Learning device, learning method, beverage supply system, and beverage supply robot control method

Abstract

A learning device (70) generates a trained model to be used for controlling a beverage supply robot (10) that autonomously moves within a facility and supplies a beverage to a user. The learning device (70) comprises: a data acquisition unit; and a model generation unit. The data acquisition unit acquires: learning data including state data representing the state of a user in a facility; and actual use data indicating the actual use of the beverage supply robot by the user. The model generation unit uses the learning data to generate a trained model for which: the user state data serves as input; and a use schedule serves as output, said schedule indicating the expected use of the beverage supply robot by the user.

Description

Learning device, learning method, beverage supply system, and beverage supply robot control method

[0001] The present disclosure relates to a learning device, a learning method, a beverage supply system, and a control method for a beverage supply robot.

[0002] Japanese Patent Laid-Open Publication No. 7-286 (Patent Document 1) discloses a tea dispensing system that uses an automatically traveling tea dispensing device to dispense tea to a user. In this tea dispensing system, the automatically traveling tea dispensing device includes a tea dispensing unit, a traveling device that carries the tea dispensing unit, and a battery, and uses the power stored in the battery to move to the user's location and dispense tea.

[0003] Japanese Unexamined Patent Publication No. 7-286

[0004] The tea-making system described above is configured so that, in response to a user's request, the tea-making device moves to the user's location and dispenses tea. As a result, there is a time lag between the user's request for tea and the user's receipt of the tea, depending on the time it takes for the tea-making device to move. If this time lag becomes too long, there is a concern that it will reduce user satisfaction.

[0005] The present disclosure has been made to solve such problems, and the purpose of the present disclosure is to realize a beverage supply system that can supply beverages at a time appropriate for the user using an autonomously mobile beverage supply robot.

[0006] A learning device according to one aspect of the present disclosure generates a trained model used to control a beverage dispenser robot that autonomously moves within a facility and dispenses beverages to users. The learning device includes a data acquisition unit and a model generation unit. The data acquisition unit acquires training data including status data representing a user's status within the facility and usage history data indicating the user's usage history of the beverage dispenser robot. The model generation unit uses the training data to generate a trained model that takes the user's status data as input and outputs a usage schedule indicating the user's planned use of the beverage dispenser robot.

[0007] A beverage supply system according to one aspect of the present disclosure includes a beverage supply robot that autonomously moves within a facility to supply beverages to users, and a control device that controls the beverage supply robot. The control device includes a data acquisition unit, an inference unit, and a movement path determination unit. The data acquisition unit acquires status data representing a user's status within the facility. The inference unit infers a user's usage schedule from the user's status data input from the data acquisition unit, using a trained model for inferring a usage schedule indicating a planned use of the beverage supply robot from the status data. The movement path determination unit determines a movement path for the beverage supply robot according to the usage schedule output from the inference unit.

[0008] According to the present disclosure, a beverage supply system can be realized that uses an autonomously mobile beverage supply robot to supply beverages at a time that is appropriate for the user.

[0009] 1 is a block diagram showing the configuration of a beverage supply system according to embodiment 1. FIG. 2 is a diagram showing an example configuration of a beverage supply robot and a station. FIG. 3 is a diagram for explaining the basic operation of the beverage supply system. FIG. 4 is a block diagram showing an example hardware configuration of a learning device. FIG. 5 is a schematic diagram showing an example functional configuration of a learning device. FIG. 6 is a diagram showing an example of behavioral history data and usage history data. FIG. 7 is a flowchart showing an example of learning processing executed by the learning device. FIG. 8 is a diagram showing an example hardware configuration of a server. FIG. 9 is a schematic diagram showing an example functional configuration of a server. FIG. 10 is a diagram showing a movement route based on usage schedules of multiple users. FIG. 11 is a flowchart showing an example of beverage supply processing executed by the server. FIG. 12 is a block diagram showing the configuration of a beverage supply system according to embodiment 2. FIG. 13 is a schematic diagram showing an example functional configuration of a learning device. FIG. 14 is a diagram showing an example of biometric data and usage history time. FIG. 15 is a schematic diagram showing an example functional configuration of a server. FIG. 16 is a flowchart showing an example of beverage supply processing executed by the server.

[0010] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are designated by the same reference numerals, and description thereof will not be repeated.

