Customized drip coffee manufacturing device based on robot arm and drip coffee manufacturing system using the same

Abstract

A robot arm-based customized drip coffee manufacturing device according to an embodiment of the present invention comprises: a shelf unit having a plurality of dosing cups arranged therein for storing different types of coffee beans; a hand robot unit including at least one collaborative robot module for performing a drip coffee manufacturing process; a grinder unit for grinding the coffee beans stored in the dosing cups to produce ground coffee beans; a dripper unit for extracting drip coffee by passing hot water through the ground coffee beans; and a processor unit for controlling the hand robot unit to perform the drip coffee manufacturing process. The hand robot unit comprises: a grip box having a rotatable structure in which a motor is arranged; a grip including a grip upper part and a grip lower part in which the gap and angle are adjusted according to the shape of the object to be gripped; and a grip connecting part connecting the grip and the grip box. The gripper unit includes a horizontal grip rail formed in the grip box so that the grip connection part moves left and right to adjust the position of the grip; wherein at least one of the upper grip and the lower grip is structured to be capable of individual rotation and frontal protrusion according to the shape of the object to be grasped; and the processor unit controls the hand robot unit to perform at least one of a grasping operation and a pressing operation by determining the gap, rotation angle, and protrusion of the grip based on data sensing the shape of the object to be grasped. (This patent was obtained as a result of research conducted under the Food Tech Research and Technology Development Project of the Gyeongsangbuk-do Provincial Office and Pohang City Hall (GBTP2023129001))

Description

Technology Field

[0001] The present invention relates to a robot arm-based customized drip coffee manufacturing device and a drip coffee manufacturing system using the same, which uses a collaborative robot to accurately select coffee beans according to an order and extracts the selected beans at an accurate ratio and time to manufacture drip coffee through an automated process. Background Technology

[0002] The modern coffee market is seeing a significant increase in demand for diverse beans, brewing methods, and customized recipes. In particular, since the taste and aroma of drip coffee are heavily influenced by various variables such as bean characteristics, temperature, flow rate, and brewing time, meticulous management and technical expertise are required to consistently produce high-quality coffee.

[0003] Meanwhile, although there have been increasing attempts recently to automate the coffee extraction process using collaborative robots, general automation devices often only provide pre-set fixed recipes or remain at the level of simply monitoring sensor data. As a result, it is difficult to respond in real time to unexpected situations such as the roasting status of the beans, indoor ambient temperature, and minute fluctuations in the equipment, leading to variations in extraction quality.

[0004] Therefore, there is a growing need for technology that enables the more precise and intelligent operation of collaborative robot-based coffee extraction processes by combining machine learning algorithms and sensor data to control and correct extraction variables, such as temperature, flow rate, and bean expansion, in real time. Prior art literature

[0005] (Patent Document 0001) KR 102093088 B1(Patent Document 0002) KR 102214778 B1(Patent Document 0003) KR 101963652 B1 The problem to be solved

[0006] The present invention was devised to solve the problems of the prior art described above, and aims to provide a robot arm-based customized drip coffee manufacturing device and a drip coffee manufacturing system using the same, which uses a collaborative robot to accurately select coffee beans according to an order, extracts the selected beans at an accurate ratio and time, and manufactures drip coffee through an automated process.

[0007] In addition, the present invention aims to provide a robot arm-based customized drip coffee manufacturing device and a drip coffee manufacturing system using the same, which manufactures mixed drip coffee by blending two or more types of coffee beans according to consumer preferences.

[0008] In addition, the present invention aims to provide a robot arm-based customized drip coffee manufacturing device and a drip coffee manufacturing system using the same, which perform the entire coffee serving process—from dosing cup gripping, coffee bean grinding, dripping, serving drip coffee to customers, to washing used dosing cups—using multiple collaborative robots configured with an optimal movement path.

[0009] In addition, the present invention aims to provide a robot arm-based customized drip coffee manufacturing device and a drip coffee manufacturing system using the same, which provides smart dripping control technology that dynamically adjusts optimal extraction conditions for each coffee bean in real time during the drip coffee manufacturing process.

[0010] In addition, the smart dripping control technology according to one embodiment of the present invention aims to provide a robot arm-based customized drip coffee manufacturing device and a drip coffee manufacturing system using the same, which enables an efficient robotic process by automatically determining whether extraction is finished (d_flag), and by inputting sensor data such as temperature, flow rate, coffee bean expansion, and extract color collected during the extraction process into a machine learning model in real time to dynamically control and correct the nozzle (collaborative robot gripper) angle, water flow rate (valve opening), and hot water temperature (heater setpoint), thereby manufacturing uniform and high-quality drip coffee even if the type of coffee beans and environmental conditions vary.

[0011] However, the technical problems that the present invention and the embodiments of the present invention aim to solve are not limited to the technical problems described above, and other technical problems may exist. means of solving the problem

[0012] A robot arm-based customized drip coffee manufacturing device according to an embodiment of the present invention comprises: a shelf unit having a plurality of dosing cups arranged therein for storing different types of coffee beans; a hand robot unit including at least one collaborative robot module for performing a drip coffee manufacturing process; a grinder unit for grinding the coffee beans stored in the dosing cups to produce ground coffee beans; a dripper unit for extracting drip coffee by passing hot water through the ground coffee beans; and a processor unit for controlling the hand robot unit to perform the drip coffee manufacturing process. The hand robot unit comprises: a grip box having a rotatable structure in which a motor is arranged; a grip including a grip upper part and a grip lower part in which the gap and angle are adjusted according to the shape of the object to be gripped; and a grip connecting part connecting the grip and the grip box. A gripper unit comprising a horizontal grip rail formed in the grip box so that the grip connection part moves left and right to adjust the position of the grip; wherein at least one of the upper and lower grips is structured to be capable of individual rotation and frontal protrusion according to the shape of the gripping object, and the processor unit controls the hand robot unit to perform at least one of a gripping operation and a pressing operation by determining the gap, rotation angle, and protrusion of the grips based on data sensing the shape of the gripping object.

[0013] In addition, the processor unit maps a unique number to each different type of coffee bean and maps the position of each dosing cup within the shelf unit to which the unique number is mapped into position information in the form of a matrix corresponding to the unique number.

[0014] In addition, at least one collaborative robot module included in the hand robot unit is a multi-joint module structure that adjusts the angle, direction, and position of the gripper unit, is equipped with a collision detection sensor for each joint module, and is programmed with a movement algorithm to simultaneously perform different processes included in the drip coffee manufacturing process.

[0015] Additionally, the grip of the gripper part includes a grip upper portion and a grip lower portion spaced apart from each other, and the grip upper portion and the grip lower portion form a space having a predetermined curvature, and the spacing between the grip upper portion and the grip lower portion is narrowed or widened according to the shape of the object to be gripped.

[0016] Additionally, the grinder unit includes a bean receiving unit into which beans are fed, a grinder blade for grinding beans, a bean discharge port for discharging ground beans, and an operating button for grinding beans, and the processor unit controls the gripper unit so that at least one of the upper grip and lower grip included in the grip protrudes and rotates to perform a pressing operation for the operating button.

[0017] In addition, the dripper part includes a funnel-shaped dripper that secures ground coffee beans and a filter, and a support frame that secures the dripper to the top of a beverage cup.

[0018] In addition, the processor unit monitors the real-time dripping state through at least one sensor included in the sensor system, analyzes at least one of the temperature, flow rate, expansion and color data of the hot water obtained during dripping through a deep learning model that has learned the optimal extraction conditions for each type of coffee bean, and dynamically adjusts the gripper unit according to the analyzed result.

[0019] In addition, the shelf unit further includes a beverage washing unit equipped with a water spray nozzle for automatically washing a used dosing cup or beverage cup, and the beverage washing unit is characterized by automatically spraying water to wash the cup when a predetermined object is detected through an infrared sensor.

[0020] In addition, the processor unit acquires orderer preference data entered based on the kiosk, determines a first bean and a second bean corresponding to the acquired preference data among the bean information stored in memory, automatically calculates the blending ratio of the determined first bean and second bean using a blending ratio recommendation algorithm, and controls the hand robot unit to manufacture customized blended drip coffee according to the calculated blending ratio.

[0021] Meanwhile, in a smart dripping control method for dynamically controlling extraction conditions by analyzing sensor data in a robot arm-based customized drip coffee manufacturing device according to claim 1, the method comprises: 1) a step of measuring the real-time extraction temperature inside the dripper through an extraction temperature sensor and acquiring sensor data including hot water temperature, extraction speed, coffee bean expansion amount, and extraction color; 2) a step of inputting the sensor data into a machine learning model to analyze the deviation from the target extraction condition and predicting the nozzle angle correction value, the water flow speed correction value, the hot water temperature correction value, and whether the extraction is finished, respectively; 3) a step of correcting the extraction conditions in real time by adjusting the nozzle angle of the collaborative robot, the valve opening, and the water heater temperature according to the predicted correction values; and a step of stopping the extraction when the extraction finish status is above a threshold value, and repeating steps 1) to 3) at a certain interval in other cases to enable dynamic extraction control corresponding to changes in coffee bean type and environment.

[0022] In addition, the machine learning model receives sensor data and adopts a machine learning-based multi-output model to simultaneously predict nozzle angle correction values, water flow rate correction values, hot water temperature correction values, and whether extraction is complete, and each parameter is optimized through hyperparameter tuning.

