Artificial intelligence coffee blending system and method

Abstract

An artificial intelligence coffee blending system is provided. An artificial intelligence coffee blending system according to one aspect of the present invention comprises: a support body that supports a plurality of cylinders containing a plurality of different coffee beans; an input terminal that receives coffee characteristics preferred by a user as taste information; a first controller that generates coffee bean combinations based on coffee bean data regarding the plurality of coffee beans, and generates output information regarding a coffee bean combination having characteristics corresponding to the taste information among the generated coffee bean combinations; and a robot arm that brings the relevant cylinders from the support body to a blending container according to the output information, pours the coffee beans contained in the relevant cylinders into the blending container, and blends them.

Description

Technology Field

[0001] The present invention relates to an artificial intelligence coffee blending system and method. Background Technology

[0002] This invention was developed with the support of the Food Tech Research and Development Project of the Gyeongsangbuk-do Provincial Government and Pohang City Hall. (GBTP-2023-129001)

[0003] The present invention was developed with funding from the Ministry of Agriculture, Food and Rural Affairs and supported by the High Value-Added Food Technology Development Project of the Korea Institute of Planning and Evaluation for Agricultural and Food Technology, the Deep Learning-Based Food Cooking Robot System Development Project (RS-2024-00403998), and the Agri-Food Science and Technology Convergence Research Personnel Training Project (RS-2024-00402136).

[0004] Recently, coffee has been gaining popularity among consumers. In line with this trend, major coffee franchises and manufacturers are devoting significant effort to developing coffee that appeals to consumers.

[0005] Traditionally, coffee franchises and manufacturers have been developing new coffees by employing expert groups, such as perfumers, to sample various beans.

[0006] However, this method has the problem that the range of bean combinations that a perfumer can combine is very limited, and since blending is done manually, there is a discrepancy between the combination intended for sampling and the actual result. The problem to be solved

[0007] The present invention aims to solve the aforementioned problems, and the objective of the present invention is to provide an artificial intelligence coffee blending system and method configured to perform various combinations and reduce blending errors.

[0008] The problems of the present invention are not limited to those mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art to which the present invention pertains from the description below. means of solving the problem

[0009] According to one aspect of the present invention, an artificial intelligence coffee blending system is provided, comprising: a support member supporting a plurality of cylinders containing a plurality of different coffee beans; an input terminal receiving a coffee characteristic preferred by a user as preference information; a first controller generating a coffee bean combination based on coffee bean data regarding the plurality of coffee beans, and generating output information regarding a coffee bean combination having a characteristic corresponding to the preference information among the generated coffee bean combinations; and a robot arm that brings the relevant cylinders from the support member to a blending container according to the output information, pours the coffee beans contained in the relevant cylinders into the blending container, and blends them.

[0010] At this time, the taste information includes three or more first characteristic items among the characteristic items and scores for the first characteristic items, and the characteristic items may include acidity, sweetness, body, balance, aftertaste, nose, aroma (fruit, nut, vegetable and various natural or artificial flavors), and roasting point.

[0011] At this time, the above coffee bean data may include the ID of each of the plurality of coffee beans, the characteristic items, and the scores of the characteristic items.

[0012] At this time, the first controller derives a coffee bean combination in which the sum of the score differences between the first characteristic items constituting the taste information and the second characteristic items of the coffee bean combination generated based on the coffee bean data is minimized, selects the derived coffee bean combination as a coffee bean combination having characteristics corresponding to the taste information, generates output information regarding the selected coffee bean combination, and the second characteristic items may be characteristic items corresponding to the first characteristic items.

[0013] At this time, the first controller can derive a coffee bean combination in which the total sum of the score differences between the first characteristic items and the second characteristic items is minimized based on a machine learning model.

[0014] Meanwhile, the above output information may include the IDs and blending amounts of beans constituting a bean combination having characteristics corresponding to the above taste information.

[0015] Meanwhile, the first controller can control the blending operation of the robot arm based on the output information and previously stored basic robot operation data.

[0016] At this time, the robot motion basic data includes position data of the plurality of cylinders, movement path data for each combination of the plurality of cylinders, coffee bean discharge data for each inclination of the plurality of cylinders, and position information of the blending container, and the output information includes the IDs and blending amounts of coffee beans constituting a coffee bean combination having characteristics corresponding to the taste information, and the first controller derives position information of related cylinders containing coffee beans constituting a coffee bean combination having characteristics corresponding to the taste information from the position data, derives movement path information of a combination composed of the related cylinders from the movement path data, derives inclination information of the related cylinders from the coffee bean discharge data based on the blending amounts of coffee beans constituting a coffee bean combination having characteristics corresponding to the taste information, and the first controller can control the blending operation of the robot arm based on the position information of the related cylinders, movement path information of a combination composed of the related cylinders, inclination information of the related cylinders, and position information of the blending container.

[0017] Meanwhile, the artificial intelligence coffee blending system may further include a second controller that controls the blending operation of the robot arm based on the output information and previously stored basic robot operation data.

[0018] Meanwhile, the first controller can receive user feedback information regarding the blending result from the input terminal and use it to improve the accuracy of the machine learning model.

