of components that have already been described will be omitted as appropriate.

[0025] (Configuration of automatic cooking device) FIG. 1 is a schematic diagram illustrating the configuration of an automatic cooking device according to this embodiment. The automatic cooking device 1 according to this embodiment is a device that can automatically create and cook a menu by adding ingredients M. The automatic cooking device 1 includes a main body unit 10, a cooking utensil unit 20, a control unit 30, a main body communication unit 40, and a terminal device 50.

[0026] Main body 10 is, for example, box-shaped and has an openable and closable lid 15 attached. Main body 10 houses pot 11, which contains food ingredients M and cooks them. Main body 10 has heater 12 arranged around pot 11. Heater 12 is provided on the outside of the periphery or bottom of pot 11. Heater 12 may be an electric heating wire type or an IH (Induction Heating) type. It is preferable that heater 12 be able to control the heating of subdivided areas on the periphery or bottom of pot 11.

[0027] Cooking utensil section 20 is a cooking utensil inserted into pot section 11. Cooking utensil section 20 has rotating tool 21 housed in main body section 10. Rotating tool 21 has shaft section 211 and spatula section 212 attached to the tip of shaft section 211. For example, shaft section 211 is rotated by motor 25 attached to lid 15, which causes spatula section 212 to rotate inside pot section 11 and stir ingredients M. Rotating tool 21 is detachable from motor 25 and can be replaced depending on the cooking contents.

[0028] The control unit 30 is a part that controls cooking. The control unit 30 is provided, for example, in the lid 15 or the main body unit 10. The control unit 30 may also be provided externally via a network. The functional configuration of the control unit 30 will be described later.

[0029] The main body side communication unit 40 is provided in the main body unit 10, for example, and inputs and outputs information to and from the terminal device 50.

[0030] The terminal device 50 is, for example, a smartphone or a tablet terminal. The terminal device 50 inputs and outputs information to and from the main body unit 10. The terminal device 50 is operated by a user and can send various settings and instructions related to cooking to the main body unit 10, and can also receive and display information related to cooking sent from the main body unit 10. The functional configuration of the terminal device 50 will be described later. The terminal device 50 may be configured separately from the main body unit 10, like a smartphone, or may be integrated into the main body unit 10.

[0031] (Functional configuration of the automatic cooking device) FIG. 2 is a functional block diagram illustrating the automatic cooking device according to this embodiment. The automatic cooking apparatus 1 has an image input unit 110, a temperature detection unit 120, an umami component detection unit 130, and a main body communication unit 40. The image input unit 110 is, for example, a camera, and has the function of acquiring images of ingredients M placed in the pot unit 11 and images of ingredients M being cooked. The image input unit 110 is provided, for example, inside the lid 15 (see FIG. 1). The camera lens may be provided with an anti-fogging function. For example, a motor 25 that rotates the rotating device 21 may be used as a drive source to operate a wiper to remove fogging from the lens, or an anti-fogging film may be attached or applied to the surface of the camera lens. The image input unit 110 may also have the function of importing images from a database server SV or the like via a network N.

[0032] Temperature detection unit 120 has the function of detecting the temperature of pot unit 11 and the temperature of food material M being cooked. Examples of temperature detection unit 120 include an infrared sensor (infrared camera), a resistance temperature detector, and a thermocouple. Umami component detection unit 130 has the function of detecting umami components contained in food material M being cooked. Umami component detection unit 130 detects the amount of amino acids and nucleic acids that become umami components by, for example, ultraviolet absorption or fluorescence analysis. The method of detecting umami components by umami component detection unit 130 will be described later.

[0033] The main body side communication unit 40 has a function of communicating information with the terminal device 50 by wireless communication or wired communication. The main body side communication unit 40 also has a function of communicating information with the database server SV via the network N.

[0034] The control unit 30 of the automatic cooking device 1 has an ingredient calculation unit 101, a menu generation unit 102, and a cooking control unit 103. The ingredient calculation unit 101 has a function of calculating the type and amount of ingredient M from an image of ingredient M received by the main body communication unit 40. The menu generation unit 102 generates cooking candidate menus from the type and amount of ingredient M using artificial intelligence (hereinafter also referred to as "AI") based on preset training data. The cooking control unit 103 controls at least one of the operation of the cooking utensil unit 20 and the temperature of the pot unit 11 based on the menu generated by the menu generation unit 102.

[0035] The terminal device 50 has an information input unit 510, an information output unit 520, a terminal-side communication unit 530, and a control unit 540. The information input unit 510 has a function of inputting information such as text, images, and audio. Examples of the information input unit 510 include a touch panel, buttons, a camera, and a microphone. The information output unit 520 has a function of outputting information such as text, images, and audio. Examples of the information output unit 520 include a display, a lamp, and a speaker.

[0036] The terminal communication unit 530 has a function of communicating information with the main body 10 via wireless communication or wired communication. The terminal communication unit 530 also has a function of communicating information with the database server SV via the network N. The control unit 540 controls each unit of the terminal device 50. The control unit 540 also executes application software (hereinafter also referred to as "app") that performs various settings and instructions for automatic cooking.

[0037] (Cooking method using automatic cooking equipment) 3 is a flowchart illustrating a cooking method using the automatic cooking device according to this embodiment. The automatic cooking device 1 according to this embodiment uses AI for various processes. While the AI ​​training data and AI engine may be stored in the main body 10, it is preferable to use those stored in a database server SV connected via the network N in order to reduce the storage capacity of the main body 10.

[0038] First, as shown in step S101, the type and amount of ingredients are calculated. For example, a user arranges ingredients M on a table and captures an image using the information input unit 510 (e.g., a camera) of the terminal device 50, which is then sent from the terminal communication unit 530 to the main unit 10. The main unit communication unit 40 receives the image sent from the terminal device 50 and sends it to the control unit 30. Note that the image of ingredients M may also be captured by the camera of the image input unit 110, capturing an image of ingredients M placed in the pot unit 11. The ingredient calculation unit 101 of the control unit 30 calculates the type and amount of ingredients M based on the image of ingredients M. For example, the type and amount of ingredients M are automatically calculated using AI based on training data that indicates the correspondence between the image of ingredients M and the type and amount.

[0039] Next, as shown in step S102, menu generation is performed using AI. The menu generation unit 102 of the control unit 30 generates cooking candidate menus from the types and amounts of ingredients M using AI, based on training data for generating menu patterns based on the preset types and amounts of ingredients M. Since there may be multiple menus that can be predicted from the types and amounts of ingredients M, multiple menu candidates may be generated.

[0040] Cooking is essentially a heat sequence (the correlation between heating time and heating temperature). Therefore, by utilizing the self-learning function, adjustments, and flexibility of AI, a wide range of menu candidates can be generated based on the type and amount of ingredients M. For example, it is not limited to menus that use all of the ingredients M currently prepared, but it is also possible to generate menu candidates that use some of the ingredients M, or menus that use the prepared ingredients M but lack some ingredients M. When generating menus using AI, the menu may be generated taking into account the user's past menu selections and preference history.

[0041] Next, as shown in step S103, a menu is selected. The menu candidates generated in the previous step S102 are transmitted from the main body communication unit 40 to the terminal device 50. The information of the menu candidates transmitted from the main body communication unit 40 is displayed on the information output unit 520 (for example, a display) of the terminal device 50. The user selects a preferred menu from the displayed menu candidates using the information input unit 510 (for example, a touch panel) of the terminal device 50. The number of people (how many people to cook for) may also be selected along with the menu selection. Information on the selected menu is transmitted from the terminal communication unit 530 to the main body unit 10.