[0011] [Embodiment 1] <Configuration of beverage supply system> Fig. 1 is a block diagram showing the configuration of a beverage supply system according to embodiment 1. Beverage supply system 100 is a system that supplies beverages to users present in a facility by a beverage supply robot that autonomously moves within the facility.

[0012] As shown in FIG. 1 , the beverage supply system 100 includes a beverage supply robot 10 , a station 20 , a plurality of wireless communication devices 30 , and a server 40 .

[0013] The beverage supply robot 10 is an autonomous mobile beverage supply device. The beverage supply robot 10 includes a beverage supply unit 12 and a drive unit 13. The beverage supply robot 10 is equipped with a battery, and by using power stored in the battery to operate the beverage supply unit 12 and the drive unit 13, the beverage supply robot 10 can move within a facility and supply a low-temperature beverage to a user at a predetermined set temperature. In the example of FIG. 1 , the beverage supply robot 10 is configured to supply hot tea, hot water, and hot coffee as beverages.

[0014] The beverage supply robot 10 is provided with an operation unit 15. The operation unit 15 includes a display unit that displays the types of beverages that the beverage supply robot 10 can supply, and an operation button for dispensing the beverage.

[0015] The beverage dispensing robot 10 is provided with a wireless communication device 11. The wireless communication device 11 outputs a signal for detecting the position of the beverage dispensing robot 10 using a communication method conforming to the BLE (Bluetooth Low Energy; "Bluetooth" is a registered trademark) communication standard. Instead of the BLE communication standard, a communication method conforming to the UWB (Ultra Wide Band) communication standard may be used. The wireless communication device 11 also transmits a signal indicating an ID for identifying the beverage dispensing robot 10 to the server 40 using a communication method conforming to a wireless communication standard such as LTE (Long Term Evolution).

[0016] The multiple wireless communication devices 30 are installed, for example, at an appropriate distance from each other on the ceiling 35, and receive signals transmitted from the beverage dispensing robot 10 and detect the reception strength thereof using a communication method conforming to the same communication standard as the wireless communication device 11 of the beverage dispensing robot 10. The position of the beverage dispensing robot 10 within the facility can be detected from the reception strength of each wireless communication device 30. The wireless communication devices 30 output the reception strength of the signals received from the beverage dispensing robot 10 to the server 40. The wireless communication devices 30 may be installed on a wall.

[0017] The station 20 is installed in the facility and accommodates the beverage dispensing robot 10 while it is waiting to move. The station 20 is equipped with a charger for charging the battery mounted on the beverage dispensing robot 10. The beverage dispensing robot 10 charges its battery while waiting in the station 20. The charging method in the station 20 may be contact type or non-contact type.

[0018] The server 40 manages and controls the beverage dispensing robot 10. The server 40 receives the reception strength of the signal received by the wireless communication devices 30 from the wireless communication devices 30, and detects the position of the beverage dispensing robot 10 within the facility from the reception strength of each wireless communication device 30.

[0019] The server 40 is connected to a communication network NW. The communication network NW may be appropriately selected from, for example, the Internet, a wireless communication network, a mobile communication network, a telephone network, a dedicated network, etc. Connected to the communication network NW are a facility management server 50 that manages the use of the facility, a user terminal 60 that is a communication terminal owned by a user, and a learning device 70.

[0020] By communicating with the facility management server 50, the server 40 obtains information regarding the routes that the beverage supply robot 10 can move within the facility, information regarding the location of the station 20 within the facility, and temperature information indicating the temperature within the facility.

[0021] The user terminal 60 is, for example, a mobile terminal such as a smartphone or tablet, or a personal computer. A user who wishes to have a beverage dispensed can output a beverage dispense instruction to the server 40 using their own user terminal 60. The server 40 receives the beverage dispense instruction from the user by communicating with the user terminal 60. When the server 40 receives the beverage dispense instruction from the user, the server 40 exchanges data with the user terminal 60 that output the dispense instruction to obtain location information of the user terminal 60 within the facility, i.e., location information of the user within the facility.

[0022] The learning device 70 creates a trained model to be used in the beverage supply system 100. The trained model is created so that, when status data representing the status of a user in a facility is input, a usage schedule indicating the user's planned use of the beverage supply robot 10 is output. The server 40 can acquire the trained model created by the learning device 70 via the communication network NW. The server 40 can determine the usage schedule of the beverage supply robot 10 by using the acquired trained model. The server 40 can determine the movement route of the beverage supply robot 10 based on the determined user's usage schedule.