[0023] In addition, the extraction termination status is predicted in the form of a continuous value, and extraction is determined to be stopped if it exceeds the threshold value; the allowable error for the threshold value and the deviations in extraction temperature and flow rate can be set variably.

[0024] Meanwhile, a robot arm-based customized drip coffee manufacturing system according to an embodiment of the present invention comprises: a shelf unit having a plurality of dosing cups arranged to each store different types of coffee beans; a hand robot unit including at least one collaborative robot module that performs a drip coffee manufacturing process; a grinder unit that grinds the coffee beans stored in the dosing cups to produce ground coffee beans; a dripper unit that extracts drip coffee by passing hot water through the ground coffee beans; a beverage providing unit that provides a beverage cup containing the drip coffee to the customer; and a beverage washing unit that automatically washes the used dosing cup or beverage cup. and a processor unit that controls the hand robot unit to perform the drip coffee manufacturing process; wherein the processor unit acquires order information through orderer input detected by an interface module, determines a target dosing cup containing the corresponding beans according to the bean information included in the acquired order information, controls the hand robot unit to move to and grasp the determined target dosing cup, controls the hand robot unit to move the grasped target dosing cup to the grinder unit to drop and grind a preset amount of beans, controls the hand robot unit to move the target dosing cup containing the ground beans after grinding to the dripper unit to perform dripping, controls the hand robot unit to provide the drip coffee completed according to the dripping to the beverage serving unit, and controls the hand robot unit to scan and grasp the used dosing cup or beverage cup through an image sensor to perform automatic washing in the beverage washing unit. Effects of the invention

[0025] The robot arm-based customized drip coffee manufacturing device and drip coffee manufacturing system using the same, according to an embodiment of the present invention, have the effect of enabling the production of various types of drip coffee with high accuracy while reducing labor costs by accurately selecting coffee beans using a collaborative robot, extracting the selected beans at an accurate ratio and time based on an optimal movement path, and manufacturing drip coffee through an automated process.

[0026] In addition, the robot arm-based customized drip coffee manufacturing device and the drip coffee manufacturing system using the same, according to an embodiment of the present invention, produce blended drip coffee by blending two or more types of coffee beans according to consumer preferences, thereby responding to the demand for a wide range of flavors and aromas of drip coffee consumers, and contributing to strengthening the competitiveness of coffee shops and increasing customer satisfaction.

[0027] In addition, the robot arm-based customized drip coffee manufacturing device and the drip coffee manufacturing system using the same, according to an embodiment of the present invention, perform the entire coffee serving process—from dosing cup gripping, bean grinding, dripping, serving drip coffee to customers, to washing used dosing cups—using multiple collaborative robots configured with optimal movement paths. This provides the latest technology and innovative experience through robot technology, enhances aesthetic appeal, and simultaneously guarantees the precision and speed of the manufacturing process.

[0028] In addition, the robot arm-based customized drip coffee manufacturing device and the drip coffee manufacturing system using the same according to the embodiment of the present invention provide smart dripping control technology that dynamically adjusts optimal extraction conditions for each coffee bean in real time during the drip coffee manufacturing process, thereby not only dramatically improving the consistency of coffee extraction quality but also enabling the provision of coffee flavors optimized according to the condition of the coffee beans and environmental conditions, rather than uniformly, to customers, which has the effect of greatly increasing customer satisfaction.

[0029] In addition, since the machine learning algorithm actively controls the extraction temperature, flow rate, time, etc., the coffee quality can be maintained uniformly even if the bean roasting state or surrounding environment changes.

[0030] In addition, the embodiment can produce high-quality coffee without relying on the barista's skill level, resulting in reduced labor costs and increased work efficiency.

[0031] In addition, the embodiment allows the system to be easily updated by collecting sensor logs and taste evaluation results and retraining the model, even if new coffee beans or extraction recipes are added in the future.

[0032] However, the effects obtainable from the present invention are not limited to those mentioned above, and other unmentioned effects can be clearly understood from the description below. Brief explanation of the drawing

[0033] FIG. 1a is an internal block diagram of a drip coffee robot according to an embodiment of the present invention. FIG. 1b is a flowchart of a smart dripping control technology according to an embodiment of the present invention. FIGS. 2 and FIGS. 3 are overall perspective views of a drip coffee robot according to an embodiment of the present invention. FIG. 4 is a front view of a drip coffee robot according to an embodiment of the present invention. FIG. 5 is a side view of a drip coffee robot according to an embodiment of the present invention. FIG. 6 is a plan view of a drip coffee robot according to an embodiment of the present invention. FIG. 7 is an overall perspective view of a gripper part according to an embodiment of the present invention. FIG. 8 is an enlarged view of a gripper part according to an embodiment of the present invention. FIG. 9 is a flowchart illustrating a drip coffee manufacturing process according to an embodiment of the present invention. FIG. 10 is an example of a drawing showing a plurality of dosing cups set on a shelf according to an embodiment of the present invention. FIG. 11 is an example of a drawing for explaining the types of coffee beans mapped for each coffee bean stored in each dosing cup according to an embodiment of the present invention. FIG. 12 is an example showing the gripper part of the hand robot part rotated at a predetermined angle according to an embodiment of the present invention. Specific details for implementing the invention

[0034] The present invention is capable of various modifications and may have various embodiments; therefore, specific embodiments are illustrated in the drawings and described in detail in the detailed description. The effects and features of the present invention, and the methods for achieving them, will become clear by referring to the embodiments described in detail below together with the drawings. However, the present invention is not limited to the embodiments disclosed below but can be implemented in various forms. In the following embodiments, terms such as "first," "second," etc., are used not in a limiting sense but for the purpose of distinguishing one component from another. Furthermore, singular expressions include plural expressions unless the context clearly indicates otherwise. Also, terms such as "include" or "have" mean that the features or components described in the specification exist, and do not preclude the possibility that one or more other features or components may be added. Additionally, in the drawings, the size of components may be exaggerated or reduced for convenience of explanation. For example, the size and thickness of each component shown in the drawings are arbitrarily depicted for convenience of explanation, so the present invention is not necessarily limited to what is illustrated.

[0035] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings. When describing with reference to the drawings, identical or corresponding components are given the same reference numerals, and redundant descriptions thereof will be omitted.

[0037] A drip coffee robot assembly (hereinafter referred to as a drip coffee robot) according to an embodiment of the present invention may be at least one device that performs an automated drip coffee manufacturing process at a location, such as a cafe, where drip coffee is required. Hereinafter, a person who orders drip coffee at the location may be referred to as the “orderer.” Additionally, the entire process of performing processes such as bean selection, grinding, and dripping based on the drip coffee robot may be referred to as the “drip coffee manufacturing process.”

[0038] FIG. 1a is an internal block diagram of a drip coffee robot according to an embodiment of the present invention. FIG. 1b is a flowchart of a smart dripping control technology according to an embodiment of the present invention.

[0039] Referring to FIG. 1a, from a functional perspective, the drip coffee robot (10) may include a memory (110), a processor (120), a communication processor (130), an interface module (140), an input system (150), a sensor system (160), and a display system (170). These components may be configured to be included within the housing of the drip coffee robot (10).

[0040] Specifically, in memory (110), an application is stored, and the application can store one or more of various applications, data and commands to provide a drip coffee manufacturing automation service environment.

[0041] That is, the memory (110) can store commands and data that can be used to create a drip coffee manufacturing automation service environment.

[0042] In addition, the memory (110) may include a program area and a data area.

[0043] Here, the program area according to the embodiment may be linked between the operating system (OS) and functional elements that boot the drip coffee robot (10), and the data area may store data generated according to the use of the drip coffee robot (10).

[0044] The processor (120) may include at least one processor capable of executing instructions of an application stored in memory (110) to perform various tasks for creating a drip coffee manufacturing automation service environment.

[0045] In an embodiment, the processor (120) can control the overall operation of the components through an application of the memory (110) to provide an automated drip coffee manufacturing service.

[0046] This processor (120) may be a system-on-chip (SOC) suitable for a drip coffee robot (10) including a central processing unit (CPU) and / or a graphics processing unit (GPU), etc., and may execute an operating system (OS) and / or application program, etc. stored in memory (110), and may control each component mounted on the drip coffee robot (10).

[0047] In addition, in the embodiment, the processor (120) can perform various deep learning for drip coffee manufacturing automation services by linking with a deep-learning neural network.

[0048] Here, the deep learning neural network according to the embodiment may include a Convolutional Neural Network (CNN), R-CNN (Regions with CNN features), Fast R-CNN, Faster R-CNN, Mask R-CNN, etc., and may include any deep learning neural network that includes an algorithm capable of performing the embodiments described below, and the embodiments of the present invention do not limit or restrict such deep learning neural networks themselves.

[0049] At this time, according to the embodiment, the deep learning neural network may be installed directly on the drip coffee robot (10) or operate as a separate device from the drip coffee robot (10) to perform deep learning for the drip coffee manufacturing automation service.

[0050] Specifically, the processor (120) can control the entire components of the drip coffee robot (10) by learning the data required for the drip coffee manufacturing automation service through machine learning and linking it with big data and / or AI (Artificial Intelligence).

[0051] In various embodiments, the drip coffee robot (10) may include a machine learning system (121) separately from the processor (120).

[0052] The communication processor (130) may include one or more devices for communicating with an external device. The communication processor (130) may communicate via a wireless network.