[0042] Next, as shown in step S104, it is determined whether or not ingredients are to be replenished. Upon receiving the menu information transmitted from terminal device 50, main body communication unit 40 transmits the menu information to menu generation unit 102. Menu generation unit 102 extracts replenishment ingredients based on the difference between the types and amounts of ingredients needed to cook the menu selected by the user and the types and amounts of ingredients M calculated from the images of ingredients M by ingredient calculation unit 101. For example, if there are not enough ingredients M for the number of people desired by the user, or if a specific ingredient M is missing, the type and amount of the missing ingredient M is extracted. If replenishment ingredients are available, information about the replenishment ingredients is transmitted to terminal device 50 as shown in step S105. The user can refer to the information about replenishment ingredients transmitted to terminal device 50 to prepare the replenishment ingredients.

[0043] Next, as shown in step S106, ingredients M are placed into pot section 11. If replenishment ingredients are needed, these are also placed into pot section 11. If ingredients M have already been placed in pot section 11, replenishment ingredients are placed in step S106, or if no replenishment ingredients are needed, the process proceeds to the next step.

[0044] Next, cooking processing is performed as shown in step S107. Cooking control unit 103 of control unit 30 controls cooking utensil unit 20 according to the menu for the ingredients M placed in pot unit 11. Cooking control unit 103 controls, for example, motor 25 in accordance with the cooking procedure corresponding to the menu. As a result, cooking utensil unit 20 rotates at a predetermined timing, and cooking such as kneading, stirring, and turning ingredients M in pot unit 11 is performed.

[0045] If it is necessary to replace cooking utensil unit 20 during cooking, information on the timing of replacement and the cooking utensil unit 20 to be replaced is transmitted to terminal device 50. During cooking, cooking control unit 103 controls the supply of electricity to heater 12 at a predetermined timing to apply heat to ingredients M in pot unit 11. In pot unit 11, the necessary cooking (stirring and heating) of ingredients M is automatically performed according to the cooking procedure corresponding to the menu, and the dish corresponding to the menu is automatically completed.

[0046] (Cooking process) The automatic cooking device 1 according to this embodiment can grasp the state of the ingredients M during cooking and automatically control the cooking process. 4 is a flowchart illustrating the cooking process, which is an example of the cooking process in step S107 shown in FIG.

[0047] First, as shown in step S201, the state of the food ingredients M is detected. For example, at a predetermined timing during cooking, an image of the food ingredients M in the pot unit 11 is acquired by the image input unit 110 (e.g., a camera) provided in the main body unit 10. The temperature of the food ingredients M during cooking is detected by the temperature detection unit 120. The amount of umami components in the food ingredients M during cooking may also be detected by the umami component detection unit 130. Information on the detected state of the food ingredients M during cooking may be transmitted to the terminal device 50. This allows the user to refer to information such as an image and temperature of the food ingredients M during cooking on the information output unit 520 (e.g., a display) of the terminal device 50.

[0048] Next, as shown in step S202, an AI-based cooking check is performed. The control unit 30 uses the AI ​​to determine whether cooking is progressing correctly based on the information on the state of the food ingredient M being cooked detected in the previous step S201. For example, the cooking control unit 103 of the control unit 30 recognizes the cooking status based on color components based on the Maillard reaction obtained from an image of the food ingredient M in the pot unit 11 during cooking. The cooking status may also be recognized based on the amount of umami components detected by the umami component detection unit 130.

[0049] Next, as shown in step S203, it is determined whether or not cooking modification is necessary. The control unit 30 determines whether or not cooking modification is necessary based on the results of the cooking status check by the cooking control unit 103. For example, the cooking state of the food material M is ascertained based on color components based on the Maillard reaction obtained from an image of the food material M, and whether or not adjustments such as heat level are necessary. If it is determined that cooking modification is necessary (Yes in step S203), cooking modification is performed as shown in step S204. For example, the cooking control unit 103 controls the amount of power supplied to the heater 12 to adjust the heat level for the food material M. Furthermore, the heat level or the degree of stirring for the food material M may be adjusted based on the amount of umami components detected by the umami component detection unit 130.

[0050] The cooking correction shown in step S204 can be performed in the following manner. (Method 1) At least one of the operation of the cooking utensil section 20 and the temperature of the pot section 11 is controlled so that the amount of umami components detected by the umami component detection section 130 approaches the amount of umami components corresponding to the menu using the ingredient umami data and menu umami data stored in the database server SV. (Method 2) At least one of the operation of the cooking utensil section 20 and the temperature of the pot section 11 is controlled so that the amount of umami components detected by the umami component detection section 130 is greater or less than the amount of umami components corresponding to the menu using the ingredient umami data and menu umami data stored in the database server SV. (Method 3) At least one of the operation of the cooking utensil section 20 and the temperature of the pot section 11 is controlled so that the amount of umami components detected by the umami component detection section 130 becomes the type and amount of umami components determined by AI based on preset training data for umami components.

[0051] On the other hand, if no cooking correction is required (No in step S203), the process proceeds to step S205 without any cooking correction.

[0052] Next, as shown in step S205, it is determined whether cooking is complete. If cooking is not complete (No in step S205), the process returns to step S201 and the subsequent processes are repeated. If cooking is complete (Yes in step S205), cooking ends.

[0053] In this way, in this embodiment, AI is used to perform everything from ingredients to menu options and automatic cooking, so the more it is used, the more training data is accumulated, and through learning, the range of menu options expands and cooking control for creating delicious dishes improves.

[0054] (Example of a rotating tool) 5 to 7 are schematic diagrams showing examples of rotary instruments. 5 shows an example of a rotating tool 21 suitable for stirring food ingredients M. This rotating tool 21 includes a shaft 211 and a spatula 212 provided on the tip side of the shaft 211. A turning part 213 is provided at the tip of the spatula 212. In this rotating tool 21, the spatula 212 rotates together with the shaft 211 to stir the food ingredients M, and the stirred food ingredients M can be mixed by being turned over by the turning surface 213a of the turning part 213. The rotation direction of the spatula 212 may be unidirectional, bidirectional, or may be switched randomly.

[0055] FIG. 6 shows an example of a rotating tool 21 suitable for kneading dough. As shown in the side view of FIG. 6(a), the rotating tool 21 includes a shaft 211, a spatula 212 provided at the tip of the shaft 211, and a lid 214 provided at the center of the shaft 211 (above the spatula 212). The surface 214a of the lid 214 facing the spatula 212 is circularly recessed. In this rotating tool 21, the spatula 212 rotates together with the shaft 211 to stir the ingredients M, while the circularly recessed surface 214a of the lid 214 rotates the ingredients M scooped upward. This facilitates a higher degree of kneading of the ingredients M, such as dough.

[0056] 6(b), the shaft 211 of the rotating tool 21 may be rotated about its axis, and the entire rotating tool 21 may be rotated within the pot 11. This allows the spatula 212 to rotate about its axis of the shaft 211 and revolve within the pot 11, thereby effectively kneading the food material M such as dough.

[0057] FIG. 7 shows an example of a rotary tool 21 equipped with a clutch function. 8(a) to 8(c) are schematic plan views illustrating clutch operation. For ease of explanation, Figs. 8(a) to 8(c) show the positional relationship between the shaft portion 211 and the elliptical hole 215a as viewed in the axial direction, and omit the lid portion 214 shown in Fig. 7. This rotating tool 21 has a clutch 215 between the lid portion 214 and the shaft portion 211. In this rotating tool 21, the shaft portion 212a of the spatula portion 212 is attached at a position offset from the center of the lid portion 214, and the rotation of the lid portion 214 causes the spatula portion 212 to rotate, thereby stirring the ingredients M. When stirring the ingredients M with the spatula portion 212, the ingredients M scooped upward hit the circular recessed surface 214a of the lid portion 214, causing the lid portion 214 and the ingredients M to rotate.

[0058] Lid portion 214 is rotated intermittently by clutch 215. As an example, clutch 215 has an elliptical hole 215a provided in the center of lid portion 214. Shaft portion 211 passes through elliptical hole 215a, and stoppers 211a are provided on shaft portion 211 at positions above and below elliptical hole 215a, and the lid portion 214 is positioned relative to shaft portion 211 in the up and down direction.