[0023] The server 40 uses the location information of the beverage supply robot 10, various information obtained from the facility management server 50, and the user's location information obtained from the user terminal 60 to perform various controls on the beverage supply robot 10, such as movement and beverage supply, and provides the necessary information to the beverage supply robot 10.

[0024] <Configuration Example of Beverage Dispensing Robot> Fig. 2 is a diagram showing a configuration example of the beverage dispensing robot 10 and the station 20. As shown in Fig. 2, the beverage dispensing robot 10 includes a wireless communication device 11, a beverage dispensing unit 12, a driving unit 13, a camera 14, an operation unit 15, a battery 16, and a control unit 18.

[0025] 1, the wireless communication device 11 uses a communication method conforming to, for example, the BLE communication standard to transmit a signal for detecting the position of the beverage dispensing robot 10. The wireless communication device 11 also uses a communication method conforming to, for example, a wireless communication standard such as LTE to transmit to the server 40 various information such as an ID for identifying the beverage dispensing robot 10, a supply start signal indicating that the beverage dispensing robot 10 has started, and a supply end signal indicating that the beverage dispensing has ended.

[0026] The beverage supply unit 12 is a component for supplying a beverage to a user. The beverage supply unit 12 includes, for example, a liquid storage tank 120, a liquid supply path 122, a temperature adjustment unit 124, a temperature sensor 126, a stirring unit 128, a discharge port 130, and an opening / closing mechanism 132. The beverage supply unit 12 can be operated by receiving power from a battery 16.

[0027] Liquid storage tank 120 stores a liquid such as water. The temperature of the liquid stored in liquid storage tank 120 is adjusted by temperature adjustment unit 124. In the example of Fig. 1, temperature adjustment unit 124 includes a heater and heats the liquid stored in liquid storage tank 120. Temperature sensor 126 detects the temperature of the liquid (hot water) stored in liquid storage tank 120 and outputs a signal indicating the detected value to control unit 18.

[0028] The liquid (hot water) stored in the liquid storage tank 120 is supplied to the stirring unit 128 through the liquid supply path 122. The stirring unit 128 produces tea or coffee by stirring the liquid (hot water) introduced from the liquid supply path 122 with tea leaf powder or coffee powder supplied from a grinding unit (not shown).

[0029] The stirring unit 128 is provided with a discharge port 130 for discharging the prepared beverage. The opening / closing mechanism 132 is a component for opening and closing the discharge port 130. In response to a user operating the operation button on the operation unit 15, the opening / closing mechanism 132 opens the discharge port 130 for a certain period of time. This causes a predetermined amount of beverage to be discharged from the stirring unit 128 through the discharge port 130 and supplied to the user.

[0030] The camera 14 captures an image of the surroundings of the beverage supply robot 10 and outputs the captured image to the control unit 18. Instead of the camera 14, a laser rangefinder or the like may be provided that measures the distance between the beverage supply robot 10 and an object present in the vicinity of the beverage supply robot 10.

[0031] The driving unit 13 generates a driving force for moving the beverage dispensing robot 10. The driving unit 13 includes, for example, wheels for moving the beverage dispensing robot 10 and a motor for driving the wheels. The driving unit 13 (motor) can be operated by receiving a supply of power from the battery 16.

[0032] The control unit 18 controls the entire beverage dispensing robot 10. Although not shown in the figure, the control unit 18 has, as its main components, a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), and an I / F (Interface) device. These components exchange various data via a communication bus. The CPU loads a program stored in the ROM into the RAM and executes it. The program stored in the ROM describes the processing to be executed by the beverage dispensing robot 10. Note that the processing is not limited to being performed by software, but can also be performed by dedicated hardware (electronic circuits).

[0033] The control unit 18 controls the driving unit 13 so that the beverage dispensing robot 10 moves autonomously based on the captured image from the camera 14. The control unit 18 also controls the temperature adjusting unit 124 (heater) based on the detection signal from the temperature sensor 126 so that a beverage at an appropriate temperature according to the set supply temperature is dispensed from the dispensing outlet 130.

[0034] The battery 16 supplies power for operating the beverage supply unit 12, the drive unit 13, and other devices of the beverage supply robot 10. The beverage supply robot 10 can charge the battery 16 while waiting in the station 20.