[0019] Meanwhile, the support member may include a plurality of upper plates spaced apart vertically and supported by a plurality of dispersed cylinders.

[0020] Seating grooves for seating the plurality of cylinders may be formed in a distributed manner on the plurality of upper plates.

[0021] At this time, the robot arm is positioned in front of the plurality of upper plates, and the plurality of upper plates can be arranged in a step structure that rises in a direction away from the robot arm.

[0022] Meanwhile, the plurality of upper plates are provided with a concave arc portion in a direction away from the robot arm, and the arc portion of the upper plate located relatively higher may have a larger radius of curvature than the arc portion of the upper plate located relatively lower.

[0023] Meanwhile, the robot arm is positioned in front of the plurality of upper plates, and the plurality of upper plates have the same shape and can be arranged in an upward and downward alignment.

[0024] Meanwhile, the support body may include a disc supporting the plurality of cylinders; and legs supporting the disc.

[0025] Meanwhile, the support body has a hollow cylindrical shape extending vertically, and a plurality of support grooves are formed on the side wall of the support body to support the plurality of cylinders, each cylinder is supported on the bottom surface of each support groove, and the robot arm can be positioned inside the support body.

[0026] At this time, the plurality of support grooves can be aligned in the circumferential direction and the vertical direction of the support body and arranged at predetermined intervals.

[0027] Meanwhile, the support body has a hollow spherical shape, and a plurality of support grooves are formed in the inner wall of the support body to support the plurality of cylinders, each cylinder is supported on the bottom surface of each support groove, and the robot arm can be disposed inside the support body.

[0028] Meanwhile, the support member includes a plurality of support plates that each support the plurality of cylinders, and the plurality of support plates can be suspended and supported on the ceiling surface of the space where the robot arm is installed.

[0029] Meanwhile, according to another aspect of the present invention, an artificial intelligence coffee blending method is provided, comprising the steps of: receiving coffee characteristics preferred by a user as preference information; generating output information regarding a coffee bean combination corresponding to the preference information based on coffee bean data regarding a plurality of coffee beans; and blending the relevant coffee beans according to the output information. Effects of the invention

[0030] According to the above configuration, the artificial intelligence coffee blending system according to one aspect of the present invention generates a coffee bean combination based on coffee bean data to derive a coffee bean combination having characteristics corresponding to the first controller preference information, and thus can generate and sample various coffee bean combinations that significantly exceed the range that a user, such as a perfumer, can attempt.

[0031] Furthermore, the robotic arm performs the blending operation based on the output information of the AI ​​coffee blending system, thereby accurately implementing the coffee bean combination preferred by the user.

[0032] The effects of the present invention are not limited to the effects described above, and should be understood to include all effects that can be inferred from the configuration of the invention described in the detailed description or claims of the present invention. Brief explanation of the drawing

[0033] FIG. 1 is a perspective view of an artificial intelligence coffee blending system according to one embodiment of the present invention, viewed from one direction. Figure 2 is an enlarged view of part A of Figure 1. FIG. 3 is a top view of an artificial intelligence coffee blending system according to one embodiment of the present invention. FIG. 4 is a perspective view of a robot arm of an artificial intelligence coffee blending system according to one embodiment of the present invention. FIG. 5 is a drawing showing a support according to another embodiment of the present invention. FIG. 6 is a drawing showing a support according to another embodiment of the present invention. FIG. 7 is a drawing showing a support according to another embodiment of the present invention. FIG. 8 is a drawing showing a support according to another embodiment of the present invention. FIG. 9 is a drawing showing a support according to another embodiment of the present invention. FIG. 10 is a flowchart of an artificial intelligence coffee blending method according to one embodiment of the present invention. Specific details for implementing the invention

[0034] Hereinafter, embodiments of the present invention are described in detail with reference to the attached drawings so that those skilled in the art can easily implement the invention. The present invention may be embodied in various different forms and is not limited to the embodiments described herein. To clearly explain the present invention, parts unrelated to the description in the drawings have been omitted, and the same reference numerals have been used throughout the specification for identical or similar components.

[0035] The words and terms used in this specification and claims are not limited to their ordinary or dictionary meanings, but should be interpreted in a meaning and concept consistent with the technical spirit of the invention in accordance with the principles by which the inventor defines terms and concepts to best describe his invention.

[0036] Therefore, the embodiments described in this specification and the configurations illustrated in the drawings correspond to preferred embodiments of the present invention and do not represent all technical concepts of the present invention; thus, various equivalents and modifications that may replace such configurations may exist at the time of filing the present invention.

[0037] In this specification, terms such as “comprising” or “having” are intended to describe the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should not be understood as precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0038] The statement that a component is "in front," "rear," "upper," or "lower" of another component includes, unless there are special circumstances, not only being positioned "in front," "rear," "upper," or "lower" in direct contact with the other component, but also cases where another component is positioned in between. Furthermore, the statement that a component is "connected" to another component includes, unless there are special circumstances, not only being directly connected to each other, but also being indirectly connected to each other.

[0039] FIG. 1 is a perspective view of an artificial intelligence coffee blending system according to one embodiment of the present invention viewed from one direction, FIG. 2 is an enlarged view of part A of FIG. 1, FIG. 3 is a top view of an artificial intelligence coffee blending system according to one embodiment of the present invention, and FIG. 4 is a perspective view of a robot arm according to one embodiment of the present invention.