[0059] As shown in FIGS. 8(a) to 8(c), at least one of the shaft portion 211 and the elliptical hole 215a may be provided with protrusions, teeth, or irregularities to increase friction. In the example shown in FIG. 8, protrusions 211b are provided on the outer peripheral surface of the shaft portion 211, and protrusions 215c are provided on the inner wall surface of the elliptical hole 215a. A clutch shoe 215b may also be provided on the inner wall surface along the major axis of the elliptical hole 215a. Examples of materials for the clutch shoe 215b include elastic bodies such as rubber and elastomer, and engineering plastics such as nylon and polyacetal. The relative positional relationship between the shaft portion 211 and the elliptical hole 215a can be changed within the tolerance range of the shaft portion 211 and the elliptical hole 215a.

[0060] 8(a), when shaft portion 211 is not in contact with elliptical hole 215a, the rotation of shaft portion 211 is not transmitted to lid portion 214. Even in this case, when food ingredient M comes into contact with lid portion 214, the movement of food ingredient M may be transmitted to lid portion 214, causing it to rotate.

[0061] As shown in Figures 8(b) and (c), when food ingredient M comes into contact with lid portion 214, the position of lid portion 214 changes, and when shaft portion 211 comes into contact with elliptical hole 215a, the rotation of shaft portion 211 is transmitted to lid portion 214, and the rotational force of shaft portion 211 is imparted to lid portion 214.

[0062] 8(b), when the shaft 211 comes into contact with the curved surface (the curved surface with a small radius) on the major axis of the elliptical hole 215a, the cover part 214 rotates in a swinging manner (eccentric rotation) with the position of the curved surface as the approximate center. At this time, the frictional force is increased by the provision of the clutch shoe 215b, and the rotational force is transmitted more reliably.

[0063] 8(c), when the shaft 211 comes into contact with the curved surface (the curved surface with a large radius) on the minor axis of the elliptical hole 215a, the lid portion 214 rotates with a large movement component in the major axis direction (oscillating rotation). At this time, the frictional force is increased by the provision of the protrusions 211b and 215c, and the force is transmitted more reliably.

[0064] The states shown in Figures 8(a) to (c) occur randomly depending on how the food ingredient M hits the lid portion 214, so the lid portion 214 and the spatula portion 212 connected to it also undergo complex movements, which changes the stirring action on the food ingredient M and makes it possible to stir the food ingredient M efficiently.

[0065] Since the shaft portion 212a of the spatula 212 is offset from the center of the lid portion 214, when no external force is applied to the lid portion 214 (initial state), the lid portion 214 is disposed at an angle to the shaft portion 211 so that the side where the spatula 212 is provided is lowered. Therefore, by aligning the offset direction of the shaft portion 212a with the major axis direction of the elliptical hole 215a, the shaft portion 211 and the elliptical hole 215a can be brought into contact with each other as shown in FIG. 8(b) in the initial state, and rotation of the lid portion 214 can be started whenever rotation of the shaft portion 211 starts. Furthermore, although the elliptical hole 215a is shown in the above example, a circular, oval, or polygonal (e.g., Reuleaux triangle) hole may be used instead of the elliptical hole 215a.

[0066] Furthermore, the cross-sectional shape of shaft portion 211 may be other than circular, for example, polygonal, asymmetrical, or teardrop-shaped (semicircular + triangular, etc.). This ensures that shaft portion 211 hits the wall surface of elliptical hole 215a in the minor axis direction at least once per rotation of shaft portion 211, making it easier to prevent a state of non-contact between shaft portion 211 and elliptical hole 215a (so-called missed strikes). When the cross-sectional shape of shaft portion 211 is teardrop-shaped, an elastic body may be provided at the apex of the triangle. This makes it easier for shaft portion 211 to hit other wall surfaces of elliptical hole 215a (for example, wall surfaces in the major axis direction) due to the elastic restoring force of the elastic body, thereby improving the effectiveness of preventing so-called missed strikes.

[0067] In the rotating tool 21 using the clutch 215, the clutch shoe 215b is used as a brake to reverse the axial rotation from clockwise to counterclockwise (or vice versa), and the shear force generated at this time twists the dough. Alternatively, the kneading of the dough can be controlled by using a transparent lid to monitor the kneading of the dough with a camera, and reversing the axial rotation after confirming that the rotation of the rotating tool 21 has stopped.

[0068] (calorie calculation using automatic cooking equipment) When automatic cooking is performed in the automatic cooking device 1 according to this embodiment, the control unit 30 performs the following calorie calculation. First, there are two formulas for calculating heat quantity: (Equation 1) Heat quantity (Q) = mass (m) x specific heat capacity (c) x temperature change (ΔT) (Equation 2) W (watt) = J (joule) / sec (second), or J = W x sec For example, if 1000cc of water at 100°C is boiled for 10 minutes, then in equation 1, The specific heat capacity of water is approximately 4.18 J / g°C. ·The temperature change is 100℃-20℃ (room temperature) = 80℃. Therefore, energy = 1000g x 4.18J / g℃ x 80℃ = 334400J.

[0069] As shown in (Equation 2), 1W (watt) is 1J (joule) per second, so the wattage for boiling for 10 minutes (600 seconds) is 334400J ÷ 600sec = approximately 557W. In other words, it will heat for 10 minutes (600 seconds) at approximately 557W.

[0070] Next, we will explain how to calculate the amount of heat when cooking one cup of rice. The specific heat capacity (c) is assumed to be approximately 4.18 J / g°C for water and 3.75 J / g°C for rice. The mass of one cup (180ml) of rice is generally about 150g for white rice. The amount of water needed to cook 1 cup of rice is about 1.2 to 1.3 times the volume of 1 cup of rice (180 ml). Therefore, the amount of water needed is about 220 g. ·The temperature change is 100℃-20℃ (room temperature) = 80℃. Therefore, by simple calculation, energy = 220g x 4.18J / g℃ x 80℃ + 150g x 3.75J / g℃ x 80℃ = 118,568J.

[0071] The details of cooking rice will be discussed later, but as a theoretical calculation, the above calorie calculation formula can be applied to heating other foods, just like heating one cup of rice. For example, if you are cooking a certain food (such as meat), you can calculate the amount of heat using the common cooking method of 100g x 100°C x 15 minutes, calculate the J (joules) and W (watts), and then vary the temperature and heating time within that J (joule) range. These are generally referred to as low-temperature cooking, slow cooking, or time-saving cooking (which can also be interpreted as high-temperature cooking, or a derivative term, cooking in a pressure cooker), and can also be applied to the automatic cooking device 1 of this embodiment.

[0072] The assumed heating temperature conditions for low-temperature cooking and slow cooking are shown below. Slow cooking temperatures are typically between 50°C and 85°C. In general, slow cooking is a method of cooking food at a low temperature for a long period of time to preserve nutrients and flavor from the heat while tenderizing ingredients (especially meat and fish).

[0073] The heating temperature for low-temperature cooking (sous vide cooking) is approximately 50 to 85°C for meat (e.g., beef, chicken), approximately 45 to 65°C for seafood (e.g., salmon, white fish), and approximately 80 to 85°C for vegetables. In general, sous vide cooking uses similar temperature specifications to slow cooking, but typically involves cooking food in a plastic casing over a water bath. The benefits are generally the same as slow cooking.

[0074] (Temperature detection) FIG. 9 is a schematic diagram illustrating an example of temperature detection by the temperature detection unit. Figure 9(a) shows an example of the temperature distribution detected from above pot 11 by temperature detection unit 120. Figure 9(b) shows an example in which temperature sensor 121 is placed on the bottom side of pot 11. As shown in Figure 1, heater 12 is provided around pot 11 so that the entire pot 11 can be heated.

[0075] When the temperature is detected using, for example, an infrared sensor (infrared camera) as temperature detection unit 120 provided on lid 15, an infrared image distribution of pot portion 11 in a planar view can be obtained, as shown in Figure 9(a). Figure 9(a) shows an infrared image (thermograph) in which pot portion 11 in a planar view is divided into a mesh. The temperature distribution of pot portion 11 in a planar view can be obtained from the correspondence between the amount of infrared light in each mesh and the temperature detected by temperature sensor 121.