[0053] In addition, the communication processor (130) can transmit and receive various data related to the drip coffee manufacturing automation service to other terminals and / or external servers, etc.

[0054] This communication processor (130) can wirelessly transmit and receive data with at least one of a base station, an external terminal, or any server on a mobile communication network built through a communication device capable of performing technical standards or communication methods for mobile communication (e.g., LTE (Long Term Evolution), LTE-A (Long Term Evolution-Advanced), 5G NR (New Radio), WIFI) or short-range communication methods.

[0055] The interface module (140) can communicate with one or more other devices to connect the drip coffee robot (10). Specifically, the interface module (140) may include a wired and / or wireless communication device compatible with one or more different communication protocols.

[0056] Through this interface module (140), the drip coffee robot (10) can be connected to various input / output devices.

[0057] The input system (150) can detect user input related to the drip coffee manufacturing automation service (e.g., gestures, voice commands, button operation, or other types of input).

[0058] Specifically, the input system (150) may include a predetermined button, a touch sensor and / or an image sensor that receives user motion input, etc.

[0059] Additionally, the input system (150) can be connected to an external controller through an interface module (140) to receive user input.

[0060] The sensor system (160) may include various sensors such as an image sensor, a position sensor (IMU), an audio sensor, a distance sensor, a proximity sensor, a contact sensor, a load detection sensor, a weight sensor, a vision sensor, a torque sensor, and a lock sensor.

[0061] Here, the image sensor can capture images and / or videos of the physical space around the drip coffee robot (10).

[0062] In the embodiment, the image sensor can capture and acquire various images and / or videos related to the drip coffee manufacturing automation service.

[0063] Additionally, the image sensor can be positioned on the front or / and rear of the drip coffee robot (10) to capture an image of the positioned direction, and can capture a physical space through a camera positioned facing the outside of the drip coffee robot (10).

[0064] Such an image sensor may include an image sensor device and an image processing module. Specifically, the image sensor may process still images or video obtained by an image sensor device (e.g., CMOS or CCD).

[0065] In addition, the image sensor can process a still image or video acquired through the image sensor device using an image recognition process (e.g., OCR, etc.) and / or an image processing module to extract necessary information and transmit the extracted information to a processor.

[0066] Such an image sensor may be a camera assembly comprising at least one camera. The camera assembly may include a general camera that captures the visible light band, and may further include special cameras such as an infrared camera or a stereo camera.

[0067] Position sensors can measure the robot's spatial position and orientation through the joint angles of the robot arm itself, the absolute position relative to the robot base, or an IMU (Inertial Measurement Unit).

[0068] Audio sensors can detect acoustics (sound, voice) or measure specific frequencies, volumes, etc. For example, if a user issues voice commands such as starting or stopping extraction, the sensor can recognize these commands and enable the robot to execute them.

[0069] Distance sensors can detect distances in advance to prevent collisions with surrounding objects such as cups, walls, and customers when the robot moves, and can be used to determine how much coffee is in the cup (liquid level) and adjust the gripper position according to the cup height.

[0070] Proximity sensors can be used to fine-tune the robot gripper by detecting the approach of an object just before it grasps a target, and they can also be used to temporarily stop the robot when a person suddenly approaches or an unexpected object gets close.

[0071] Contact sensors can be introduced to check in real time whether the dosing cup or beverage cup has been held correctly or if it has come into contact with an unexpected object.

[0072] A load sensing sensor can be used to roughly measure the weight of the beans when the dosing cup is held to determine the amount for one serving, or to monitor the load applied to the gear or motor during the process of the grinder grinding the beans and to stop operation in the event of an abnormal overload.

[0073] Vision sensors can be applied to analyze the color of the extract falling below the dripper in real time to determine the concentration or degree of extraction of the coffee, or to use vision algorithms to calculate the layer of coffee beans expanding upon contact with hot water (blooming state) and provide feedback control of extraction conditions (flow rate, temperature, etc.).

[0074] Torque sensors are installed on the rotational axes of each joint of the robot arm to measure the torsional force of the axes. These torque sensors can sensitively detect changes in load and collisions for each joint of the robot arm.

[0075] A lock sensor can be placed at each joint of the robot arm to detect whether the joint is locked. This lock sensor can be linked with a processor (120) to support real-time reflection of the lock status of each joint in the work algorithm.

[0076] The display system (170) can output various information related to drip coffee manufacturing automation services as graphic images.

[0077] In an example, the display system (170) can display a drip coffee manufacturing automation service user interface (UI), baking management content and / or content management system (CMS) based data, etc.

[0078] Such displays may include at least one of a liquid crystal display (LCD), a thin film transistor-liquid crystal display (TFT LCD), an organic light-emitting diode (OLED), a flexible display, a 3D display, and an e-ink display.

[0079] The above components may be placed within the housing of such drip coffee robot (10), and the user interface may include a touch sensor on a display configured to receive user touch input.

[0080] The drip coffee robot (10) according to the embodiment including the above-described components may be implemented to include components necessary to perform a predetermined process involved in the manufacture of drip coffee.

[0081] In addition, referring to FIG. 1b, the smart dripping control method (S10) according to an embodiment of the present invention inputs data periodically measured from a collaborative robot arm (nozzle) and a sensor system (temperature sensor, flow sensor, vision sensor, etc.) into a machine learning model to predict the nozzle angle (d_theta), water flow rate (d_Qw), hot water temperature (d_Tc), and whether extraction has ended (d_flag) in real time, and can dynamically control the drip coffee extraction process based on this. That is, by controlling the collaborative robot, heater, and valve device based on the correction values ​​predicted by the model, the taste and quality of the coffee can be maintained or improved according to the type of coffee beans, environmental variables, and real-time extraction conditions. Specifically, the smart dripping control method (S10) according to an embodiment of the present invention

[0082] It may include an initial setup and environment configuration step (S11), a data collection and preprocessing step (S12), a hyperparameter tuning step (S13), a model training and validation step (S14), a model deployment and real-time extraction control step (S15), and a feedback loop and continuous model update step (S16).

[0083] The embodiment initializes the hardware (collaborative robot arm, heater, valve, sensor, etc.) required for the drip coffee extraction process and the software APIs (e.g., robot_arm_api, heater_api, valve_api) for controlling them, loads libraries required for machine learning (ML) analysis, configures a basic environment for setting model parameters, and can also perform a connection with a database (DB) where sensor data (temperature, flow rate, image sensor, etc.) is stored and managed.

[0084] In addition, the embodiment reads the coffee extraction temperature (Tr) and hot water temperature (Tw) during dripping from a temperature sensor, measures the extraction flow rate (Fr) using a flow sensor, and extracts image-based indicators such as the coffee bean expansion amount (Eb) and extract color (Cc) through a vision sensor (vision_module). Furthermore, depending on the extraction recipe, label data can be constructed by setting the nozzle angle (d_theta), water flow rate (d_Qw), hot water temperature (d_Tc), and extraction completion status (d_flag) to standard values ​​or values ​​provided by previous logs or experts (baristas). The collected sensor data and label data can be stored in a DB or local memory for future use in model training. After a sufficient amount of collected data has accumulated, standardization operations, such as removing missing values ​​and outliers, can be performed to create the data in the form of machine learning model input (X, Y).

[0085] In addition, the embodiments may attempt various combinations of parameters such as n_estimators and max_depth, for example, the number of trees or maximum tree depth, for Random Forest, XGBoost, and other ensemble-based models. In various embodiments, the embodiments may use MultiOutputRegressor to predict four outputs (d_theta, d_Qw, d_Tc, d_flag) simultaneously in the form of regression or regression and classification. In addition, in various embodiments, the process of finding optimal parameters may be performed using techniques such as GridSearch, RandomSearch, and Bayesian Optimization. That is, the embodiments adopt an ensemble-random forest (or a similar boosting-bagging model) based on MultiOutputRegressor, and each parameter (n_estimators, max_depth, etc.) can be optimized through hyperparameter tuning.

[0086] In addition, the embodiment can divide the collected data (X, Y) into training and validation sets, perform training (fit) on the optimal model (best_model), and evaluate model performance by comparing the prediction results (predictions) obtained by inputting validation data (X_val) with the actual correct answer (Y_val). Also, d_theta, d_Qw, and d_Tc are continuous values ​​(regression), and mean squared error (MSE), RMSE, MAE, etc., can be used. d_flag is predicted as a continuous value in the range of 0 to 1, but accuracy, etc., can be measured by classifying based on 0.5.

[0087] In addition, the embodiment loads the finally selected optimal model (best_model) into the robot control system, and when the actual drip coffee extraction operation begins, sensor values ​​(Tr, Tw, Fr, Eb, Cc) can be read in real time at regular intervals (e.g., 0.5 seconds). In addition, when the sensor values ​​are input into the machine learning model, the nozzle angle (d_theta), water flow rate (d_Qw), hot water temperature (d_Tc), and extraction termination status (d_flag_cont) are output, and the robot arm, valve, and heater can be controlled according to the corresponding correction values. If it is determined that d_flag_cont ≥ 0.5, the extraction can be stopped (e.g., by shutting off the valve) and the loop can be terminated.

[0088] In addition, the embodiments can automatically accumulate sensor logs and taste evaluation / extraction results, which are continuously newly generated in the real-time control loop, into a database. In various embodiments, retraining is performed with new data in an offline environment at regular intervals (e.g., once a day, once a week, etc.), and only models with improved performance can be deployed to the field. Since the model can be continuously improved even during operation, it is possible to flexibly respond to equipment aging, environmental changes, and the introduction of new coffee beans.