[0040] Referring to FIGS. 1 to 4, the artificial intelligence coffee blending system (10) includes a support (200), an input terminal (300), a first controller (400), and a robot arm (600).

[0041] For reference, in FIGS. 1 to 3, numbers were assigned only to the lowest plate (210) and the highest plate (210) among the plurality of plates (210) for convenience. Also, for convenience, numbers were assigned only to one cylinder (100) supported by the lowest plate (210) and one cylinder (100) supported by the highest plate (210) among the plurality of cylinders (100).

[0042] The support body (200) supports multiple cylinders (100) containing multiple different coffee beans.

[0043] Multiple coffee beans possess different characteristics. Here, characteristics are a concept used to distinguish one bean or coffee from another. The characteristics of a bean or coffee can be defined by a combination of characteristic items and their scores.

[0044] The fact that the characteristics of the coffee beans differ means that at least one score among the characteristic items differs.

[0045] For example, the characteristic items that constitute the characteristics of coffee beans may include acidity, sweetness, body, balance, aftertaste, nose, aroma, and roasting point. These characteristic items can be expressed as scores, allowing for quantitative comparison.

[0046] However, the characteristic items mentioned above as examples are merely a part of the generally used characteristic items, and there are various other characteristic items that have not been mentioned.

[0047] The characteristics of coffee beans may vary depending on the origin and roasting of the beans.

[0048] Multiple different coffee beans are each placed in multiple cylinders (100). The multiple cylinders (100) may have the same shape. The multiple coffee beans are each placed in the multiple cylinders (100) in a ground state.

[0049] For example, the cylinder (100) may include a cylinder body that receives coffee beans and a cylinder lid that closes the upper opening of the cylinder body. In this case, the robot arm (600) described later can discharge the coffee beans received in the cylinder body to the outside by opening the cylinder lid and tilting the cylinder body.

[0050] As another example, the cylinder (100) may consist only of a cylinder body without a lid, with an opening formed on one side for putting in or taking out coffee beans.

[0051] The support body (200) supports a plurality of cylinders (100) in a distributed manner. At this time, each cylinder (100) is positioned at a specific location on the support body (200). The position information of each cylinder (100) is stored in advance in the form of three-dimensional coordinates and can be used as basic data for controlling the blending operation of the robot arm (600) described later. The position information of the cylinder (100) may be defined as the inner center of the cylinder (100), but is not limited thereto.

[0052] Each cylinder (100) may be assigned an ID. Through the ID, the cylinder (100) can be distinguished from other cylinders (100). Each coffee bean contained in each cylinder (100) may also be assigned an ID. The ID of each cylinder (100) corresponds to the ID of each coffee bean.

[0053] In this embodiment, the support (200) includes a plurality of upper plates (210) spaced apart vertically. The plurality of upper plates (210) are spaced apart parallel to each other. At this time, the plurality of upper plates (210) may be spaced apart horizontally.

[0054] A plurality of top plates (210) are supported by a support member (220).

[0055] For example, the support member (220) may have a bar shape that extends vertically. The support member (220) may support the corners of a plurality of top plates (210).

[0056] However, the support member (220) is not limited to the above example and may have various shapes to support a plurality of upper plates (210) spaced apart vertically.

[0057] Multiple cylinders (100) are distributed and supported on multiple top plates (210).

[0058] A plurality of mounting grooves (211) as shown in FIG. 3, on which a plurality of cylinders (100) are seated, are formed in a distributed manner on a plurality of top plates (210). A plurality of cylinders (100) can be seated in a distributed manner in the plurality of mounting grooves (211). In this case, the cylinders (100) seated in the mounting grooves (211) can be stably supported on the top plates (210).

[0059] A robot arm (600), described later, is positioned in front of a plurality of top plates (210). At this time, the plurality of top plates (210) may be arranged in a stepped structure that rises in a direction away from the robot arm (600). In other words, the front edge area of ​​each top plate (210) forms a stepped structure.

[0060] At this time, multiple cylinders (100) may be distributed in the front edge area of ​​each top plate (210). In this case, compared to the case where multiple top plates (210) are simply arranged vertically rather than in a stepped shape, the movement path of the robot arm (600) moving between the lower top plate (210) and the upper top plate (210) can be shortened and diversified.

[0061] Each of the multiple top plates (210) may have a concave arc portion (213) on one side that is concave in a direction away from the robot arm (600). The arc portion (213) of the top plate (210) located relatively higher may have a larger radius of curvature than the arc portion (213) of the top plate (210) located relatively lower. The arc centers of the arc portions (213) of the multiple top plates (210) may be arranged on a coaxial line extending vertically.

[0062] At this time, the front edge region of each top plate (210) may have an arc shape. At this time, a plurality of cylinders (100) may be distributed in the front edge region of each top plate (210) having an arc shape.

[0063] The robot arm (600) can be positioned to face the arc portion (213) of the plurality of top plates (210).