[0076] In addition, the temperature of the bottom side of pot 11 can be detected by temperature sensor 121 provided on the bottom side of pot 11 as shown in Figure 9(b). Based on the thermograph from above pot 11 and the temperature detected on the bottom side, cooking control unit 103 precisely controls the power supply to heater 12 and the stirring of ingredients M by rotating device 21. At this time, cooking is checked by AI, leading to the completion of delicious food.

[0077] Next, a specific example of cooking and an example of detecting a state during cooking will be described.

[0078] (Example of baking bread) Figures 10(a) to (c) are schematic diagrams showing examples of bread baking. Figure 10(a) shows an example of baking bread B, Figure 10(b) shows an example of baking using a baking cup CP, and Figure 10(c) shows an example of baking roll bread RB. In all examples, the rotating device 21 (see Figure 1) is shown removed. When baking bread B shown in Figure 10(a), a metal reflector 13 is placed at the upper opening of pan 11 to increase thermal efficiency. By providing reflector 13, heat from the sides of pan 11 can be more easily transferred upward, and heat loss from the upper opening of pan 11 can be prevented. Reflector 13 may have an opening so as not to interfere with the operation of image input unit 110 or temperature detection unit 120. Instead of a metal reflector 13, a reflector 13 made of heat-resistant glass or heat-resistant plastic may be used.

[0079] As shown in Figure 10(b), baking can be performed by placing a baking cup CP containing batter in the pan section 11. By capturing the state of the batter from above the baking cup CP using the image input section 110 and detecting the temperature using the temperature detection section 120, it is possible to obtain optimal baking results while checking the degree of doneness.

[0080] As shown in Figure 10(c), roll-shaped dough can be placed in the pan unit 11 to bake a bread roll RB. In this case, the state of the dough can be captured from above the bread roll RB using the image input unit 110, and the temperature can be detected using the temperature detection unit 120, allowing the user to check the degree of doneness and obtain optimally baked bread.

[0081] (Example of cooking rice) Next, an example of automatic rice cooking using the automatic cooking device 1 according to this embodiment will be described. There is an old saying, "Start with a trickle, then let it boil..." and I will use this to explain. This is a method of cooking rice in an iron pot or a stove, but it can still be applied today. This is because electric rice cookers are made of iron, stainless steel, aluminum, or copper.

[0082] Below, we will explain the details of each stage based on the saying, "Start with a trickle, then pop, and don't take the lid off even if the baby cries." Here, the cooking conditions are 1 cup (150g) of white rice.

[0083] (Stage 1: Beginning trickle) This means "simmering at low heat at first," and is the stage known as pre-cooking or pre-cooking. Allow the rice to absorb the water. Hydrolysis of starch produces sweetness (sugars). It takes about 10 to 15 minutes to boil the water, so the calculation is based on 13 minutes (780 seconds). The automatic cooking device 1 calculates the energy consumption in the same way as the simple calculation above. Energy = 220g x 4.18J / g℃ x 80℃ + 150g x 3.75J / g℃ x 80℃ = 118568J 118568J ÷ 780 seconds = 152W (watts) In other words, it will heat for 780 seconds at 152W (watts).

[0084] (Stage 2: Medium) This means "increasing the heat in the middle," and is the stage known as "hon-taki" or "takiage." -Softens rice and makes it sticky. -Part of the starch is gelatinized (gelatinized) using heat and water. The boiling state (=100°C) is kept for about 15 to 20 minutes, so the calculation is based on 18 minutes (1080 seconds). The automatic cooking device 1 calculates the energy consumption in the same way as the simple calculation above. Energy = 220g x 4.18J / g℃ x 80℃ + 150g x 3.75J / g℃ x 80℃ = 118568J 118568J ÷ 1080 seconds = 110W (watts) In other words, it will heat for 1080 seconds at 110W (watts).

[0085] (Step 3: Don't take the lid off, even if the baby cries) This means "do not remove the lid under any circumstances" and is the stage known as steaming. -Furthermore, it brings out the sweetness, stickiness, and aroma of the rice. The temperature is kept at 90°C or above for about 10 to 15 minutes, so the calculation is based on 13 minutes (780 seconds). The automatic cooking device 1 calculates the energy consumption in the same way as the simple calculation above. 220g×4.18J / g℃×70℃+150g×3.75J / g℃×70℃=103747J 103747J ÷ 780 seconds = 133W (watts) In other words, it will heat for 780 seconds at 133W (watts). Since cooking is done with preheating, 0J or 0W is fine.

[0086] The total amount of heat generated from steps 1 to 3 is 340,883 J (237,136 J if step 3 is 0 J, 0 W).

[0087] However, these are theoretically calculated values. Therefore, the automatic cooking device 1 starts cooking using these theoretically calculated values, and then performs cooking while detecting the state using the image input unit 110 (including a camera and an infrared camera) and a temperature sensor and correcting the amount of heat. In addition, AI machine learning reduces the discrepancy between the theoretically calculated values ​​and the actual amount of heat.

[0088] (The role of the camera) In the automatic cooking apparatus 1 according to this embodiment, the camera, which is an example of the image input unit 110, plays various roles. By using an infrared camera, which is an example of image input unit 110, an image such as that shown in Fig. 9(a) can be obtained, similar to temperature detection unit 120. For example, by photographing rice being cooked from above pot unit 11 with an infrared camera, which is an example of image input unit 110, an image such as that shown in Fig. 9(a) can be obtained. For example, a cylindrical pot is divided with a reticle, and each part is displayed as a thermograph (e.g., color-coded). Naturally, the temperature decreases as you move away from the wall of the pot 11. Since these are only surface temperatures, to grasp the total heat quantity, the temperature across the entire depth is estimated from the bottom surface temperature obtained from the temperature sensor set at the bottom and the temperature of each reticle obtained from the thermograph. Then, the amount of heat Q per unit time (seconds) is calculated, and these are integrated to approach the theoretical value calculated earlier. For example, when calculating the amount of heat Q required to heat 1000cc of water in a pot, where the temperature at the bottom of the pot is 100℃, the temperature of the water surface is 80℃, and the room temperature is 20℃, the calculation can be done using the following formula. Q=m×c×ΔT Here, Q is the amount of heat (J: Joules), m is mass (g or kg), c is the specific heat capacity (J / g°C or J / kg°C), and ΔT is the temperature change (initial temperature - final temperature). Assuming there is 1000cc of water in a pot, and the density of water is approximately 1g / cc, the mass m is 1000g. The specific heat capacity c of water is generally approximately 4.18J / g℃. The initial temperature is 100℃, which is the temperature of the bottom of the pot, and the final temperature is 80℃, which is the temperature of the water surface. Substituting these into the above formula, the amount of heat Q is calculated as follows: Q=1000g×4.18J / g℃×20℃=83600J Therefore, when 1000cc of water is poured into a pot and heated, the amount of heat Q is approximately 83,600J. In this way, the amount of heat (J: Joules) per unit time (seconds) is calculated and integrated using the above formula, and approaches the initial calculated value. The upper surface temperature is the sum of the temperatures of each reticle, and the lower surface temperature is obtained from the temperature sensor. Note that convection is ignored in this case. The above calculation applies when only the surface temperature is measured using an infrared sensor (thermograph), but this does not apply if the infrared sensor can measure the entire area.

[0089] Here, temperature sensors can be tightly attached to the bottom or wall of pot 11, or the temperature can be estimated from the amount of power input (W). However, using a thermograph has many advantages, as it allows you to detect temperature variations and stir the food immediately.

[0090] Stirring the rice during cooking allows for more even heating. Furthermore, after cooking is complete, rotating the stirrer left and right eliminates the need to use a rice paddle.