[0035] <Configuration of Station> Station 20 includes a charger 22, a wireless communication device 24, and a control unit 26. Wireless communication device 24 receives signals transmitted from beverage dispensing robot 10 and detects the reception strength thereof using a communication method conforming to the same communication standard as wireless communication device 11 of beverage dispensing robot 10. Control unit 26 can determine whether beverage dispensing robot 10 is located at station 20 based on the reception strength of wireless communication device 24.

[0036] When it is determined that the beverage dispensing robot 10 is located at the station 20, the control unit 26 activates the charger 22. The charger 22 converts AC power supplied from an external power source (e.g., a commercial power source) into power that can be charged into the battery 16 and supplies the power to the battery 16. As a result, while the beverage dispensing robot 10 is waiting at the station 20, the battery 16 is charged using the power supplied from the charger 22.

[0037] In the beverage supply system 100 according to the first embodiment, the server 40 , together with the control unit 18 of the beverage supply robot 10 and the control unit 26 of the station 20 , constitutes a “control device” that controls the beverage supply robot 10 .

[0038] <Basic Operation of Beverage Supply System> Next, a basic operation of the beverage supply system 100 according to the first embodiment will be described.

[0039] Fig. 3 is a diagram for explaining the basic operation of the beverage supply system 100. Fig. 3 shows a schematic plan view of a floor in a facility as viewed from the ceiling 35. In the following explanation, it is assumed that the facility is an office building.

[0040] A plurality of pieces of equipment, such as desks, chairs, and racks, are arranged on the floor. Also arranged on the floor are stations 20. The number of stations 20 arranged on the floor may be one or more.

[0041] Areas on the floor that are passable and areas that are not passable for the beverage supply robot 10 are designated. Based on these designated areas, the facility management server 50 sets a route that the beverage supply robot 10 can move along.

[0042] As shown in Fig. 3, the beverage supply robot 10 is waiting at the station 20. As described above, the beverage supply robot 10 charges the battery 16 while waiting at the station 20. While waiting at the station 20, the beverage supply robot 10 controls the temperature adjustment unit 124 to maintain the temperature of the liquid in the liquid storage tank 120 at the set supply temperature. This allows the beverage supply robot 10 to supply a beverage at an appropriate temperature according to the set supply temperature to a user even while waiting at the station 20.

[0043] While working or taking a break in the facility, each user can use their own user terminal 60 (e.g., a smartphone) to issue a beverage supply instruction to the beverage supply system 100. When a supply instruction is received from the user terminal 60, the server 40 exchanges data with the user terminal 60 that sent the instruction, thereby obtaining location information of the user who owns the user terminal 60.

[0044] The server 40 determines a movement route for the beverage supply robot 10 using the acquired user's location information and information stored in a storage device about routes that the beverage supply robot 10 can travel. Specifically, the server 40 determines a beverage supply point from the user's location information. The server 40 then searches for candidate movement routes from the station 20 (the current location) to the supply point by referring to the information about routes that the beverage supply robot 10 can travel. The server 40 determines, for example, the candidate movement route that has the shortest travel time as the movement route for the beverage supply robot 10. FIG. 3 shows a beverage supply point SP1 and a movement route Rt1 for the beverage supply robot 10, which are determined based on the user's location information. The server 40 transmits information about the determined supply point SP1 and movement route Rt1 to the beverage supply robot 10.

[0045] When the beverage supply robot 10 receives information about the supply point SP1 and the movement route Rt1 from the server 40, it moves along the movement route Rt1 to the supply point SP1. As a result, the beverage supply robot 10 that has arrived at the supply point SP1 can supply a beverage to the user who issued the supply instruction.

[0046] In addition to the configuration for moving to a user and dispensing a beverage in response to a dispensing instruction from the user as described above, the beverage dispensing system 100 according to the first embodiment is also configured to use a trained model to infer a usage schedule for the user's planned use of the beverage dispensing robot 10 from the user's state within the facility. In this configuration, the beverage dispensing robot 10 moves autonomously within the facility and dispenses beverages to the user based on the inferred usage schedule.

[0047] <Configuration of Learning Device> The trained model used in the beverage supply system 100 according to the first embodiment is created by the learning device 70. Next, the configuration of the learning device 70 will be described with reference to Figs. 4 to 6 .

[0048] 4 is a block diagram showing an example of the hardware configuration of the learning device 70. The learning device 70 receives input of the state of a user in a facility and creates a trained model for inferring a usage schedule in which the user plans to use the beverage dispensing robot 10.