[0064] At this time, although not clearly illustrated in FIGS. 1 to 3, the base (610) of the robot arm (600) may be positioned at the center of the arc of the arc portion (213) of the plurality of top plates (210). In this case, the standard deviation between the base (610) of the robot arm (600) and the plurality of cylinders (100) supported on each top plate (210) is reduced, so that the movement path of the robot arm (600) can be effectively shortened.

[0065] The input terminal (300) is configured to receive preference information regarding the characteristics of the coffee preferred by the user. The input terminal (300) includes a kiosk, tablet PC, etc., with an app or program installed to receive preference information.

[0066] The input terminal (300) may be placed on a table (710) positioned in front of the support body (200), but is not limited thereto.

[0067] The input terminal (300) has various types of user interfaces, such as a voice user interface or a text user interface, and the user inputs preference information through the user interface.

[0068] Taste information includes three or more first characteristic items among the characteristic items and the scores of the first characteristic items. The characteristic items include acidity, sweetness, body, balance, aftertaste, nose, aroma, and roasting point.

[0069] For example, the user can enter scores of 4, 5, 3, and 3 for acidity, sweetness, body, and balance among the characteristic items. In this case, the coffee preferred by the user has characteristics of acidity 4, sweetness 5, body 3, and balance 3.

[0070] As another example, the user can enter scores of 3, 3, 1, 5, and 4 for acidity, sweetness, balance, nose, and aroma among the characteristic items. In this case, the coffee preferred by the user has the characteristics of acidity 3, sweetness 3, balance 1, nose 5, and aroma 4.

[0071] The first controller (400) generates a coffee bean combination based on coffee bean data regarding a plurality of coffee beans, and generates output information regarding a coffee bean combination having characteristics corresponding to taste information among the generated coffee bean combinations.

[0072] In other words, the first controller (400) generates a coffee bean combination based on coffee bean data regarding a plurality of coffee beans, and generates output information regarding the coffee bean combination that has characteristics most similar to the taste information among the generated coffee bean combinations.

[0073] At this time, the number of coffee beans constituting the coffee bean combination can be set to three or more. The number of coffee beans constituting the coffee bean combination can be set by the user inputting it along with preference information through the input terminal (300), or it can be pre-set by the system administrator.

[0074] For example, the first controller (400) may be configured to include a communication unit (not shown), a storage unit (not shown), and a processor (not shown).

[0075] The communication unit receives preference information from the input terminal (300). The communication unit receives and transmits information or signals using a known wired or wireless communication method.

[0076] The storage unit stores coffee bean data regarding a plurality of coffee beans. The storage unit includes a known storage medium that stores software or data, such as RAM, HDD, ROM, etc.

[0077] The processor generates coffee bean combinations based on coffee bean data, and generates output information regarding coffee bean combinations among the generated combinations that have characteristics corresponding to taste information.

[0078] The coffee bean data includes the ID, characteristic items, and scores of each of the multiple coffee beans supported on the support (200). The characteristic items of each coffee bean include acidity, sweetness, body, balance, aftertaste, nose, aroma, roasting point, etc.

[0079] For example, coffee bean 1 may have acidity, sweetness, body, balance, aftertaste, nose, aroma, and roasting point of 2, 3, 5, 3, 1, 3, 3, and 2, respectively, and coffee bean n may have acidity, sweetness, body, balance, aftertaste, nose, aroma, and roasting point of 3, 3, 4, 2, 1, 4, 2, and 3, respectively. The characteristic items and scores of the coffee bean with ID 1 and the characteristic items and scores of the coffee bean with ID n constitute the coffee bean data.

[0080] The above coffee bean data can be stored in advance in the storage unit of the first controller (400).

[0081] Specifically, the first controller (400) derives a coffee bean combination in which the sum of the score differences between the first characteristic items constituting taste information and the second characteristic items of the coffee bean combination generated based on the coffee bean data is minimized. Here, the second characteristic items are characteristic items corresponding to the first characteristic items.

[0082] In other words, the first controller (400) derives a coffee bean combination in which the sum of the differences between the scores for each of the first characteristic items constituting taste information and the scores for each of the second characteristic items of the coffee bean combination generated based on the coffee bean data is minimized.

[0083] In other words, the first controller (400) derives a coffee bean combination in which d of the following formula (1) is minimized among the coffee bean combinations generated based on the coffee bean data.

[0084] - Formula (1)

[0085] In formula (1), d is the sum of the score differences between the first characteristic items and the second characteristic items, and x i is the score of the i-th first characteristic item constituting the preference information, y i is the score of the i-th second characteristic item of the coffee bean combination, and n is the number of first characteristic items constituting the taste information.

[0086] For example, the first characteristic items constituting taste information are acidity, sweetness, body, balance, and aroma, and their scores may be 2.0, 4.0, 5.0, 3.0, and 4.0.

[0087] And the second characteristic items of a coffee bean combination generated from the coffee bean data are acidity, sweetness, body, balance, and aroma corresponding to the first characteristic items, and their scores may be 2.5, 4.0, 5.0, 3.0, and 5.0.

[0088] In this case, if the score for the first characteristic item and the score for the second characteristic item are substituted into formula (1), It becomes. Here, d for a certain coffee bean combination becomes 1.12.