[0089] In various embodiments, whether extraction is terminated (d_flag) may be predicted in the form of a continuous value and determined to be an extraction stoppage if it is 0.5 or higher, but is not limited thereto, and as needed, the threshold value may be variably set, such as in the range of 0.4 to 0.6, and the allowable error for extraction temperature and flow rate deviations (Tr, Fr, etc.) may also be adjusted, for example, in the range of ±1°C to ±5°C or ±10% to ±20%.

[0090] The embodiment interconnects the collaborative robot arm (nozzle) control unit, the hot water heater control unit, and the valve control unit to input data collected in real time from sensors (temperature, flow rate, vision) into a machine learning model, thereby dynamically correcting the nozzle angle, water flow rate, hot water temperature, and extraction end time. Through this, drip coffee of consistent quality can be automatically extracted even if the type of coffee beans or surrounding environmental conditions vary, and extraction accuracy can be further improved by retraining the model as log data accumulates during actual use.

[0091] FIGS. 2 to 6 are drawings for explaining the structure of a drip coffee robot according to an embodiment of the present invention. Specifically, FIGS. 2 and 3 are overall perspective views of the drip coffee robot, FIG. 4 is a front view, FIG. 5 is a side view, and FIG. 6 is a top view.

[0092] Referring to FIGS. 2 to 6, from a structural perspective, the drip coffee robot (10) can be divided into a shelf section (1), a hand robot section (2), a grinder section (3), a dripper section (4), a beverage serving section (5), and a processor section (6).

[0093] The shelf section (1) may refer to a section that includes a shelf structure capable of storing multiple dosing cups that accommodate various types of different coffee beans.

[0094] This shelf section (1) may include a shelf holder (11), a dosing cup section (12), a beverage washing section (16), and a work table (18).

[0095] The shelf holder (11) may refer to a rack formed along a horizontal line so that the dosing cup portion (12) can be placed upright.

[0096] Each of the above shelf supports (11) can support a row of dosing cups, and multiple such shelf supports (11) can be formed on the shelf section (1).

[0097] In addition, the dosing cup portions (12) placed on the shelf holder (11) can be arranged horizontally and vertically at regular intervals.

[0098] The dosing cup portion (12) may include a plurality of dosing cup assemblies, each comprising a dosing cup (13) and a dosing cup lid (14).

[0099] The dosing cup (13) may be in a shape that is easy for the gripper of the hand robot part (2) to grasp. The dosing cup lid (14) may cover the top of the dosing cup (13) to block air from the outside.

[0100] In addition, the dosing cup (13) can store at least one type of coffee bean inside.

[0101] The beverage washing section (16) may refer to a section in which a water spray nozzle (17) is formed to wash the used dosing cup (13) and / or beverage cup (15).

[0102] The hand robot unit (2) can wash the beverage cup (15) before and after serving the beverage with water sprayed through the water spray nozzle (17) of the beverage washing unit (16).

[0103] The work table (18) may be equipped with certain components necessary for making drip coffee. Specifically, the work table (18) may be equipped with a beverage cup (15), a beverage washing unit (16) and a grinder unit (3), a water heater (100) and / or a hot water storage cup (101).

[0104] In addition, a heater and a line for supplying or discharging water may be installed at the bottom of the above work table (18).

[0105] The hand robot section (2) may refer to a section equipped with a hand robot that mixes one or at least two coffee beans to produce drip coffee. The hand robot section (2) may be in the form of a collaborative robot module (not shown) comprising a plurality of hand robots.

[0106] The hand robot part (2) may include a robot body (21), a gripper part (22), and a pendant (23).

[0107] The robot body (21) is a main body part made of a multi-joint module structure.

[0108] Specifically, the robot body (21) may include a plurality of adjustable multi-joint modules to change the angle, direction, and position of the gripper part (22).

[0109] At this time, when the hand robot unit (2) is implemented in the form of a collaborative robot, various drip coffee manufacturing environments may exist, such as when the paths of other hand robots overlap or collide. Therefore, the robot body (21) can be implemented in a multi-joint module structure so that the hand robot unit (2) can change its posture to an optimized structure each time.

[0110] In addition, each hand robot unit (2) is programmed with a movement algorithm to simultaneously perform different processes included in the drip coffee manufacturing process (e.g., dosing cup transfer, coffee bean grinding, etc.), and can perform parallel operations without collision by detecting real-time collision situations with a collision detection sensor (e.g., torque sensor) equipped for each joint module.

[0111] To this end, the robot body (21) can operate on an optimized work path derived by pre-setting the height, length, rotation radius, and angle radius of the robot body (21) suitable for the drip coffee manufacturing environment by performing a simulation on the devices to be used during the initial setup.

[0112] The gripper part (22) is the end (grip) part of the main body for gripping components required for the automation process, such as a dosing cup, a hot water storage cup, and a button, or for performing a press input.

[0113] For example, the gripper part (22) can be changed or replaced in various forms depending on the purpose, such as a clamp-type gripper and a hook-type gripper. In this specification, the description is based on the implementation of a clamp-type gripper for gripping a dosing cup, a beverage cup, etc.

[0114] FIGS. 7 and FIGS. 8 are drawings for explaining the structure of a gripper part according to an embodiment of the present invention. Specifically, FIG. 7 is an overall perspective view of the gripper part, and FIG. 8 is an enlarged view of the gripper part.

[0115] Referring to FIGS. 7 and 8, the gripper part (22) may include a grip box (200), a grip (210), a grip connecting part (220), and a grip rail (230).

[0116] The grip box (200) may be a part in which certain parts (e.g., motors) are arranged to enable the grip (210) to grip the cup.

[0117] This grip box (200) is also a part that connects the robot body (21) and the gripper part (22) of the hand robot part (2).

[0118] Additionally, the grip box (200) is implemented with a rotatable structure so that it can be rotated to have a predetermined angle with respect to the robot body (21) of the hand robot part (2). This may be for the purpose of rotating the grip (210).

[0119] Additionally, a specific pneumatic or vibration module may be disposed in the grip box (200). A specific vibration may be applied to the grip (210) through the module.

[0120] The grip (210) is connected to the grip box (200) through the grip connecting part (220) and can be formed in a C-shaped clamp shape having a predetermined curvature so as to be able to grip a cup.

[0121] Specifically, the grip (210) includes a grip upper portion (211) and a grip lower portion (212) spaced apart from each other. At this time, the grip upper portion and the grip lower portion (211, 212) form a space between them having a predetermined curvature that can hold a cup.

[0122] The upper and lower grip portions (211, 212) may extend outward from the grip connecting portion (220). At this time, the upper and lower grip portions (211, 212) may each be individually connected to the grip connecting portion (220).

[0123] Additionally, the upper and lower parts of the grip (211, 212) can be spread apart in opposite directions (as illustrated, left and right directions). In other words, the spacing between the upper and lower parts of the grip (211, 212) can be widened. This may be a spreading action performed when gripping an object with a diameter larger than the default diameter of the grip (210). Naturally, a closing action may also be performed.

[0124] That is, depending on the left and right movement of the upper and lower grips (211, 212), the width of the space formed by the upper and lower grips (211, 212) can be adjusted to match the size and diameter of the object.

[0125] In addition, at least one of the upper and lower grips (211, 212) may be implemented as a rotatable and protruding structure.

[0126] The upper grip (211) and lower grip (212) may each include an independent rotary drive module and a linear protrusion drive module.

[0127] The rotary drive module may include a motor and a gear device so that the upper and lower parts of the grip can rotate independently up to 180 degrees around individual rotation axes.

[0128] Additionally, the linear protrusion drive module may be configured to include a piston or a linear actuator so that a part of the grip protrudes forward to press a button.

[0129] For example, when a button is pressed, the processor (120) recognizes the position and shape of the gripping object (e.g., an operating button) through the vision sensor and distance sensor of the sensor system (160). Additionally, based on the recognized data, the rotation drive module and the linear protrusion drive module are controlled to individually rotate and protrude the upper grip (211) forward to an angle and position suitable for pressing the button. Accordingly, the posture is changed so that the end portion of the upper grip (211) can accurately contact the surface of the button, and a stable button pressing operation can be performed.

[0130] In addition, the rotation axes of the upper and lower parts of the grip can be set in mutually different directions. Accordingly, the range of responsiveness to gripping objects of various shapes and sizes can be expanded, and even for objects that are difficult to grip or operate buttons on with conventional simple opening / closing or single-type grippers, the gripper part according to the present invention provides the effect of performing stable and precise gripping and operation through individual rotation and protrusion structures.

[0131] That is, at least one of the upper and lower parts of the grip (211, 212) can be rotated so as to be perpendicular to the grip connecting part (220) and protrude in the front direction.

[0132] Accordingly, a portion of the rotating and protruding grip can be converted into a form that facilitates pressing a button depending on the shape of the object. That is, the edges of the rotating and protruding grip can come into contact with a predetermined button to perform a button pressing action.

[0133] The grip connecting portion (220) extends from the grip box (200) toward the grip (210) and connects the grip box (200) and the grip (210).

[0134] The grip lane (230) refers to a groove formed horizontally in the grip box (200). Additionally, the grip lane (230) is provided with a grip connecting part (200).