[0049] 4, the learning device 70 includes a CPU 71, a RAM 72, a ROM 73, an I / F device 74, a storage device 75, and a drive 76. The CPU 71, RAM 72, ROM 73, I / F device 74, storage device 75, and drive 76 exchange various data via a communication bus 77.

[0050] The CPU 71 loads a program stored in the ROM 73 into the RAM 72 and executes the program. The program stored in the ROM 73 describes the processes to be executed by the server 40. Although Fig. 4 illustrates an example of a configuration with a single CPU, the server 40 may be configured to have multiple CPUs.

[0051] The I / F device 74 is an input / output device for exchanging signals and data with the server 40 and the facility management server 50. The I / F device 74 receives status data indicating the status of the user in the facility from the facility management server 50. The I / F device 74 receives usage history data indicating the usage history of the beverage dispensing robot 10 from the server 40.

[0052] The storage device 75 is a storage for storing various types of information, and stores various types of information for creating a trained model. The various types of information include a learning program 751 executed by the CPU 71, training data 752 used for training, and a trained model 753 created by executing the learning program 751. The storage device 75 is, for example, a hard disk drive, a solid state drive, or the like.

[0053] The drive 76 is, for example, a CD drive, a DVD drive, or the like, and is a drive device for reading a program stored in the storage medium 80. The type of the drive 76 may be selected appropriately depending on the type of the storage medium 80. The learning program 751 and the learning data 752 may be stored in the storage medium 80.

[0054] The storage medium 80 is a medium that stores information such as a recorded program by electrical, magnetic, optical, mechanical, or chemical action so that the information can be read by a computer or other device, machine, etc. The learning device 70 may acquire the learning program 751 and the learning data 752 from the storage medium 80.

[0055] Note that the hardware configuration example shown in FIG. 4 can have components omitted, replaced, or added as appropriate depending on the embodiment. For example, the learning device 70 may include multiple hardware processors. The hardware processor may be configured with a microprocessor, an FPGA (Field-Programmable Gate Array), or the like. The learning device 70 may be configured with multiple information processing devices. The learning device 70 may be an information processing device designed specifically for the service provided, or may be a general-purpose server or PC (Personal Computer), etc.

[0056] (Example of Functional Configuration of Learning Device) Fig. 5 is a schematic diagram showing an example of the functional configuration of the learning device 70. Each functional configuration shown in Fig. 5 is typically realized by the CPU 71 of the learning device 70 executing a learning program 751.

[0057] As shown in FIG. 5, the learning device 70 includes, as its mechanical configuration, a learning data acquisition unit 700, a model generation unit 702, and a trained model storage unit 703.

[0058] The learning data acquisition unit 700 acquires learning data 752 stored in the storage device 75. The learning data 752 includes status data that indicates the status of the user within the facility. In the first embodiment, the status data uses behavior history data 710 that indicates the user's behavior and the location of the behavior within the facility. The learning data 752 further includes usage history data 720 that indicates the user's usage history of the beverage dispensing robot 10.

[0059] 6 is a diagram showing an example of behavior history data 710 and usage history data 720. As shown in FIG. 6, the behavior history data 710 is data indicating the content and location of a user's behavior within the facility. The behavior history data 710 includes information indicating when, where, and what behavior the user performed. When multiple users are present within the facility and each user is able to use the beverage dispensing robot 10, the behavior history data 710 is generated for each user.

[0060] 6 shows behavioral history data 710 for a user with a user ID of "U1." The behavioral history data 710 includes, for example, information about the user's daily behavior and the location of the behavior. Specifically, the information includes, for a given day, what time the user arrives at work, what time the user leaves work, what work the user did at what location in the facility during what time period, and what location the user took a break during what time period. As shown in FIG. 6, the behavioral history data 710 includes information about the user's behavior and the location of the behavior for several days (for example, one week).

[0061] The user's location information within the facility, which is included in the behavioral history data 710, can be obtained, for example, by the facility management server 50 exchanging data with the user terminal 60 owned by the user. Alternatively, the facility management server 50 can obtain the user's location information based on a signal from a card reader that communicates with a security card (e.g., an ID card with an RFID (Radio Frequency Identifier) ​​function) owned by the user for authenticating identification information.

[0062] 6 further shows usage history data 720 of the beverage dispensing robot 10 for a user with a user ID of "U1." The usage history data 720 includes information on the time (beverage supply time) and location (beverage supply point) at which the user used the beverage dispensing robot 10 for each day. Specifically, the information includes information on what time and where the user received a beverage from the beverage dispensing robot 10 on a given day. The usage history data 720 includes information on the beverage supply time and beverage supply point for the same day as the behavior history data 710.