[0089] The first controller (400) derives the coffee bean combination with the smallest d among the coffee bean combinations generated from the coffee bean data in the manner described above.

[0090] In formula (1), the above d represents a type of similarity, and when d is 0, the characteristics of the taste information and the characteristics of the coffee bean combination are identical, and when d is minimum, the characteristics of the coffee bean combination and the characteristics of the taste information can be seen as having the highest similarity.

[0091] The scores of the second characteristic items of the coffee bean combination in formula (1) may be weighted average scores. In this case, the scores for each second characteristic item of the coffee bean combination are derived based on the blending ratio in which weights are assigned to the coffee beans constituting the coffee bean combination. The weights may be assigned arbitrarily in the machine learning model.

[0092] For example, if weights of 1.0:1.5:2.0 are assigned to the three beans constituting the bean combination, a weighted average score for each second characteristic item of the bean combination can be derived based on this. In this case, the weights of the three beans become the blending ratio.

[0093] However, the score for each second characteristic item of the coffee bean combination in formula (1) may be a general average score. In this case, the mixing ratio of the coffee beans constituting the coffee bean combination becomes 1:1:1.

[0094] In one embodiment of the present invention, the first controller can derive a coffee bean combination in which the total sum of the score differences between the first characteristic items and the second characteristic items is minimized based on an artificial intelligence machine learning model.

[0095] At this time, the artificial intelligence machine learning model is trained using machine learning algorithms such as linear regression, random forest, or gradient boosting.

[0096] Specifically, the machine learning model is pre-trained to derive a coffee bean combination that minimizes the sum of the score differences between the first characteristic items constituting taste information and the second characteristic items of the coffee bean combination generated based on the coffee bean data.

[0097] In this way, when the first controller derives a coffee bean combination in which the sum of the score differences between the first characteristic items constituting taste information based on a machine learning model and the second characteristic items of the coffee bean combination generated based on coffee bean data is minimized, rapid, accurate, and efficient data processing is possible.

[0098] The first controller (400) generates output information regarding a coffee bean combination having characteristics corresponding to taste information. The output information includes the IDs and blending amounts of the coffee beans constituting the coffee bean combination having characteristics corresponding to taste information.

[0099] In this regard, if a combination of coffee beans with characteristics corresponding to taste information is derived, the IDs and blending ratios of the beans constituting the combination can be derived.

[0100] The blending amount of coffee beans constituting a coffee bean combination with characteristics corresponding to preference information can be derived based on the blending ratio of the beans constituting the coffee bean combination with characteristics corresponding to preference information and a pre-set total blending amount. Here, the total blending amount can be set by the user inputting it along with preference information through an input terminal, or it can be pre-set by a system administrator.

[0101] For example, if the beans constituting a combination of beans having characteristics corresponding to taste information are bean number 3, bean number 5, and bean number 11, and their blending ratio is 1.0;2.0;1.0, and the preset total blending amount is 40g, then the first controller (400) sets the blending amounts of bean number 3, bean number 5, and bean number 11 to 10g each ( ), 20g( ), 10g( Derived as ).

[0102] The robot arm (600) takes the relevant cylinders from the support (200) to the blending container (730) according to the output information and pours the beans contained in the relevant cylinders into the blending container (730) to blend.

[0103] Here, the related cylinders refer to cylinders containing coffee beans related to the output information. In other words, the related cylinders refer to cylinders containing coffee beans that constitute a combination of beans with characteristics corresponding to the preference information.

[0104] The blending container (730) can be placed on a table (710) on which the input terminal (300) is supported. The blending container (730) is supported on the table (710) through a separate support member. The blending container (730) is stored in advance in the form of three-dimensional coordinates and is used as basic data to control the blending operation of the robot arm (600).

[0105] The blending container (730) can be a dripper.

[0106] The robot arm (600) may include a base (610), a plurality of arms (620) extending from the base (610) and rotatably coupled to each other, and an end effector (630) coupled to an end arm (620a). The end effector (630) includes a gripper that grips a cylinder (100) containing coffee beans.

[0107] The robot arm (600) may be supported on a table (710) on which the input terminal (300) is supported, but is not limited thereto.

[0108] The joint portion of the multiple arms (620) and the joint member of the end arm (620a) and the end effector (630) become joints of the robot arm (600), and the rotational movements of the multiple arms (620) are combined to perform robot movements. The multiple arms (620) can rotate by means of actuators such as motors.

[0109] This robot arm (600) is a known multi-joint robot that operates to perform movements to perform a specific mission.

[0110] In one embodiment of the present invention, the robot arm (600) performs a blending operation. This means a series of operations in which the robot arm (600) grasps a cylinder (100) associated with the coffee beans to be blended, moves the grasped cylinder (100) toward a blending container (730), tilts the grasped cylinder (100) to pour the coffee beans contained in the cylinder (100) into the blending container (730), returns the grasped cylinder (100) to a predetermined position on a support (200), and releases the grasping of the cylinder (100). This series of operations is applied sequentially to all cylinders (100) containing the coffee beans to be blended.

[0111] In one embodiment of the present invention, the first controller (400) can control the blending operation of the robot arm (600) based on output information and previously stored basic robot motion data. The basic robot motion data may be stored in advance in the storage unit of the first controller (400).