[0091] Another method of use when cooking rice is to submerge the rice in water (this method is said to be tastier because the rice is more saturated with water), and the automatic cooking device 1 can take images of the rice with a camera and detect the degree of submersion of the rice based on the captured images.Specific examples include a method of observing images of the rice at fixed points and determining the degree of swelling, or a method of observing the rice absorbing water and turning white using images (whitening can also be managed using a white index).In addition, the stirring function can be used to take images of areas other than the surface with a camera to detect the overall degree of submersion of the rice.

[0092] (The camera's role as a safety mechanism) In the automatic cooking device 1, when the camera detects an abnormality, such as sudden boiling or over-boiling during cooking, or the generation of a large amount of white smoke or black smoke, the heating can be stopped immediately, even when cooking simmered dishes other than rice.

[0093] (Image recognition function of the camera and linking of the built-in camera and the camera function of the terminal device) The automatic cooking device 1 can have the AI ​​recognize the raw materials (amount and type) added before cooking and set the amount of heat in advance based on their specific heat capacities. For this purpose, the AI ​​can be made to recognize the raw materials using various methods, but this function can also be achieved by the image input unit 110 (built-in camera) provided in the main body 10, as well as the camera of the terminal device 50 (smartphone, tablet, etc.) used as the UI (user interface).

[0094] (Regarding ingredients other than cameras) The automatic cooking apparatus 1 has a function to determine the type and amount of ingredients M from images of ingredients M captured by the information input unit 510 (camera) of the terminal device 50 or the camera of the image input unit 110 of the main body unit 10, but other methods besides cameras (images) may also be used, such as having the user read out the name of ingredients M and provide it to the control unit 30 as voice data, or providing code numbers and recipe data associated with menus on recipe websites or cookbooks (recipe books) to the control unit 30. In this case, it is preferable to use an AI dialogue function or the like to ask the user whether the ingredients recognized from the provided data match the ingredients actually added.

[0095] (Regarding recipe data input using a camera) In the automatic cooking apparatus 1, an image captured by the image input unit 110 (camera) can be displayed on the information output unit 520 (display) of the terminal device 50 directly or via the network N. A user can view the image captured by the image input unit 110 (camera) on the information output unit 520 (display) of the terminal device 50.

[0096] In addition, when a code number, one-dimensional or two-dimensional code, etc. printed in a recipe book is photographed and read using the information input unit 510 (camera) of the terminal device 50, the read image can be confirmed on the information output unit 520 (display). Furthermore, when the ingredients M (including water) to be added are photographed with the information input unit 510 (camera) and the AI ​​is made to recognize their contents and amounts, the image and the recognition results can be confirmed on the information output unit 520 (display). At this time, the AI ​​can be made to recognize the type and amount of ingredients from the captured image using image recognition technology, and information for confirmation as to whether the contents are correct can be output from the information output unit 520 (display or speaker).

[0097] (A function that recognizes changes in the color of ingredients during cooking) The automatic cooking device 1 may have a function for performing white indexing using an image taken by a camera as a method for determining the degree of soaking of the rice. When rice is completely immersed in water, it turns white, so compare the value digitized by the camera's solid-state imaging element (CCD image sensor, COMOS image sensor, etc.) with the white value below and complete the process with an approximate value.

[0098] As for the "white value," the white value in the Lab value (CIELAB color space) is an L value of 100. This value represents brightness, and indicates that white is the brightest. The a and b values ​​are generally close to 0. In the RGB color space, which uses the three primary colors of light, red (R), green (G), and blue (B), as the "white value," values ​​between 0 and 255 are used to represent the intensity of each color, with all values ​​being 0 being "black" and all values ​​being 255 being "white." It should be noted that Lab values ​​and RGB values ​​can be converted into each other.

[0099] In addition to the above, it is also possible to use training image data of whitened rice to train the AI ​​in machine learning of images of whitened rice in advance, and then have the AI ​​compare the captured images.

[0100] (Maillard reaction) Next, a method for measuring the color of food material M in the Maillard reaction will be illustrated. Two methods are used to measure color: direct reading of stimulus values ​​and spectrophotometric colorimetry. The direct reading of stimulus values ​​is a method of measuring the three colors red, green, and blue in the same way that humans perceive color. On the other hand, spectrophotometric colorimetry measures color by measuring the wavelength of light, and is therefore used to measure more precise shades than direct reading methods.

[0101] In the automatic cooking device 1, when cooking pork miso soup, for example, it is possible to insert a cooking step of lightly frying ingredients beforehand. Therefore, it is necessary to use a camera to monitor the heating state of the food M to prevent the food M from burning. The heating state can be monitored, for example, by the following methods. - AI-based judgment using training image data Direct reading of stimulus values ​​using a camera (solid-state image sensor) A method of adding visible light (spectroscopically reflecting light) to an ultraviolet spectrophotometer (spectroscopically reflecting transmitted light) used in quantifying amino acids and nucleic acids (i.e., spectrophotometric method)

[0102] The Maillard reaction (browning) plays a major role in the discoloration of food M when heated (boiled, baked, roasted, stir-fried, steamed). If the primary color of food material M is transparent or whitish (including cream or yellow), it can be indexed by increasing the amount of brown or black. This method allows the promotion of the Maillard reaction due to heating to be indexed by preparing typical color samples (for example, a color chart in which brown increases by 10%) for each reaction stage. For example, in the case of black, the brightness decreases step by step, eventually carbonizing to black. In the case of brown, the image is of the primary color being overlaid with brown, so the starting point is brown minus the primary color, starting at zero, then the brown increases step by step, and the brightness decreases, eventually becoming black. The color change of the food material M obtained from the camera is compared with a color sample to determine the reaction stage, and heating is stopped when the ideal (specified) state is reached.

[0103] However, color changes due to heating are not limited to the Maillard reaction; they are also caused by a wide range of factors, including a decrease in brightness and saturation due to dehydration, an increase in brightness due to water absorption, the elution of pigments and decomposition due to heating, carbonization of ingredient M, yellowing due to oxidation, and color changes due to the absorption of sauces, etc. In addition, if the primary color of the ingredient is not white or transparent, browning is not detectable due to light absorption. For this reason, it is preferable to use the combination of ingredients M added according to the menu being cooked and the color changes due to heating as training data, and use images of the ingredients M taken during cooking to determine the degree of cooking using AI.

[0104] (AI cooking check) The automatic cooking device 1 according to this embodiment may be equipped with a cooking check function using AI. FIG. 11 shows an example of an interactive cooking check between a user and an AI regarding the degree of browning of bread. In this example, the automatic cooking device 1 detects the degree of browning of the finished toasted bread and performs a cooking check using AI. First, the control unit 30 checks the change over time in the toasting state of the side of the bread (thermograph history). The control unit 30 performs an AI check based on this toasting state history, and if it determines that there is an abnormality, it sends an AI response to the terminal device 50. An example of a response would be, "It appears that overheating occurred between XX and △△ minutes during heating. There is a possibility of a heater malfunction, so in future, if this occurs, we will temporarily stop heating and contact the user." If it determines that there is no abnormality, the AI ​​responds, "We will check other possibilities," and sends this response to the terminal device 50.

[0105] Next, the control unit 30 checks the browning of the top surface of the bread over time due to the Maillard reaction. The control unit 30 performs an AI check based on this browning over time, and if it determines that there is an abnormality, it sends an AI response to the terminal device 50. An example of a response would be, "Overall, the heat was too strong, and the sides in particular appear to have burned because they are close to or in contact with the heater. In the future, please request a lower heat setting when setting the oven." If it determines that there is no abnormality, an AI response such as, "We will check other possibilities," is sent to the terminal device 50.

[0106] Next, the control unit 30 checks the time-dependent change in the toasting state of the top surface of the bread (thermograph history). The control unit 30 performs an AI check based on this toasting state history, and if it determines that there is an abnormality, it sends an AI response to the terminal device 50. An example of a response is, "Overall, the toasting was too strong, and the sides in particular appear to have burned because they are close to or in contact with the heater. In the future, please request a lower heating setting when setting the oven." If it determines that there is no abnormality, the AI ​​responds to the terminal device 50 by saying, "The top surface is toasting without any problems, but the sides appear to have been heated too much. In the future, we will review the heating correlation ratio between the top and sides. We also recommend using a heat reflector." If a reflector was used, the AI ​​responds by saying, "We will check other possibilities."