[0063] Usage history data 720 can be acquired by exchanging data with server 40. As described in Fig. 3, when server 40 receives a supply instruction from user terminal 60, it determines a beverage supply point based on location information of the user who owns user terminal 60, and controls beverage supply robot 10 to move to the supply point and supply the beverage to the user. Learning device 70 acquires from server 40 the ID of the user to whom beverage supply robot 10 supplied a beverage, the point where the beverage was supplied to the user, and information regarding the time of supply.

[0064] As described above, within the facility, each user can receive a beverage by issuing a beverage supply instruction using the user terminal 60 at any timing. Therefore, for example, when a user wants to recover from fatigue while working or when a user wants to relax during a break, the user can receive a beverage from the beverage supply robot 10 by issuing a supply instruction. As such, the timing of issuing a supply instruction (the timing of requesting beverage supply) is subjectively determined by the user, and therefore the timing at which each user issues a supply instruction varies.

[0065] In the first embodiment, learning data 752 is generated for each user by associating behavior history data 710 with usage history data 720. That is, the learning data 752 associates, for each user, information about what behavior the user performed at what time of day and at what location (behavior history) with information about when and at what location the user received a beverage (usage history).

[0066] 5 , the model generation unit 702 learns a usage schedule indicating the user's planned use of the beverage dispensing robot 10 based on the learning data 752 from the learning data acquisition unit 700. The behavioral history data 710 is used as input data, and the usage history data 720 is used as training data (correct answer data). That is, the model generation unit 702 generates a trained model 753 for inferring the user's usage schedule of the beverage dispensing robot 10 from the user's behavioral history data 710 and usage history data 720. The usage schedule includes information regarding the planned locations and planned times at which beverages are to be dispensed to the user.

[0067] In this way, learning device 70 is used to learn the usage schedule of beverage dispensing robot 10, and as shown in FIG. 1 , learning device 70 may be connected to server 40 of beverage dispensing system 100 via communication network NW and may be a separate device from server 40. Alternatively, learning device 70 may be built into server 40. Furthermore, learning device 70 may exist on a cloud server.

[0068] The learning algorithm used by the model generation unit 702 may be a known algorithm such as supervised learning, unsupervised learning, reinforcement learning, etc. As an example, a case where a neural network is applied will be described.

[0069] The model generation unit 702 learns the usage schedule of the beverage dispensing robot 10, for example, by so-called supervised learning according to a neural network model. Here, supervised learning refers to a technique in which a learning device is provided with sets of data consisting of inputs and results (labels), and the learning device learns the features contained in the learning data and infers the results from the inputs. Note that a neural network is composed of an input layer consisting of multiple neurons, an intermediate layer (hidden layer) consisting of multiple neurons, and an output layer consisting of multiple neurons.

[0070] In the first embodiment, the neural network learns the usage schedule of each user of the beverage supply robot 10 by so-called supervised learning in accordance with the learning data 752 created based on a combination of the behavioral history data 710 and usage history data 720 of each user acquired by the learning data acquisition unit 700.

[0071] The model generation unit 702 performs the above-described learning to generate a trained model 753. The trained model storage unit 703 stores the trained model 753 generated by the model generation unit 702.

[0072] 7 is a flowchart showing an example of a learning process executed by the learning device 70. Each step of this flowchart is typically executed by the CPU 71 of the learning device 70.

[0073] 7 , in step S01, learning device 70 (learning data acquisition unit 700) acquires learning data 752. In S01, learning device 70 communicates with facility management server 50 to acquire behavior history data 710 indicating the behavior history of each user within the facility. Learning device 70 further communicates with server 40 to acquire usage history data 720 indicating the usage history of beverage dispensing robot 10 by each user. Note that learning device 70 only needs to input behavior history data 710 and usage history data 720 in association with each other for each user, and may acquire behavior history data 710 and usage history data 720 at different times.

[0074] Next, in step S02, the learning device 70 (model generation unit 702) learns the usage schedule of each user's beverage supply robot 10 by so-called supervised learning in accordance with learning data 752 created based on a combination of each user's behavioral history data 710 and usage history data 720 acquired by the learning data acquisition unit 700, and generates a learned model.

[0075] In step S03, the learning device 70 (trained model storage unit 703) stores the trained model 753 generated by the model generation unit 702.