[0112] The robot motion basic data includes position data of a plurality of cylinders (100), movement path data for each combination of the plurality of cylinders (100), coffee bean discharge data for each inclination of the plurality of cylinders (100), and position information of the blending container (730).

[0113] The movement path data for each combination of multiple cylinders (100) per cylinder combination is for cylinder combinations that can be generated from multiple cylinders (100) and includes a movement path having the optimal sequence number of the cylinders constituting the cylinder combinations. The movement path having the optimal sequence number has the shortest movement path length.

[0114] The output information includes the IDs and blending amounts of the beans constituting a bean combination having characteristics corresponding to the taste information, as previously explained.

[0115] The first controller (400) derives location information of related cylinders containing beans that constitute a combination of beans having characteristics corresponding to taste information from location data, derives movement information of a combination composed of said related cylinders from movement data, and derives slope information of said related cylinders from said bean discharge data based on the blending amount of beans that constitute a combination of beans having characteristics corresponding to said taste information.

[0116] The first controller (400) can control the blending operation of the robot arm based on the position information of the related cylinders, the movement path information of the combination composed of the related cylinders, the inclination information of the related cylinders, and the position information of the blending container.

[0117] The first controller (400) generates a blending operation command based on the position information of the related cylinders, the movement path information of the combination composed of the related cylinders, the inclination information of the related cylinders, and the position information of the blending container, and the robot arm (600) performs a blending operation according to the blending operation command.

[0118] In one embodiment of the present invention, the first controller (400) simultaneously performed the function of generating output information and the function of controlling the blending operation of the robot arm (600).

[0119] In another embodiment of the present invention, a first controller (not shown) generates output information, and a second controller (not shown) can control the blending operation of the robot arm (600). In this case, the second controller can control the blending operation of the robot arm (600 of FIG. 1) based on the output information generated by the first controller and the previously stored robot motion basic data.

[0120] In one embodiment of the present invention, the relevant coffee beans are blended in the blending container (730) due to the blending operation of the robot arm (600).

[0121] For example, the user places a cup (760) under a blending container (730) which is a dripper, and pours hot water into the blending container (730) to drip coffee into the cup (760). Afterward, the user can taste and evaluate the coffee contained in the cup (760).

[0122] As another example, the first controller (400) can control a dripping operation in addition to the blending operation of the robot arm (600). The dripping operation is the operation of grasping a kettle (not shown) containing hot water and pouring it into a blending container (730), and the first controller (400) can control the dripping operation of the robot arm (600) based on the position information of the kettle, etc.

[0123] Coffee is dripped into a cup (760) located below the blending container (730) by the dripping action of the robot arm (600). Afterwards, the coffee in the cup (760) can be tasted and evaluated.

[0124] In one embodiment of the present invention, the first controller (400) can receive user feedback information regarding the blending result from the input terminal (300) and use it to improve the accuracy of the machine learning model.

[0125] The artificial intelligence coffee blending system (10) according to one embodiment of the present invention described above can generate and sample various coffee bean combinations that significantly exceed the range that a user, such as a perfumer, can attempt, because the first controller (400) generates coffee bean combinations based on coffee bean data to derive coffee bean combinations having characteristics corresponding to taste information.

[0126] Furthermore, the artificial intelligence coffee blending system (10) can accurately implement the coffee bean combination preferred by the user by having the robot arm perform the blending operation according to the output information.

[0127] FIG. 5 is a drawing showing a support according to another embodiment of the present invention.

[0128] Referring to FIG. 5, the support body (200-1) includes a plurality of upper plates (210) spaced apart vertically. The plurality of upper plates (210) are spaced apart parallel to each other. At this time, the plurality of upper plates (210) may be spaced apart horizontally. A plurality of cylinders (100) are distributed on the plurality of upper plates (210).

[0129] Multiple top plates (210) have the same shape and are arranged vertically. For example, the multiple top plates (210) may have a rectangular plate shape extending in the horizontal direction, but are not limited thereto.

[0130] A plurality of top plates (210) are supported by a support member (220). For example, the support member (220) may have a plate shape that extends vertically. The support member (220) may support both sides of the extension direction of the plurality of top plates (210).

[0131] A robot arm (not shown) may be positioned in front of a plurality of top plates (210).

[0132] FIG. 6 is a drawing showing a support according to another embodiment of the present invention. Referring to FIG. 6, the support (200-2) includes a disc (230) that supports the plurality of cylinders (100) and a leg (231) that supports the disc (230).

[0133] A plurality of cylinders (100) may be aligned and spaced apart in the circumferential and radial directions on the upper surface of the disc (230). A robot arm (not shown) may be positioned at the center of the disc (230). In this case, the standard deviation between the robot arm and the plurality of cylinders (100) is reduced, so that the movement path of the robot arm can be effectively shortened. In addition to the robot arm, a blending container (not shown) may be positioned in the central area of ​​the disc (230).

[0134] FIG. 7 is a drawing showing a support according to another embodiment of the present invention.

[0135] Referring to FIG. 7, the support (200-3) has a hollow cylindrical shape that extends vertically.

[0136] The support body (200-3) may have an open upper side as shown in FIG. 7, but may also be closed.