[0135] That is, the grip connecting part (220) moves left and right along the grip lane (230) and can move the grip (210) left and right.

[0136] The pendant (23) is a part that includes a controller for controlling the movement of the hand robot unit (2). Additionally, the pendant (23) may be equipped with an interface module (140) and / or a display system (170) that can operate the hand robot unit (2) to perform a predetermined operation.

[0137] This hand robot part (2) may further include a motor part that provides power to operate the entire included component. For example, the motor part may include parts such as a link, cam, gear, latch, gear train, etc., for converting the movement of the motor into the opening and closing of the gripper part (22).

[0138] Additionally, the hand robot unit (2) may further include vertical and / or horizontal tracks for moving to the shelf unit (1) and the beverage supply unit (5) according to the manufacturing process.

[0139] The above track can be implemented in a structure that is positioned at a predetermined distance from the shelf unit (1) and the beverage supply unit (5), and moves the hand robot unit (2) according to the shape formed by the track while supporting the robot body (21).

[0140] Additionally, a predetermined rotating part capable of rotating the hand robot part (2) in place may be provided between the track and the hand robot part (2). Accordingly, the hand robot part (2) can rotate at a predetermined angle in the direction of the shelf part (1) and the beverage supply part (5).

[0141] The grinder section (3) may refer to a section equipped with a grinder for grinding the coffee beans contained in the dosing cup (13).

[0142] The grinder part (3) may include a grinder cover (31), a bean receiving part (32), a bean discharge port (33), and an operating button (34).

[0143] The grinder cover (31) is formed in a structure that can cover the upper part of the bean receiving portion (32) so that foreign matter does not enter the bean receiving portion (32) when the grinder is not in use.

[0144] The bean receiving portion (32) is a portion that receives the beans located inside the dosing cup (13) held by the hand robot portion (2).

[0145] The beans received in the bean receiving portion (32) can be discharged to the bean discharge outlet (33) after undergoing a predetermined grinding process when the pressing of the operating button (34) is detected.

[0146] Although not shown, a grinder blade for grinding beans may be placed between the bean receiving portion (32) and the bean discharge portion (33).

[0147] That is, since the coffee beans discharged through the coffee bean outlet (33) are in a ground state, the ground coffee beans are referred to as “ground coffee beans” below.

[0148] The dripper part (4) may refer to a part that filters ground coffee beans with a predetermined filter and slowly passes hot water through them to produce an extract.

[0149] The dripper part (4) may include a dripper (41) and a support frame (42).

[0150] The dripper (41) is a funnel-shaped device that holds a filter and ground coffee beans. The filter is a piece of paper that prevents the ground coffee beans from being carried away with the coffee when hot water falls from the dripper (41).

[0151] The support frame (42) is a device that fixes the dripper (41) on a server (e.g., a beverage cup (15)) where the extract is collected.

[0152] That is, an extract is produced from ground coffee beans by passing through the dripper part (4), and the produced extract can be provided to the customer as drip coffee.

[0153] The beverage serving section (5) may refer to a section that provides drip coffee, which is produced after extraction is complete, to the customer.

[0154] The coffee serving table (51) of the beverage serving unit (5) may be equipped with a dripper unit (4), a water heater (100) and / or a hot water storage cup (101).

[0155] The processor unit (6) is a part that controls the drip coffee robot (10) to receive an order from the orderer and to manufacture and provide drip coffee according to the order.

[0156] The processor unit (6) may include an interface module (61) such as a kiosk and a processor (62).

[0157] In addition, the processor (120) of the drip coffee robot (10) according to an embodiment of the present invention can control the entire components of the drip coffee robot (10) to perform at least one drip coffee manufacturing process.

[0158] Hereinafter, drip coffee manufacturing processes performed in a baking system by an application and / or program executed by at least one processor (120) of a drip coffee robot (10) according to an embodiment of the present invention will be described in detail with reference to the attached FIGS. 9 to 12.

[0159] Hereinafter, the process of the at least one processor (120) operating to execute instructions of the application and / or program to perform the drip coffee manufacturing process of the drip coffee manufacturing system described above is described in a shortened manner as the processor (120) performing the process.

[0160] FIG. 9 is a flowchart illustrating a drip coffee manufacturing process according to an embodiment of the present invention.

[0161] Referring to FIG. 9, in an embodiment, the processor (120) is, Order information can be obtained from the interface module (61). (S101)

[0162] The interface module (61) is, for example, a kiosk, and the kiosk can display various types of drip coffee menus on the display of the kiosk.

[0163] For example, various order information, such as menu information, price information, payment information, and / or beverage pickup time information, can be displayed on the display.

[0164] Accordingly, the orderer can order a predetermined menu, and in the embodiment, the processor (120) can obtain order information, which is information about the ordered menu.

[0165] At this time, for each of the above menus, a manufacturing process for producing the corresponding menu may be pre-programmed and mapped. The mapped content may include the attributes of the menu and the movement path of the hand robot unit (2) for producing the menu.

[0166] Specifically, the attributes of the above menu may include the type of coffee beans required for each menu, the amount of coffee beans, the amount of water, the coffee bean grinding specifications, the coffee bean grinding time, the coffee bean dripping time, and / or the hot water ratio. In addition, the movement path, gripper width, rotation angle, etc. of the hand robot unit (2) for making drip coffee may be pre-programmed according to the attributes of the above menu.

[0167] To this end, a unique number is pre-set for each coffee bean, which is the raw material for each drip coffee menu, and / or for each dosing cup containing the coffee beans, and can be set on the shelf (1).

[0168] FIG. 10 is an example of a drawing showing a plurality of dosing cups set on a shelf according to an embodiment of the present invention. FIG. 11 is also an example of a drawing explaining the types of coffee beans mapped to each coffee bean stored in each dosing cup according to an embodiment of the present invention.

[0169] Referring to FIGS. 10 and 11, a shelf section (1) can be set up by arranging multiple dosing cups (13) (e.g., 6 to 12) containing different types of coffee beans on a single row of shelf holders (11). That is, the shelf section (1) may have a structure in which multiple dosing cups (13) are arranged in horizontal and vertical directions.

[0170] FIG. 10 above is illustrated based on the assumption that two dosing cups (13) containing the same coffee beans are set up in succession. Additionally, it is illustrated that one hand robot unit (2) corresponds to one section of the shelf unit (1). However, if the shelf unit (1) is implemented with multiple sections, additional hand robot units (2) may be set up for each section, and accordingly, multiple hand robot units (2) may operate as collaborative robots for a wider shelf unit (1) than illustrated.

[0171] In an embodiment, the processor (120) can map a unique number to each different type of coffee bean.

[0172] Additionally, in the embodiment, the processor (120) can map the position of each dosing cup (13) within the shelf section (1) where the unique number is mapped to a matrix-shaped position information corresponding to the unique number.

[0173] Accordingly, when a unique number of the coffee bean (e.g., 88) is selected from the order information, the processor (120) in the embodiment can determine the position of the dosing cup (13) based on the mapped position information. Additionally, an optimized movement path to the determined position can be automatically programmed and provided to the hand robot unit (2).

[0174] In addition, specific coffee bean information (CI) can be further mapped and stored for each type of coffee bean.

[0175] Here, the coffee bean information (CI) according to the embodiment may be information including a unique number for each coffee bean, origin of the coffee bean, variety of the coffee bean, processing method of the coffee bean, and at least one flavor. For example, in FIG. 11, the first coffee bean information with a unique number of 1 may have information mapped and stored including “1 (unique number) / India Akaru Microlot (origin of the coffee bean) / Sin.9 (variety of the coffee bean) / Raspberry (first flavor) / Cinnamon (second flavor) / Blackberry (third flavor)”.

[0176] Returning to the example, the processor (120) can move to the location of the unique number when it receives only the unique number.

[0177] That is, in the embodiment, the order information obtained based on the interface module (61) may include a unique coffee bean number and a manufacturing process programmed into the corresponding coffee bean.

[0178] Accordingly, in the embodiment, the processor (120) can extract the unique number of the coffee beans included in the order information and start the drip coffee manufacturing process.

[0179] Next, in the embodiment, the processor (120) is, A target dosing cup containing coffee beans can be determined based on the acquired order. (S102)

[0180] Here, the target dosing cup refers to the dosing cup set at a position matching the unique coffee bean number included in the order information above.

[0181] In other words, in the embodiment, the processor (120) can extract a unique coffee bean number included in the acquired order information and determine a target dosing cup set at a location that matches the extracted unique coffee bean number.

[0182] For example, if a customer orders a menu using coffee beans number 88, the dosing cup set at position 88 can be determined as the target dosing cup (13-T).

[0183] Accordingly, in the embodiment, the processor (120) has a hand robot part (2) It can be controlled to move to and grasp the target dosing cup determined above. (S103)

[0184] In detail, in an embodiment, the processor (120) can control the gripper part (22) of the hand robot part (2) to first open the dosing cup lid (14) of the target dosing cup (13-T) when the gripper part (22) moves in front of the determined target dosing cup (13-T). Referring again to FIG. 7, the dosing cup lid (14) has a structure in which the central part protrudes outward, and the protruding part is similar to a rectangular structure so that it can be implemented in a form that is easy for the gripper part (22) to grasp. In addition, the grasped dosing cup lid (14) can be temporarily placed in the space between the target dosing cup (13-T) and an adjacent dosing cup.