[0076] <Configuration of Server> The server 40 is configured to infer the usage schedule of each user's beverage dispensing robot 10 by utilizing the trained model 753 generated by the learning device 70. In other words, the server 40 constitutes an "inference device." Next, the configuration of the server 40 will be described with reference to Figures 8 to 10.

[0077] (Example of Hardware Configuration of Server) Fig. 8 is a diagram showing an example of the hardware configuration of the server 40. As shown in Fig. 8, the server 40 is configured to include a CPU 41, a RAM 42, a ROM 43, an I / F device 44, and a storage device 45. The CPU 41, RAM 42, ROM 43, I / F device 44, and storage device 45 exchange various types of data via a communication bus 46.

[0078] The CPU 41 loads a program stored in the ROM 43 into the RAM 42 and executes the program. The program stored in the ROM 43 describes the processes to be executed by the server 40. Although Fig. 8 illustrates an example of a configuration with a single CPU, the server 40 may be configured to have multiple CPUs.

[0079] The I / F device 44 is an input / output device for exchanging signals and data with the wireless communication device 30, the beverage dispensing robot 10, the facility management server 50, the user terminal 60, and the learning device 70. The I / F device 44 receives, from the wireless communication device 30, the reception strength of a signal received by the wireless communication device 30. The I / F device 44 also receives, from the beverage dispensing robot 10, an ID for identifying the beverage dispensing robot 10, a signal indicating the start / end of beverage dispensing by the beverage dispensing robot 10, and the like, using a communication method conforming to a wireless communication standard such as LTE. The I / F device 44 also receives a trained model 753 from the learning device 70.

[0080] The memory device 45 is a storage for storing various types of information, and stores various types of information for managing and controlling beverage supply by the beverage supply robot 10. The various types of information include information about the beverage supply robot 10, information about the facility (information about the routes that the beverage supply robot 10 can travel, location information of the stations 20, temperature information within the facility, etc.), location information of the user, and location information of the beverage supply robot 10. The various types of information further include a program 451 executed by the CPU 41 and a trained model 753 created by the learning device 70. The memory device 75 is, for example, a hard disk drive, a solid state drive, etc.

[0081] (Example of Functional Configuration of Server) Fig. 9 is a schematic diagram showing an example of the functional configuration of the server 40. Each functional configuration shown in Fig. 9 is typically realized by the CPU 41 of the server 40 executing a program 451.

[0082] As shown in FIG. 9, the server 40 includes, as its mechanical configuration, an inference data acquisition unit 400, an inference unit 402, a trained model storage unit 403, and a travel route determination unit 404.

[0083] The inference data acquisition unit 400 acquires an action schedule 410 for each user by communicating with the facility management server 50. The action schedule 410 is data indicating the planned content and location of the user's actions within the facility. The action schedule 410 includes information indicating when, where, and what actions the user will take. The action schedule 410 is generated for each user who is allowed to use the beverage dispensing robot 10.

[0084] For example, the activity schedule 410 includes information about the user's planned activities and locations for each day, such as what time the user is scheduled to arrive at work, what time the user is scheduled to leave work, what work the user is scheduled to do at what location in the facility, and what time the user is scheduled to take a break.

[0085] The inference unit 402 uses the trained model 753 stored in the trained model storage unit 403 to infer a usage schedule 420 indicating each user's planned use of the beverage supply robot 10. Specifically, the inference unit 402 inputs the behavior schedule 410 of each user acquired by the inference data acquisition unit 400 into the trained model 753. The inference unit 402 acquires the usage schedule 420 of each user's beverage supply robot 10 from the trained model 753. This usage schedule 420 is information regarding the planned use of the beverage supply robot 10, and includes information regarding the planned location and planned time of beverage supply for the user. The usage schedule 420 is generated for each user in association with the behavior schedule 410. The usage schedule 420 of each user is provided to the movement route determination unit 404.

[0086] The movement path determination unit 404 determines the movement path of the beverage supply robot 10 according to each user's usage schedule 420 provided by the inference unit 402 so that the beverage supply robot 10 can move to each user and supply beverages to them.

[0087] For example, assume that it is inferred from the daily activity schedule 410 of a user with user ID "U1" that the user plans to use the beverage supply robot 10 on that day. In this case, the movement path determination unit 404 determines the movement path of the beverage supply robot 10 using the planned beverage supply point included in the user's usage schedule 420 and information about routes that the beverage supply robot 10 can travel, stored in the storage device 45. For example, the server 40 searches for candidate movement paths from the station 20 to the planned beverage supply point by referring to the information about routes that the beverage supply robot 10 can travel. The server 40 determines the candidate movement path that has the shortest travel time from among the candidate movement paths as the movement path of the beverage supply robot 10.