[0137] A plurality of support grooves (240) are formed on the side wall of the support body (200-3) to support a plurality of cylinders (100). The plurality of support grooves (240) may be formed in a shape that penetrates the inner and outer sides of the support body (200) as shown in FIG. 7. Alternatively, although not illustrated, the plurality of support grooves may be formed in a concave shape only on the inner surface of the support body (200).

[0138] Multiple cylinders (100) can be supported on the bottom surface of multiple support grooves (240).

[0139] A plurality of support grooves (240) are aligned in the circumferential direction and the vertical direction of the support body (200) and arranged at predetermined intervals. A robot arm (not shown) is positioned on the inner side of the support body (200).

[0140] In this case, the standard deviation between the robot arm and the multiple cylinders (100) is reduced, so the movement path of the robot arm can be effectively shortened.

[0141] An entrance (250) may be formed in the side wall of the support body (200). Through the entrance (250), all or part of the user's body may enter or exit the inside of the support body (200).

[0142] FIG. 8 is a drawing showing a support according to another embodiment of the present invention.

[0143] Referring to FIG. 8, the support (200-4) may have a hollow spherical shape.

[0144] The support body (200-4) may have an open upper side as shown in FIG. 8, but may also be closed.

[0145] A plurality of support grooves (240) are formed on the side wall of the support body (200-4) to support a plurality of cylinders (100). The plurality of support grooves (240) may be formed in a shape that penetrates the inner and outer sides of the support body (200-4) as shown in FIG. 8. Alternatively, although not illustrated, the plurality of support grooves may be formed in a concave shape only on the inner surface of the support body (200-4).

[0146] For reference, in FIG. 8, a groove unrelated to the use of supporting the cylinder (100) may be formed on the outer surface of the support body (200-4).

[0147] Multiple cylinders (100) can be supported on the bottom surface of multiple support grooves (240).

[0148] A plurality of support grooves (240) are aligned in the circumferential direction and the vertical direction of the support body (200-4) and arranged at predetermined intervals. A robot arm (not shown) is positioned on the inner side of the support body (200-4).

[0149] In this case, the standard deviation between the robot arm and the multiple cylinders (100) is reduced, so the movement path of the robot arm can be effectively shortened.

[0150] An entrance (250) may be formed in the side wall of the support body (200-4). Through the entrance (250), all or part of the user's body may enter or exit the inside of the support body (200).

[0151] FIG. 9 is a drawing showing a support according to another embodiment of the present invention.

[0152] Referring to FIG. 9, the support body (200-5) may include a plurality of support plates (270) that each support a plurality of cylinders (100). The support plates (270) support the bottom surface of the cylinders (100). A seating groove (not shown) on which the bottom surface of the cylinders (100) is seated may be formed on the upper surface of the support plates (270).

[0153] A plurality of support plates (270) are suspended and supported on the ceiling surface (280) of the space where the robot arm (not shown) is installed. At this time, the plurality of support plates (270) may be supported by a support rod (271) that is fixed to the ceiling surface (280) and extends downward. A support member (273) that supports the central area of ​​the cylinder (100) may be formed on one side of the support rod (271).

[0154] The support structure (200-5) configured in this way has a form that is suspended from the ceiling, thereby improving space utilization.

[0155] FIG. 10 is a flowchart of an artificial intelligence coffee blending method according to an embodiment of the present invention. Referring to FIG. 10, the artificial intelligence coffee blending method according to an embodiment of the present invention includes the step of receiving coffee characteristics preferred by a user as taste information (S110), the step of generating output information regarding a coffee bean combination corresponding to the taste information based on coffee bean data regarding a plurality of coffee beans (S130), and the step of blending the relevant coffee beans according to the output information (S150).

[0156] The above artificial intelligence coffee blending method is implemented as the aforementioned artificial intelligence coffee blending system (10), and a specific explanation thereof is replaced by the explanation of the aforementioned artificial intelligence coffee blending system (10).

[0157] Although embodiments of the present invention have been described above, the spirit of the present invention is not limited by the embodiments presented in this specification. Those skilled in the art who understand the spirit of the present invention may easily propose other embodiments within the scope of the same spirit by adding, changing, deleting, or adding components, and such are also to be considered to fall within the scope of the spirit of the present invention. Explanation of the symbols

[0158] 10 : AI Coffee Blending System 100 : Cylinder 200 : Support 210 : Top plate 211 : Seating groove 213 : Arc section 220 : Support member 400 : First controller 600 : Robot arm 610 : Base 620: Multiple cancers 620a: Distal cancer 630 : End Effector 710 : Table 730 : Blending container 760 : Cup