[0107] Next, the control unit asks the user whether or not they have applied egg yolk or oil to the dough to prevent it from burning. If they have not, an AI-generated response is sent to the terminal device 50. An example of a response is, "We recommend applying egg yolk or oil to prevent the surface of the bread from burning." If they have, an AI-generated response such as, "We will check other possibilities, such as the temperature history," is sent to the terminal device 50.

[0108] The automatic cooking device 1 according to this embodiment may also have a cooking check function based on dialogue between the user and AI. In determining whether or not a recipe modification is necessary, as shown in step S203 of FIG. 4, the determination of whether or not a modification is necessary may be made interactively between the user and the AI. For example, the control unit 30 transmits to the terminal device 50 a recipe modification proposal determined by the AI ​​based on detection results such as images of the ingredients M during cooking, the temperature distribution on the surface of the ingredients M, and the amount of umami components. If the user refers to the AI's recipe modification proposal transmitted to the terminal device 50 and decides to follow it, the AI's recipe modification proposal is executed by sending information indicating consent to the control unit 30. On the other hand, if the user refers to the AI's recipe modification proposal and wishes to modify the recipe at their own discretion, the user sends the control unit 30 their own modification proposal (which may be a fine-tuning of the AI's modification proposal). This results in the recipe modification based on the user's modification proposal. By incorporating such interactions with the user regarding recipe modifications as training data, the accuracy of the AI's recipe modifications can be improved.

[0109] (User Review) The automatic cooking device 1 according to this embodiment may have a function for inputting and reflecting user reviews. Fig. 12 is a diagram showing an example of input of user reviews. As shown in FIG. 12, a screen that allows user reviews to be entered is displayed on the display (touch panel) of the terminal device 50. From this screen display, the user can enter an evaluation (user review) of the menu cooked by the automatic cooking device 1. The user can specify the cooking evaluation by associating it with a score. In addition, the user can enter their impressions as text data in the comments field. Such user reviews are received by the control unit 30 and reflected in the training data. In addition to the above-mentioned AI cooking check function, traceability, machine learning functions, and self-diagnosis functions can be utilized in automatic cooking.

[0110] (Detection of umami components) Next, an example of detection of umami components in the automatic cooking device 1 according to this embodiment will be described.

[0111] (Amount of glutamic acid and umami nucleic acid (GMP and IMP) in each food) The synthesis process of inosinic acid and guanylic acid, which are known as nucleic acid substances and are one of the umami components, is complex. Inosinic acid is often produced when ATP (adenosine triphosphate) is decomposed by enzymes into its original substance, inosinic acid (IMP), when the phosphate energy is released due to cell death and AMP (adenylic acid) is broken down by enzymes. One of the reasons why aged meat is delicious is that glutamic acid is produced when proteins are broken down.

[0112] It is known that dried shiitake mushrooms contain large amounts of guanylic acid, which is often produced when the cell wall breaks down due to cell death, allowing degradative enzymes to come into contact with the nucleic acid base guanine, which is then broken down into guanylic acid (GMP).

[0113] However, inosinic acid and guanylic acid are naturally present in free form in food. In addition, they can be produced by biodegradation through fermentation, and of course by "thermal decomposition." This is why the umami flavor of food is enhanced by "thermal cooking" such as baking, steaming, smoking, roasting, boiling, and frying. Therefore, it is theoretically possible to produce glutamic acid, guanylic acid, and inosinic acid by thermal decomposition using the "thermal cooking" of the automatic cooking device 1.

[0114] (Quantification of umami components) Next, an example of detection of umami components by the umami component detection unit 130 of the automatic cooking apparatus 1 will be described. First, umami components can be detected by optical detection from the soup of stews and soups, and from the broth or stock of simmered dishes. It is possible to create a database of the amount of free glutamic acid in each food ingredient, and also to create a database of the free aromatic amino acids tyrosine, phenylalanine, and tryptophan.

[0115] Since the automatic cooking device 1 uses AI to recognize in advance which ingredients and how much of each ingredient have been added, the amount of free amino acids can be calculated if the amount of free amino acids for each ingredient is stored in a database. Therefore, if any of the free aromatic amino acids is optically detected, it is thought that the amount of free glutamic acid can also be estimated from the correlation ratio.

[0116] The amino acids newly generated by the thermal decomposition of proteins can be considered to be in excess of the amount of free amino acids. In other words, the amount of glutamic acid (mg) newly generated by thermal decomposition is considered to be approximately equal to the total amount of glutamic acid (mg) in the aqueous solution estimated by optical detection of aromatic amino acids in the aqueous solution - the total amount of free glutamic acid (mg) in each food ingredient.

[0117] Regarding proteinogenic amino acids (non-free amino acids), the number of residues of 20 types of constituent amino acids in each protein is known, so if these are compiled into a database, especially if glutamic acid and the three aromatic amino acids are compiled into a database, the amount of free and non-free glutamic acid can be estimated from the correlation ratio with the amount of optically detected aromatic amino acids. Here, the amount of proteinogenic amino acids is shown in the Standard Tables of Food Composition in Japan. Note that the free amounts of guanylic acid and inosinic acid are already known.

[0118] Here, for guanylic acid and inosinic acid, the amount of free aromatic amino acids was quantified and estimated to have dissolved into the broth in relation to the free aromatic amino acids, as with free glutamic acid. The content and content ratio of guanylic acid and inosinic acid in the database above was used. The elution of free amino acids and free guanylic acid and inosinic acid into aqueous solution differs from that of thermal decomposition, so it is thought that all substances are roughly proportional. Regarding guanylic acid and inosinic acid produced by thermal decomposition, the thermal decomposition temperature and the active temperature of the decomposing enzyme are different, so the amounts are calculated using the method shown below, as an example.

[0119] (A method for quantifying nucleic acids and nucleic acid-related substances using 260 nm ultraviolet absorption spectrometry to estimate the amount of eluted guanylic acid and inosinic acid and the amount of newly produced guanylic acid and inosinic acid.) First of all, nucleic acids and nucleic acid-related substances refer to nucleic acids (DNA and RNA) and nucleic acid-related substances (polydeoxynucleotides, deoxynucleotides, and polynucleotides, nucleotides), both of which are water-soluble. Incidentally, inosinic acid and guanylic acid are nucleotides.

[0120] Next, a method for quantifying guanylic acid and inosinic acid without using aromatic amino acids is described below. First, the "slow cooking and low temperature cooking" method described above will be shown again. Slow cooking temperatures are typically between 50°C and 85°C. The heating temperature for low-temperature cooking (sous vide cooking) is approximately 50°C to 85°C for meat (e.g., beef, chicken), approximately 45°C to 65°C for seafood (e.g., salmon, white fish), and approximately 80°C to 85°C for vegetables. Including the above, the normal cooking temperature of 90°C to 100°C is also included in the quantification.

[0121] (1) A database will be created of the amount of free glutamic acid, free guanylic acid, and free inosinic acid per unit weight (100g) of each food ingredient. (2) Each ingredient is boiled in water, and the amount of nucleic acids and nucleic acid-related substances, glutamic acid, guanylic acid, and inosinic acid in the aqueous solution is analyzed, and the amount of elution is recorded as data. The detection conditions include temperature conditions such as slow cooking, and are set to, for example, 100°C in 5°C increments.

[0122] The cooking time is set from 0 to 12 hours in 15-minute increments, which is 1 / 4 of an hour, to accommodate slow cooking, low-temperature cooking, and long simmering. Using the above detection conditions, the amount of nucleic acids and umami substances is detected in 5°C increments and 15-minute increments.