[0185] Additionally, in the embodiment, the processor (120) can adjust the width of the grip (210) of the gripper part (22) to match the width of the target dosing cup (13-T). When the width of the grip (210) is adjusted, the target dosing cup (13-T) can be gripped by controlling the grip (210) to tighten.

[0186] Meanwhile, the gripper part (22) can change the shape of the grip (210) according to the object to be gripped (e.g., a dosing cup and lid, a beverage cup, a grinder lid, an operating button, etc.).

[0187] To this end, in the embodiment, the processor (120) can obtain size and shape information of the grasping target through a vision sensor, a distance sensor and / or a contact sensor included in the sensor system (160).

[0188] Afterward, the processor (120) can automatically adjust the shape by controlling the grip (210) based on the information obtained above.

[0189] For example, when gripping a dosing cup (13), the processor (120) adjusts the upper and lower grips (211, 212) to correspond to the diameter of the dosing cup to maintain a stable gripping state. When gripping a beverage cup (15), the upper and lower grips (211, 212) are adjusted to a larger diameter so that the beverage cup can be gripped without damage. For an operating element such as a button, at least one of the upper and lower grips (211, 212) can be rotated and protruded so that its shape can be deformed into a structure suitable for performing a button pressing action.

[0190] Through this, the grip (210) can flexibly respond to various gripping targets, thereby improving the stability of the drip coffee manufacturing process.

[0191] In addition, in the embodiment, the processor (120) is, The gripped target dosing cup can be moved to the grinder section. (S104)

[0192] Since coffee beans are stored inside the held dosing cup, in the embodiment, the processor (120) can control the grip (210) to maintain a horizontal position so that the coffee beans inside the dosing cup held by the hand robot part (2) do not spill out during movement.

[0193] Additionally, in an embodiment, the processor (120) can control the grip (210) to temporarily lower the dosing cup near the grinder section (3) when the movement to the grinder section (3) is completed, and to remove the grinder cover (31) so that the space of the bean receiving section (32) of the grinder section (3) is exposed to the outside. The grinder cover (31) may be in a shape that is easy to grip by the grip (210), similar to the dosing cup lid (14).

[0194] That is, in the embodiment, the processor (120) can remove the grinder cover (31) and then move the grip (210) to the bean receiving portion (32) of the grinder portion (3) by grasping the dosing cup again to grind the beans.

[0195] In addition, in the embodiment, the processor (120) is, A preset amount of coffee beans can be fed into the bean receiving portion of the grinder and ground. (S105)

[0196] To this end, in the embodiment, the processor (120) can rotate the gripper part (22) of the hand robot part (2) by a predetermined angle.

[0197] FIG. 12 is an example showing the gripper part of the hand robot unit rotated at a predetermined angle according to an embodiment of the present invention. The gripper part (22) of the hand robot unit (2) can be rotated at a predetermined angle in the grinder part (3) and the dripper part (4), and although only the dripper part (4) is shown in FIG. 12, it can also be performed in the grinder part (3).

[0198] Referring to FIG. 12, in the embodiment, the processor (120) can rotate the gripper part (22), which is in a state of holding a dosing cup containing coffee beans, at a predetermined angle.

[0199] At this time, the dosing cup may contain only a single serving of coffee beans or a quantity of coffee beans that can be dispensed multiple times.

[0200] In the former case, in the embodiment, the processor (120) can adjust the angle and angle holding time of the gripper part (22) to drop all the coffee beans stored in the dosing cup.

[0201] Meanwhile, in the latter case, in the embodiment, the processor (120) can precisely adjust the rotation angle and the rotation angle holding time of the gripper part (22) to drop only one amount of coffee beans using the coffee bean weight data measured in real time by the weight sensor.

[0202] To this end, the attributes of each menu (bean) may include information on the amount of beans required for one serving.

[0203] In detail, in an embodiment, the processor (120) can compare the current amount of coffee beans measured by the weight sensor with a previously stored reference value for a single serving of coffee beans. Additionally, when the measured current amount of coffee beans approaches the target amount, the amount of beans can be controlled by reducing the rotation angle of the gripper part (22) or stopping the rotation state.

[0204] At this time, the processor (120) can finely adjust the rotation angle of the gripper part (22) in real time during dropping by adopting a feedback loop method.

[0205] When all the coffee beans have been dropped, in the embodiment, the processor (120) can release the grip on the empty dosing cup and perform a press input on the operation button (34). Thus, the grinder of the grinder unit (3) can be operated.

[0206] In order to perform a press input for the above-mentioned operating button (34), in the embodiment, the processor (120) can control the shape of the gripper part (22) to be changed to a structure capable of performing a press input by rotating and protruding at least one of the upper and lower parts of the grip (211, 212).

[0207] Accordingly, in the embodiment, the processor (120) can operate the grinder of the grinder unit (3). Then, the coffee beans are ground and ground coffee beans are produced.

[0208] Additionally, in the embodiment, the processor (120) can move the gripper part (22), which holds the dosing cup containing ground coffee beans, to the dripper part (4).

[0209] In addition, in the embodiment, the processor (120) is, You can insert ground coffee beans into the dripper and pour hot water. (S106)

[0210] Meanwhile, when making drip coffee, the taste of the drip coffee may vary depending on the dripping method, such as the amount of hot water poured, the pouring time, and the pouring method, so delicate control of the gripper part (22) is required to maintain optimal extraction conditions.

[0211] To this end, in the embodiment, the processor (120) may include “extraction condition information” that includes the grind particle size, amount of hot water added, adding time, adding method, extraction temperature, extraction speed, bean expansion speed, extract color and transparency, etc., which serve as criteria for optimal extraction conditions for each menu (bean) menu attribute.

[0212] That is, in the embodiment, the processor (120) can implement smart dripping control technology that dynamically adjusts the hand robot unit (2) in real time based on extraction condition information mapped for each menu (bean).

[0213] Specifically, in the embodiment, the processor (120) can monitor the real-time dripping state through the temperature sensor, vision sensor, and flow sensor included in the sensor system (160).

[0214] More specifically, the processor (120) can obtain extraction temperature data by measuring the real-time temperature change of the hot water dropped from the dripper (4) based on the temperature sensor. The measured extraction temperature data is compared with the optimal temperature range preset for each coffee bean, and if a temperature change greater than a preset error is detected, the processor (120) can perform temperature control.

[0215] Additionally, the processor (120) can obtain extraction speed data by measuring the flow rate and amount of hot water dripped through the dripper (4) based on flow sensors installed at the top and bottom of the dripper (4). If the drip speed is detected to be excessively fast or slow based on the measured extraction speed data, the processor (120) can control the dripping speed of the hot water.

[0216] Additionally, the processor (120) can acquire coffee bean expansion data, which is image data regarding the speed and degree of expansion of the coffee beans as they absorb hot water when hot water is poured onto the ground coffee beans, based on a vision sensor located on the side or top of the dripper part (4). The coffee bean expansion data can be used to control the dripping speed of the hand robot part (120) so that the condition of the coffee beans does not reach a state of under- or over-extraction.

[0217] Additionally, the processor (120) can acquire extract data, which is image data of the color, turbidity, and transparency of the coffee extract falling to the bottom of the dripper (41), based on a vision sensor or optical sensor located at the bottom of the dripper part (4). Through this, if the concentration of the extract deviates from a set standard, the processor (120) can immediately adjust the flow rate of hot water and the dripping pattern.

[0218] That is, in the embodiment, the processor (120) can collect extraction temperature data, extraction speed data, coffee bean expansion data and extract data in real time and analyze them in conjunction with a deep learning model that has been pre-trained with the extraction condition information.

[0219] The above deep learning model can be pre-trained based on a training dataset built through various coffee bean extraction experiments. That is, it is trained to determine the optimal extraction conditions for each type of coffee bean and extraction situation. In other words, the optimal extraction conditions for each type of coffee bean can be pre-set as a standard.

[0220] Afterward, the processor (120) can automatically change and optimize the extraction situation based on the real-time analysis results extracted from the deep learning model.

[0221] In detail, in an embodiment, the processor (120) can finely adjust the temperature of the hot water by controlling the water heater (100) when the temperature, color, and transparency of the drip coffee extracted according to extraction temperature data and extraction data differ from a predetermined condition.

[0222] Additionally, in the embodiment, the processor (120) can adjust the flow rate of hot water by automatically adjusting the position, movement speed, and dripping speed of the gripper part (22) when the condition of the coffee beans differs from a preset condition according to extraction speed data and coffee bean expansion data.

[0223] Additionally, in the embodiment, the processor (120) can adjust the drip pattern (e.g., circular, zigzag, spiral pattern) at the top of the dripper so that the extract is uniformly extracted according to the extraction state determined through the collected data, using the gripper part (22).

[0224] That is, in the embodiment, the processor (120) can perform dripping based on smart dripping control technology and according to extraction conditions matched to the menu (beans) being manufactured.

[0225] Accordingly, this embodiment can provide the effect of maintaining consistent and optimal coffee quality in various situations, such as not only the type of coffee beans but also environmental variables and changes in the condition of the beans.

[0226] Additionally, in the embodiment, the processor (120) can obtain a finished coffee extract (i.e., drip coffee) by performing dripping.

[0227] Additionally, in the embodiment, the processor (120) may transfer the finished drip coffee into a beverage cup (15) for serving to the customer. Alternatively, to omit the transfer process, the beverage cup (15) may be pre-set at the location where the coffee extract falls from the dripper part (4).