Claims

Claim 1 A support body that supports a plurality of cylinders containing a plurality of different coffee beans; an input terminal that receives the characteristics of coffee preferred by a user as taste information; and a first controller that generates a coffee bean combination based on coffee bean data regarding the plurality of coffee beans, and generates output information regarding a coffee bean combination having characteristics corresponding to the taste information among the generated coffee bean combinations. The system includes a robot arm that brings related cylinders from the support to a blending container according to the output information and pours the coffee beans contained in the related cylinders into the blending container to blend; the first controller controls the blending operation of the robot arm based on the output information and previously stored basic robot operation data; the basic robot operation data includes position data of the plurality of cylinders, movement path data by combination of the plurality of cylinders, coffee bean discharge data by inclination of the plurality of cylinders, and position information of the blending container; the output information includes the IDs and blending amounts of the coffee beans constituting a coffee bean combination having characteristics corresponding to the preference information; the first controller derives position information of the related cylinders containing the coffee beans constituting the coffee bean combination having characteristics corresponding to the preference information from the position data; derives movement path information of the combination composed of the related cylinders from the movement path data; and derives inclination information of the related cylinders from the coffee bean discharge data based on the blending amounts of the coffee beans constituting the coffee bean combination having characteristics corresponding to the preference information; and the first controller... An artificial intelligence coffee blending system that controls the blending operation of the robot arm based on information, tilt information of the related cylinders, and position information of the blending container. Claim 2 An artificial intelligence coffee blending system according to claim 1, wherein the taste information includes three or more first characteristic items among the characteristic items and scores for said first characteristic items, and said characteristic items include acidity, sweetness, body, balance, aftertaste, nose, aroma, and roasting point. Claim 3 In paragraph 2, the artificial intelligence coffee blending system, wherein the coffee bean data includes the ID of each of the plurality of coffee beans, the characteristic items, and the scores of the characteristic items. Claim 4 In paragraph 3, the first controller derives a coffee bean combination in which the sum of the score differences between the first characteristic items constituting the taste information and the second characteristic items of the coffee bean combination generated based on the coffee bean data is minimized, selects the derived coffee bean combination as a coffee bean combination having characteristics corresponding to the taste information, generates output information regarding the selected coffee bean combination, and the second characteristic items are characteristic items corresponding to the first characteristic items, an artificial intelligence coffee blending system. Claim 5 In paragraph 4, the first controller is an artificial intelligence coffee blending system that derives a coffee bean combination in which the sum of the score differences between the first characteristic items and the second characteristic items is minimized based on a machine learning model. Claim 6 In paragraph 4, the output information comprises the IDs and blending amounts of beans constituting a bean combination having characteristics corresponding to the taste information, in an artificial intelligence coffee blending system. Claim 7 delete Claim 8 delete Claim 9 An artificial intelligence coffee blending system according to claim 1, further comprising a second controller that controls the blending operation of the robot arm based on the output information and previously stored basic robot motion data. Claim 10 In claim 5, the first controller receives user feedback information regarding the blending result from the input terminal and uses it to improve the accuracy of the machine learning model, an artificial intelligence coffee blending system. Claim 11 An artificial intelligence coffee blending system according to claim 1, wherein the support member comprises a plurality of top plates arranged vertically spaced apart and in which the plurality of cylinders are dispersed and supported. Claim 12 An artificial intelligence coffee blending system according to claim 11, wherein the plurality of upper plates have mounting grooves formed in a dispersed manner for mounting the plurality of cylinders. Claim 13 An artificial intelligence coffee blending system according to claim 11, wherein the robot arm is positioned in front of the plurality of top plates, and the plurality of top plates are arranged in a stepped structure that rises in a direction away from the robot arm. Claim 14 An artificial intelligence coffee blending system according to claim 13, wherein the plurality of upper plates are provided with a concave arc portion in a direction away from the robot arm, and the arc portion of the upper plate located relatively higher has a larger radius of curvature than the arc portion of the upper plate located relatively lower. Claim 15 An artificial intelligence coffee blending system according to claim 11, wherein the robot arm is positioned in front of the plurality of top plates, and the plurality of top plates have the same shape and are arranged in an up-and-down alignment. Claim 16 An artificial intelligence coffee blending system according to claim 1, wherein the support comprises a disc supporting the plurality of cylinders; and a leg supporting the disc. Claim 17 An artificial intelligence coffee blending system according to claim 1, wherein the support body has a hollow cylindrical shape extending vertically, a plurality of support grooves are formed in the side wall of the support body to support the plurality of cylinders, each cylinder is supported on the bottom surface of each support groove, and the robot arm is disposed inside the support body. Claim 18 An artificial intelligence coffee blending system according to claim 17, wherein the plurality of support grooves are aligned in the circumferential direction and the vertical direction of the support body and are arranged at predetermined intervals. Claim 19 In claim 1, the support body has a hollow spherical shape, a plurality of support grooves are formed in the inner wall of the support body to support the plurality of cylinders, each cylinder is supported on the bottom surface of each support groove, and the robot arm is disposed inside the support body, an artificial intelligence coffee blending system. Claim 20 An artificial intelligence coffee blending system according to claim 1, wherein the support body comprises a plurality of support plates each supporting the plurality of cylinders, and the plurality of support plates are suspended and supported on the ceiling surface of the space where the robot arm is installed. Claim 21 A method for blending coffee using an artificial intelligence coffee blending system according to any one of claims 1 to 6 and 9 to 20, comprising: receiving coffee characteristics preferred by a user as preference information; generating output information regarding a combination of coffee beans corresponding to the preference information based on coffee bean data regarding a plurality of coffee beans; and blending the relevant coffee beans according to the output information.

Images