[0123] In addition, a heating heat cycle is also added to the above detection conditions. This is because, just as it is said that hot pot dishes taste better the second time, and soups and stews taste better the second time or later, it is said that foods that have been heated once and left to stand (aged) taste better. For example, detection is performed on food that has been heated for 2 hours, left to stand for 4 hours, and then heated for a final 30 minutes (at the same heating temperature).

[0124] (3) As with the amount of glutamic acid mentioned above, the amount exceeding the amount released in (1) above is considered to be the amount of umami substance newly produced by thermal decomposition. (4) The ratio (%) of the amount of glutamic acid, inosinic acid, and guanylic acid to the amount of nucleic acids, etc. is determined for each heating temperature and heating time for each food ingredient.

[0125] By integrating (1) through (4) above, it is theoretically possible to estimate the amounts of umami components—glutamic acid, guanylic acid, and inosinic acid—from the amounts of nucleic acids, etc. Furthermore, the correlation between the thermal decomposition temperatures of umami components from each ingredient can also be obtained from each heating temperature and heating time. The combination of glutamic acid, inosinic acid, and guanylic acid is sometimes referred to as flavor doubling, but it is also sometimes referred to as flavor synergism (synergistic effect), and is said to enhance flavor by up to seven to eight times. To bring the umami components during cooking closer to the menu umami data, it is ideal to use the umami component data of the completed menu as a reference. In this case, a database storing data on nucleic acids and nucleic acid-related substances, as well as glutamic acid, guanylic acid, and inosinic acid (completed menu umami data) for the completed menu, and training data are required.

[0126] (Quantification of amino acids and nucleic acids) Next, we will explain the quantification of amino acids and nucleic acids. (Quantitative Method 1) Quantitative Method 1 involves quantifying the increase in tryptophan (second candidate: tyrosine and phenylalanine) or all three aromatic amino acids using (a) ultraviolet absorption analysis or (b) fluorescence analysis, and estimating the amount of L-glutamic acid from the correlation ratio. At the start of quantification, the starting measurement value is set to 0 because free amino acids are present. The increase is then measured from this value. The correlation ratio is the ratio of the number of tryptophan, phenylalanine, and tyrosine residues to the number of glutamic acid residues in each protein.

[0127] Next, we will explain the fluorescence wavelengths of amino acids and their approximate content in proteins. 1. Tryptophan Tryptophan is an essential amino acid and is present in less than 1% of proteins. Its fluorescence properties are 350 nm when excited at 280 nm. 2. Phenylalanine Phenylalanine is an essential amino acid and accounts for less than 5% of protein. It is also a precursor to tyrosine. Its fluorescence characteristics are excitation at 257 nm and emission at 287 nm. 3. Tyrosine Tyrosine accounts for less than 1% of proteins. Its fluorescence properties are excitation at 275 nm and emission at 304 nm. 4. L-Glutamic Acid L-glutamic acid is present in proteins at 6 to 9%. The four amino acids are the building blocks of proteins.

[0128] (Quantitative method 2) Quantitative method 2 is a method in which nucleic acids or nucleotides are quantified using ultraviolet absorption at 260 nm, and while quantifying the increase, either (a) inosinic acid, (b) guanylic acid, or (c) both inosinic acid and guanylic acid in the nucleic acid or nucleotide is estimated from the correlation ratio.

[0129] (About the effects of quantification methods 1 and 2) Quantitative methods 1 and 2 make it possible to quantify, albeit as estimated values, the taste substances (umami) L-glutamic acid, inosinic acid, and guanylic acid. It is also possible to collect user reviews about umami, score them, and convert them into text to collect details, and then create a database that associates umami components with quantified values. Furthermore, the cooking details are recorded in a database as heating times and temperatures (heating sequences), and analyzing this big data will lead to the elucidation of cooking methods that are more flavorful, i.e., more umami-rich. As a result, cooking using the automatic cooking device 1 will continue to evolve into more delicious dishes. Note that analyzing the heating sequences in the big data means obtaining correlations between which ingredients should be heated at what temperature for what minutes to promote the hydrolysis of proteins and nucleic acids and increase umami substances such as amino acids and tasty nucleotides. Furthermore, since L-glutamic acid, inosinic acid, and guanylic acid are also best simmered in a pot for a long time, it is possible to prepare them in the automatic cooking device 1 at night and have them ready the next morning, or to prepare them before going to work in the morning and have them ready when you get home. In addition, it is possible to check the increase in the amount of amino acids and cooking images during cooking via the network N using the terminal device 50.

[0130] (ultraviolet absorption method) Next, the ultraviolet absorption method will be briefly described. In the ultraviolet absorption method, protein concentration (amino acid concentration) is calculated using the Lambert-Beer law, which uses the "extinction coefficient." Note that since proteins are polymers formed by the condensation polymerization of amino acids, proteins can also be referred to as amino acids. Here, C (concentration mg / mL) = A (absorbance at 280 nm) / ε (extinction coefficient) × L (path length).

[0131] (When estimating protein concentration by measuring absorbance at 280 nm) Protein concentration is estimated by the following procedure. (1) Add the sample solution to a 1 cm optical path length cell (such as a quartz cell that can be used for measuring in the ultraviolet range). (2) Measure the absorbance (280 nm). (3) When the absorbance is 1, the concentration is estimated as approximately 1 mg / mL. (4) For a single protein solution, the following formula applies: Molar concentration of protein = absorbance ÷ (number of tyrosine residues in protein × 1390 + number of tryptophan residues × 5800)

[0132] The value varies between proteins because the tyrosine and tryptophan content differs depending on the protein, but in the case of a crude protein solution containing various proteins, when the absorbance at 280 nm (when using a 1 cm path length cell) is 1, the protein concentration of the solution is generally about 1 mg / mL.

[0133] Generally, a protein solution exhibits an ultraviolet absorption peak (around 200 nm to 215 nm) due to peptide bonds and an absorption peak at 280 nm due to the side chains of aromatic amino acids (tyrosine, tryptophan). Furthermore, the extinction coefficients of nucleic acids, DNA and RNA are usually measured at a wavelength of 260 nm, and their values ​​are approximately 1.0 to 2.0.

[0134] For reference, the extinction coefficients are as follows: Tryptophan: approximately 5500 to 5700 cm -1 / M (usually measured at a wavelength of approximately 280 nm in the ultraviolet range) Tyrosine: approximately 1250 to 1300 cm -1 / M (usually measured at a wavelength of approximately 280 nm in the ultraviolet range) Here, to perform measurements using ultraviolet absorption and fluorescence analysis, methods include attaching the above cell and a reflector to a float and floating it inside the pot section 11, or providing a section on the side wall of the pot section 11 to place the cell.

[0135] (fluorescence analysis) Next, fluorescence analysis will be outlined. Tryptophan and some nucleic acids and nucleotides emit fluorescence, which is detected. Fluorescence analysis is characterized by its high detection accuracy.

[0136] Aromatic amino acids such as tryptophan are fluorescent, making fluorometric analysis useful. L-glutamic acid does not fluoresce, and although its rate of thermal decomposition from proteins differs from that of aromatic amino acids, if tryptophan and other amino acids are quantified, an estimated value of L-glutamic acid can be obtained from the correlation ratio.

[0137] (Timer function) The automatic cooking device 1 is well-suited for timer cooking. For example, you might prepare ingredients before going to bed at night, eat the freshly cooked meal in the morning, prepare more ingredients before going out, and eat the same freshly cooked meal when you get home. By performing slow cooking during this time, you can bring out the flavor of the ingredients even more.

[0138] An example of timer cooking is shown below. If you prepare ingredient M at 11pm and it is ready by 7am the next morning, there are 8 hours between then. This is sufficient time for simmered dishes such as stew, and various cooking patterns can be suggested to the user. (Proposal 1) Slow cooking This proposal makes it possible to cook meat and other ingredients softly and bring out the flavor of the ingredients themselves. (Suggestion 2) Simmer slowly Although this proposal is orthodox, it softens ingredient M and also brings out the flavor of ingredient M itself. (Proposal 3) Heat it once, then stop heating, let it "ripen", and then resume heating before waking up to finish. This proposal makes it possible to prevent food material M from spoiling and also to allow food material M to mature.