[0228] In addition, in the embodiment, the processor (120) is, The finished drip coffee can be moved to the beverage serving area. (S107)

[0229] In detail, in an embodiment, the processor (120) can provide the finished drip coffee to the customer by holding the beverage cup (15) containing the finished drip coffee and positioning the beverage cup (15) in the beverage serving unit (5) while maintaining the horizontal position of the gripper part (22) so that it does not spill.

[0230] Meanwhile, the above work table (18) and beverage serving area (5) may have used dosing cups (13), empty beverage cups (15) left after the customer drinks drip coffee, etc. placed on them.

[0231] At this time, in the embodiment, the processor (120) can scan a plurality of cups placed on the work table (18) and / or the beverage serving unit (5) through an image sensor.

[0232] Specifically, the processor (120) can analyze image data of a cup captured by an image sensor included in the sensor system (160) through a deep learning neural network-based image recognition algorithm. Specifically, it can automatically filter cups to be washed by recognizing whether there is residue inside the cup, whether the cup has moved, and / or the identification code of the cup.

[0233] Accordingly, in the embodiment, the processor (120) may determine the priority among the identified cups to be washed and instruct the hand robot unit (2) to sequentially move the cups to the beverage washing unit (16). For example, the cups to be washed may be used dosing cups (13) and beverage cups (15).

[0234] That is, the processor (120) can maintain the hygienic condition of the work table (18) and the beverage supply unit (5) and ensure the continuity of the drip coffee manufacturing process through this.

[0235] Additionally, in the embodiment, the processor (120) can determine a first cup among the filtered cups to be washed as a target and control the gripper part (22) to grasp the empty beverage cup.

[0236] In addition, in the embodiment, the processor (120) is, Used cups can be moved to the beverage washing unit for automatic washing. (S108)

[0237] In detail, in an embodiment, the processor (120) can move the first cup to the beverage washing unit (16) and position it at the water spray nozzle (17). At this time, the beverage washing unit (16) can be set to automatically spray water when a predetermined object is detected using an infrared sensor.

[0238] At this time, in the embodiment, the processor (120) can implement an advanced cup management logic based on a cup identification number to highly manage various cups used in the drip coffee manufacturing process.

[0239] In the embodiment, the cup management logic refers to the logic in which the hand robot unit (2) performs specific posture control according to the height, diameter, and material of each cup.

[0240] First, each dosing cup (13) and beverage cup (15) may be individually assigned a unique cup identification number. The cup identification number may be attached to the bottom or side of the cup in the form of an RFID tag, barcode, or QR code. This cup identification number may be registered and managed in a memory (110) from the time the cup is first used, thereby continuously recording and managing status information such as the cumulative number of uses, number of washes, and lifespan of each cup.

[0241] Additionally, in the embodiment, the processor (120) can automatically recognize the identification number of the cup through the sensor system (160) after the cup is used, and through this identification number, look up the status information of the cup stored in the memory (110) to classify the cup into one of 1) a reusable cup, 2) a cup to be washed, or 3) a cup to be discarded.

[0242] Reusable cups are those where cumulative use and cleaning status are within the standard range and no foreign substances are detected. Cups subject to cleaning are those where residue is detected inside the cup but the lifespan and durability are good. Cups subject to disposal are those where the lifespan has been exceeded or defects such as damage or breakage are detected.

[0243] In the embodiment, the processor (120) can issue commands to the hand robot unit (2) according to the classification result to store immediately reusable cups in a predetermined location, move cups that need to be washed to a washing process, and transfer cups that need to be discarded to a separate waste bin for disposal.

[0244] Additionally, in the embodiment, the processor (120) can control the hand robot part (2) to select a specific posture to increase cleaning efficiency according to the material of the cup (e.g., glass, plastic, stainless steel, etc.), diameter, and height.

[0245] Specifically, the processor (120) can determine the most suitable posture and motion for washing by collecting shape data of each cup through the sensor system (160) or by querying shape data of the cup from a database that has been stored in advance along with the cup identification number.

[0246] For example, in the case of a long, narrow beverage cup (15), the hand robot unit (2) can assume a position in which the cup is completely inverted to allow a high-pressure stream of water from the water nozzle (17) to be injected directly into the interior, thereby facilitating internal cleaning. On the other hand, in the case of a wide-diameter cup, the position control can be performed by tilting the cup at a certain angle to allow for rotational movement of the cup so that contaminants on the inner wall of the cup can be effectively removed.

[0247] In other words, the cup management logic and posture control technology implemented according to the embodiment of the present invention significantly improve cup management efficiency and cleaning quality, and provide the effect of further enhancing the completeness of automation of the entire drip coffee manufacturing process by maximizing the durability and usability of the cup.

[0248] Additionally, in the embodiment, the processor (120) may perform self-cleaning of the gripper part (22) in the beverage washing part (16) when the gripper part (22) is contaminated with coffee grounds and beverage residue.

[0249] To this end, in the embodiment, the processor (120) can detect the condition of the grip contamination based on a sensor system. Accordingly, if it is determined to be a target for cleaning, in the embodiment, the processor (120) can move forward to the water spray nozzle (17), automatically spread the upper and lower parts of the grip apart, and then control the surface and contact parts to be cleaned at high pressure through water spraying while maintaining an appropriate angle and posture.

[0250] In addition, after cleaning is complete, the water remaining in the gripper part can be quickly removed using a pneumatic or vibration module placed in the gripper part (22) to dry it so that an immediate subsequent process can be performed.

[0251] Accordingly, in the embodiment, the processor (120) can automatically clean an object to be cleaned, including at least one of a dosing cup, a used beverage cup, and a gripper part (22), using water sprayed from a water nozzle (17).

[0252] Additionally, in the embodiment, the processor (120) can consider the drip coffee manufacturing process to be complete when the automatically washed cups are placed in a predetermined space on the work table (18) and dried.

[0253] Meanwhile, drip coffee can be produced using only one type of coffee bean, but it can also be produced by blending two or more types of coffee beans according to the consumer's preference. Accordingly, the processor (120) according to the embodiment can implement a customized blend drip coffee production technology that automatically generates a recipe and produces it by simply surveying the customer's drip coffee preference based on a kiosk.

[0254] To this end, in the embodiment, the processor (120) may provide an interface to the kiosk that allows the orderer to input their preferred flavor, taste, and caffeine sensitivity when selecting a blend drip coffee menu. Accordingly, the orderer's preference data can be obtained.

[0255] Additionally, in the embodiment, the processor (120) can correspond to coffee bean information (e.g., taste profile data expressing acidity, sweetness, bitterness, body, etc.) stored in the memory (110) based on the preference data.

[0256] Accordingly, in the embodiment, the processor (120) can determine a first coffee bean and a second coffee bean corresponding to the preference data. The number of blended coffee bean types may be two or more, but for convenience of explanation, the description is based on two.

[0257] In addition, in the embodiment, the processor (120) can automatically calculate the optimal blending ratio between the first coffee beans and the second coffee beans by utilizing a pre-learned machine learning-based blending ratio recommendation algorithm.

[0258] In this case, the machine learning model is pre-trained to automatically predict the optimal values ​​for the combination and ratio of coffee beans based on past accumulated coffee bean blending data.

[0259] For example, if a customer selects 'fruit flavor emphasis' and 'nutty beans', the machine learning algorithm can calculate the optimal ratio of the bean combination, for example, "Ethiopia Yirgacheffe (fruit flavor) 60% : Brazil Santos (nutty) 40%".

[0260] That is, the above machine learning model can provide a blending recipe with a calculated blending ratio as output data.

[0261] Accordingly, in the embodiment, the processor (120) can automatically select and grasp the first and second beans corresponding to the blending recipe from among the plurality of dosing cups (13) placed on the shelf (1) according to the calculated blending recipe using the hand robot unit (2), grind them according to the blending ratio set in the grinder unit (3), and then extract them under optimal conditions in the dripper unit (4) to produce a customized blended drip coffee.

[0262] Through the method of the present embodiment, the present invention can provide high-quality blended drip coffee tailored to individual tastes with only brief user input, thereby increasing customer satisfaction while maintaining a convenient ordering environment.

[0264] In summary, the robot arm-based customized drip coffee manufacturing device and the drip coffee manufacturing system using the same according to the embodiment of the present invention have the effect of enabling the production of various types of drip coffee with high accuracy while reducing labor costs by accurately selecting coffee beans using a collaborative robot, extracting the selected beans at an accurate ratio and time based on an optimal movement path, and manufacturing drip coffee through an automated process.

[0265] In addition, the robot arm-based customized drip coffee manufacturing device and the drip coffee manufacturing system using the same, according to an embodiment of the present invention, produce blended drip coffee by blending two or more types of coffee beans according to consumer preferences, thereby responding to the demand for a wide range of flavors and aromas of drip coffee consumers, and contributing to strengthening the competitiveness of coffee shops and increasing customer satisfaction.

[0266] In addition, the robot arm-based customized drip coffee manufacturing device and the drip coffee manufacturing system using the same, according to an embodiment of the present invention, perform the entire coffee serving process—from dosing cup gripping, bean grinding, dripping, serving drip coffee to customers, to washing used dosing cups—using multiple collaborative robots configured with optimal movement paths. This provides the latest technology and innovative experience through robot technology, enhances aesthetic appeal, and simultaneously guarantees the precision and speed of the manufacturing process.