[0139] In the cooking patterns of proposals 1 to 3 above, the amount of umami components during cooking is detected by the umami component detection unit 130, and the increase in umami components is monitored. The production of amino acids and nucleic acids through hydrolysis involves complex correlations between heating time, heating temperature, pH value, etc., and many aspects remain unknown. The automatic cooking device 1 can analyze thermographs during cooking and monitoring data (big data) on the increase in amino acids and nucleic acids to search for the ideal heating pattern.

[0140] In addition, the automatic cooking device 1 according to this embodiment is also advantageous for cooking brown rice and multigrain rice, which require a long time to soak in water, and by using a camera to monitor the degree of soaking, it is possible to achieve the ideal cooking level.

[0141] Furthermore, the automatic cooking device 1 is also advantageous for fermenting bread. For example, by analyzing images and thermographs taken during cooking, and monitoring data (big data) on the increase in the amount of amino acids and nucleic acids, it is possible to increase the amount of amino acids by fermenting and maturing the bread slowly, and it is also possible to control the fermentation temperature.

[0142] (Integration with cloud services) The automatic cooking device 1 according to this embodiment can link with recipe websites and cooking apps via the network N. It is also compatible with other application software (apps), specifically apps that handle body record data, such as health management apps. The automatic cooking device 1 uses AI to recognize ingredients M before cooking, allowing it to independently calculate nutritional values ​​using databases such as the Standard Tables of Nutritional Composition in Japan. Furthermore, this nutritional data for one meal can be sent, shared, and managed with body record data, and menus and recipes can be recommended based on the nutrient deficiencies identified and the activity and physical condition derived from the body record data.

[0143] As described above, the automatic cooking device 1 according to this embodiment can easily and automatically prepare delicious dishes.

[0144] Although the present embodiment and its application examples (modifications and specific examples) have been described above, the present invention is not limited to these examples. For example, those skilled in the art may appropriately add, delete, or modify components of the above-described embodiments or their application examples (modifications and specific examples), or may appropriately combine features of the embodiments, as long as they include the gist of the present invention. [Explanation of symbols]

[0145] 1...Automatic cooking device 10...Main body 11...Pot section 12...heater 13...Reflector 15…Lid 20...Cooking utensils section 21...Rotating equipment 25...Motor 30...Control unit 40...Main unit communication section 50...Terminal equipment 101…Ingredient calculation department 102...Menu generation unit 103...Cooking control unit 110...Image input unit 120...Temperature detection unit 121...Temperature sensor 130...Umami component detection unit 211...shaft 211a...Stopper 211b…Protrusion 212...Spatula 212a...shaft part 213...Returned part 213a...Flip side 214…Lid part 214a…plane 215…Clutch 215a...Oval hole 215b...Clutch shoe 215c…Protrusion 510...Information input section 520...Information output unit 530...Terminal communication unit 540...Control unit B...bread CP...Baking cup M…Ingredients N...Network RB...Bread roll SV...Database server

Claims

1. a main body that accommodates the pot; a cooking utensil portion inserted into the pot portion; a control unit that controls cooking; a main body side communication unit provided in the main body unit; a terminal device having an information input unit, an information output unit, and a terminal-side communication unit; Equipped with the terminal device transmits the image of the ingredient captured by the information input unit to the main body communication unit via the terminal communication unit; The control unit an ingredient calculation unit that calculates the type and amount of the ingredient from the image of the ingredient received by the main body side communication unit; a menu generation unit that generates cooking candidate menus from the types and amounts of the ingredients using artificial intelligence based on preset training data; and a cooking control unit that controls at least one of the operation of the cooking utensil unit and the temperature of the pot unit based on the menu generated by the menu generation unit.

2. the menu generation unit extracts supplementary ingredients based on a difference between the types and amounts of ingredients required to prepare the candidate menu and the types and amounts of ingredients calculated from the image by the ingredient calculation unit; The automatic cooking apparatus according to claim 1 , wherein the main body communication unit transmits information about the supplementary ingredients to the terminal device.

3. further comprising an image input unit provided in the main body unit, The automatic cooking apparatus according to claim 1 , wherein the main body communication unit transmits the image of the pot during cooking input by the image input unit to the terminal device.

4. the terminal device transmits correction information regarding the operation of the cooking utensil unit and the cooking temperature of the pot unit received from the user from the terminal side communication unit to the main body side communication unit; The automatic cooking device according to claim 1 , wherein the cooking control unit corrects control of at least one of the operation of the cooking utensil unit and the temperature of the pot unit based on the correction information received by the main body side communication unit.

5. the information input unit of the terminal device accepts comments from the user regarding the food cooked in the pot unit based on the menu generated by the menu generation unit; the terminal-side communication unit of the terminal device transmits the comment to the main-body-side communication unit; The automatic cooking device according to claim 1 , wherein the menu generation unit of the control unit reflects the comments received by the main body communication unit in the teacher data.

6. The automatic cooking device according to claim 1 , wherein the cooking control unit controls at least one of the temperature of the pot unit and the operation of the cooking utensil unit according to color components based on the Maillard reaction obtained from an image of the ingredients in the pot unit during cooking.

7. The cooking utensil section includes: A stirring part rotatably provided in the pot part; a rotatable lid portion disposed above the stirring portion in the pot portion; 2. The automatic cooking apparatus according to claim 1, further comprising: a clutch provided between the lid portion and the shaft portion, for interrupting transmission of rotational force of the shaft portion to the lid portion.

8. the main body or the terminal device is connected via a network to a database server that stores type and amount data indicating correspondence between the images of the ingredients and the types and amounts of the ingredients, the training data, and menu data that is data on the menu; the ingredient calculation unit calculates the type and amount of the ingredient from the image of the ingredient using the type and amount data stored in the database server; The automatic cooking apparatus according to claim 1 , wherein the menu generating unit generates the menu using the menu data stored in the database server.

9. the menu data includes data on a cooking procedure for cooking based on the menu; The automatic cooking apparatus according to claim 8 , wherein the cooking control unit controls at least one of the operation of the cooking utensil unit and the temperature of the pot unit using the cooking procedure data stored in the database server.

10. The cooking method further includes a umami component detection unit that detects the amount of umami components, which are at least any one of glutamic acid, guanylic acid, and inosinic acid, from the food material during cooking, the database server stores ingredient umami component data indicating a correspondence between the types and amounts of the ingredients before cooking and the umami components, and menu umami data indicating a correspondence between the menu and the umami components; The automatic cooking device of claim 8, wherein the cooking control unit controls at least one of the operation of the cooking utensil unit and the temperature of the pot unit so that the amount of umami components detected by the umami component detection unit approaches the amount of umami components corresponding to the menu using the ingredient umami data and the menu umami data stored in the database server.

11. The cooking method further includes a umami component detection unit that detects the amount of umami components, which are at least any one of glutamic acid, guanylic acid, and inosinic acid, from the food material during cooking, the database server stores ingredient umami component data indicating a correspondence between the types and amounts of the ingredients before cooking and the umami components, and menu umami data indicating a correspondence between the menu and the umami components; The automatic cooking device of claim 8, wherein the cooking control unit controls at least one of the operation of the cooking utensil unit and the temperature of the pot unit so that the amount of umami components detected by the umami component detection unit is greater or less than the amount of umami components corresponding to the menu using the ingredient umami data and the menu umami data stored in the database server.

12. The cooking method further includes a umami component detection unit that detects the amount of umami components, which are at least any one of glutamic acid, guanylic acid, and inosinic acid, from the food material during cooking, The automatic cooking device of claim 8, wherein the cooking control unit controls at least one of the operation of the cooking utensil unit and the temperature of the pot unit so that the amount of umami components detected by the umami component detection unit becomes the type and amount of umami components determined by artificial intelligence based on preset teacher data of umami components.

Images