[0267] In addition, the robot arm-based customized drip coffee manufacturing device and the drip coffee manufacturing system using the same according to the embodiment of the present invention provide smart dripping control technology that dynamically adjusts optimal extraction conditions for each coffee bean in real time during the drip coffee manufacturing process, thereby not only dramatically improving the consistency of coffee extraction quality but also enabling the provision of coffee flavors optimized according to the condition of the coffee beans and environmental conditions, rather than uniformly, to customers, which has the effect of greatly increasing customer satisfaction.

[0268] The embodiments according to the present invention described above may be implemented in the form of program instructions that can be executed through various computer components and recorded on a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc., either individually or in combination. The program instructions recorded on the computer-readable recording medium may be those specifically designed and configured for the present invention or those known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. Hardware devices may be modified into one or more software modules to perform processing according to the present invention, and vice versa.

[0269] The specific embodiments described in this invention are examples and do not limit the scope of the invention in any way. For the sake of brevity of the specification, descriptions of prior electronic configurations, control systems, software, and other functional aspects of said systems may be omitted. Additionally, the connections of lines or connecting members between components shown in the drawings are illustrative of functional connections and / or physical or circuit connections, and may be replaced or additionally represented as various functional connections, physical connections, or circuit connections in actual devices. Furthermore, unless specifically stated as “essential,” “importantly,” etc., a component may not be strictly necessary for the application of the invention.

[0270] Furthermore, although the detailed description of the present invention has been explained with reference to preferred embodiments of the invention, those skilled in the art or those with ordinary knowledge in the relevant technical field will understand that various modifications and changes can be made to the invention without departing from the spirit and technical scope of the invention as set forth in the claims below. Accordingly, the technical scope of the present invention should not be limited to the contents described in the detailed description of the specification, but should be determined by the claims.

Claims

Claim 1 A shelf unit having a plurality of dosing cups arranged to store different types of coffee beans; a hand robot unit including at least one collaborative robot module that performs a drip coffee manufacturing process; a grinder unit that grinds the coffee beans stored in the dosing cups to produce ground coffee beans; a dripper unit that extracts drip coffee by passing hot water through the ground coffee beans; and a processor unit that controls the hand robot unit to perform the drip coffee manufacturing process; wherein the hand robot unit includes a gripper unit comprising: a grip box having a rotatable structure in which a motor is arranged; a grip including a grip upper part and a grip lower part whose spacing and angle are adjusted according to the shape of the object to be gripped; a grip connecting part connecting the grip and the grip box; and a horizontal grip rail formed on the grip box so that the grip connecting part can move left and right to adjust the position of the grip.It includes, wherein at least one of the upper grip and the lower grip is structured to be capable of individual rotation and front protrusion according to the shape of the object to be grasped, and the processor unit controls the hand robot unit to perform at least one of a grasping operation and a pressing operation by determining the gap, rotation angle, and protrusion of the grip based on data sensing the shape of the object to be grasped, and performs smart dripping control that dynamically controls extraction conditions by analyzing sensor data, wherein the smart dripping control comprises: 1) measuring the real-time extraction temperature inside the dripper unit through an extraction temperature sensor and acquiring sensor data including hot water temperature, extraction speed, coffee bean expansion amount, and extract color; 2) inputting the sensor data into a machine learning model to analyze the deviation from the target extraction condition and predicting the nozzle angle correction value, water flow speed correction value, hot water temperature correction value, and whether extraction is finished, respectively; 3) correcting the extraction conditions in real-time by adjusting the nozzle angle, valve opening, and water heater temperature of the hand robot unit according to the predicted correction values, stopping the extraction if the whether extraction is finished is above a threshold value, and otherwise, at a certain period, the above 1) to A robot arm-based customized drip coffee manufacturing device characterized by controlling dynamic extraction control that responds to changes in coffee bean type and environment by repeatedly performing the operation of 3). Claim 2 A robot arm-based customized drip coffee manufacturing device according to claim 1, wherein the processor unit maps a unique number for each different type of coffee bean and maps the position of each dosing cup within the shelf unit to which the unique number is mapped into position information in the form of a matrix corresponding to the unique number. Claim 3 A customized drip coffee manufacturing device based on a robot arm, wherein at least one collaborative robot module included in the hand robot part is a multi-joint module structure that adjusts the angle, direction, and position of the gripper part, and is equipped with a collision detection sensor for each joint module, and has a movement algorithm programmed to simultaneously perform different processes included in the drip coffee manufacturing process. Claim 4 A robot arm-based customized drip coffee making device according to claim 1, wherein the grip of the gripper part comprises a grip upper part and a grip lower part spaced apart from each other, the grip upper part and the grip lower part form a space having a predetermined curvature, and the spacing between the grip upper part and the grip lower part is narrowed or widened according to the shape of the object to be gripped. Claim 5 A robot arm-based customized drip coffee making device according to claim 1, wherein the grinder unit comprises a bean receiving unit into which coffee beans are fed, a grinder blade for grinding coffee beans, a coffee bean discharge port for discharging ground coffee beans, and an operating button for grinding coffee beans, and the processor unit controls the gripper unit such that at least one of the upper grip and lower grip included in the grip protrudes and rotates to perform a pressing operation for the operating button. Claim 6 A robot arm-based customized drip coffee making device according to claim 1, wherein the dripper part comprises a funnel-shaped dripper that fixes ground coffee beans and a filter, and a support frame that fixes the dripper to the top of a beverage cup. Claim 7 A robot arm-based customized drip coffee manufacturing device according to claim 1, wherein the processor unit monitors a real-time dripping state through at least one sensor included in a sensor system, analyzes at least one of the temperature, flow rate, expansion and color data of hot water obtained during dripping through a deep learning model learned for optimal extraction conditions for each type of coffee bean, and dynamically adjusts the gripper unit according to the analyzed result. Claim 8 A robot arm-based customized drip coffee making device, wherein, in claim 1, the shelf unit further comprises a beverage washing unit equipped with a water spray nozzle for automatically washing a used dosing cup or beverage cup, and the beverage washing unit automatically sprays water to wash the cup when a predetermined object is detected through an infrared sensor. Claim 9 A robot arm-based customized drip coffee manufacturing device according to claim 1, wherein the processor unit acquires orderer preference data input based on a kiosk, determines a first coffee bean and a second coffee bean corresponding to the acquired preference data among coffee bean information stored in memory, automatically calculates the blending ratio of the determined first coffee bean and second coffee bean using a blending ratio recommendation algorithm, and controls the hand robot unit to manufacture customized blended drip coffee according to the calculated blending ratio. Claim 10 delete Claim 11 A robot arm-based customized drip coffee manufacturing device according to claim 1, wherein the machine learning model receives sensor data and adopts a machine learning-based multi-output model to simultaneously predict a nozzle angle correction value, a water flow rate correction value, a hot water temperature correction value, and whether extraction is finished, and each parameter is optimized through hyperparameter tuning. Claim 12 A robot arm-based customized drip coffee making device according to claim 11, wherein the extraction termination is predicted in the form of a continuous value and is determined as extraction stoppage when it is greater than or equal to the threshold value, and the allowable error for the deviation of the threshold value, extraction temperature, and flow rate can be variably set. Claim 13 A shelf unit having a plurality of dosing cups arranged to store different types of coffee beans; a hand robot unit including at least one collaborative robot module that performs a drip coffee manufacturing process; a grinder unit that grinds the coffee beans stored in the dosing cups to produce ground coffee beans; a dripper unit that passes hot water through the ground coffee beans to extract drip coffee; a beverage serving unit that provides a beverage cup containing the drip coffee to a customer; a beverage washing unit that automatically washes used dosing cups or beverage cups; and a processor unit that controls the hand robot unit to perform the drip coffee manufacturing process.The processor unit includes, wherein the processor unit acquires order information through orderer input detected by the interface module, determines a target dosing cup containing the corresponding beans according to the bean information included in the acquired order information, controls the hand robot unit to move to and grasp the determined target dosing cup, controls the hand robot unit to move the grasped target dosing cup to the grinder unit to drop and grind a preset amount of beans, controls the hand robot unit to move the target dosing cup containing the ground beans after grinding to the dripper unit to perform dripping, controls the hand robot unit to provide the drip coffee completed according to the dripping to the beverage serving unit, controls the hand robot unit to scan and grasp the used dosing cup or beverage cup through an image sensor to perform automatic washing in the beverage washing unit, and performs smart dripping control that dynamically controls extraction conditions by analyzing sensor data when performing the dripping, wherein 1) through the extraction temperature sensor, the inside of the dripper unit A robot arm-based customized drip coffee manufacturing system characterized by: 1) measuring the extraction temperature in real time and acquiring sensor data including hot water temperature, extraction speed, coffee bean expansion amount, and extract color; 2) inputting the sensor data into a machine learning model to analyze deviations from target extraction conditions and predicting nozzle angle correction values, water flow speed correction values, hot water temperature correction values, and whether extraction is terminated, respectively; 3) adjusting the nozzle angle, valve opening, and water heater temperature of the hand robot unit according to the predicted correction values ​​to correct extraction conditions in real time; stopping extraction when the determination of whether extraction is terminated is above a threshold value, and otherwise repeating the operations of 1) to 3) at regular intervals to control dynamic extraction control in response to changes in coffee bean type and environment.

Images