[0088] Server 40 transmits to beverage supply robot 10 information relating to the scheduled supply point and scheduled supply time, and information relating to the movement route of beverage supply robot 10. Upon receiving the information from server 40, control unit 18 of beverage supply robot 10 controls drive unit 13 and beverage supply unit 12 (temperature adjustment unit 124) so ​​that a beverage at an appropriate temperature according to the set supply temperature can be supplied to the user at the scheduled supply time.

[0089] Specifically, the control unit 18 sets the time at which the beverage supply robot 10 departs from the station 20 (hereinafter also referred to as the "movement start time") so that the beverage supply robot 10 arrives at the scheduled supply point at the scheduled supply time. Here, the control unit 18 uses information about the movement route to calculate the movement time required for the beverage supply robot 10 to arrive at the scheduled supply point from the station 20. This movement time can be calculated by dividing the distance of the movement route by a predetermined movement speed of the beverage supply robot 10. Then, the control unit 18 sets the movement start time to a time that is earlier than the scheduled supply time by the movement time.

[0090] When the movement start time arrives, the beverage dispensing robot 10 departs from the station 20 and moves along the movement route to the scheduled supply point. During movement, the control unit 18 executes intermittent operation of the temperature adjustment unit 124 to maintain the liquid temperature at the set supply temperature. After the movement start time, the supply of power from the station 20 to the battery 16 is cut off, and the remaining capacity of the battery 16 gradually decreases due to the operation of the drive unit 13 and the temperature adjustment unit 124.

[0091] When the beverage supply robot 10 arrives at the scheduled supply point at the scheduled supply time, the control unit 18 stops the operation of the drive unit 13. Even after arrival at the scheduled supply point, the control unit 18 executes intermittent operation of the temperature adjustment unit 124 to maintain the liquid temperature at the set supply temperature.

[0092] When the supply of the beverage to the user is completed, the beverage supply robot 10 operates the drive unit 13 again to move from the supply point to the station 20. At this time, the beverage supply robot 10 can return to the original station 20 along the same movement route as the movement route to the supply point. Alternatively, the beverage supply robot 10 may acquire, from the server 40, new information regarding a movement route to another station 20 closest to the supply point, and return to the other station 20 along the acquired movement route.

[0093] When the beverage supply robot 10 returns to the station 20, the control unit 18 stops the operation of the drive unit 13. The beverage supply robot 10 receives power again from the station 20 and charges the battery 16. This restores the remaining capacity of the battery 16.

[0094] In this way, the beverage supply robot 10 moves to the user autonomously, without waiting for a supply instruction from the user, according to the usage schedule 420 inferred from the user's behavior schedule 410. This allows the beverage supply robot 10 to be positioned near the user in a situation where it is estimated that the user will desire a beverage. This allows the user to receive a beverage promptly from the beverage supply robot 10.

[0095] The server 40 can also determine the movement route of the beverage supply robot 10 so that the beverage supply robot 10 visits multiple users based on the usage schedules 420 of the multiple users. FIG. 10 schematically shows a movement route Rt2 determined based on the usage schedules 420 of the multiple users. The example in FIG. 10 shows the movement route Rt2 when the scheduled supply locations of three users working on the same floor are close to each other and the scheduled supply times are close to each other. The beverage supply robot 10 moves in a circular movement between the three users according to the movement route Rt2. This allows the beverage supply robot 10 to supply beverages to each user at an appropriate time.

[0096] 11 is a flowchart showing an example of a beverage supply process executed by the server 40. Each step of this flowchart is typically executed by the CPU 41 of the server 40.

[0097] 11 , in step S11, the server 40 determines whether or not a supply instruction has been received from the user terminal 60. If a supply instruction has been received (YES in S11), the server 40 proceeds to step S12 and exchanges data with the facility management server 50 to obtain location information of the user who owns the user terminal 60.

[0098] Next, in step S13, the server 40 determines a beverage supply point using the acquired user location information. Furthermore, in step S14, the server 40 determines a movement route for the beverage supply robot 10 using the determined supply point and information related to a route along which the beverage supply robot 10 can move, which is stored in the storage device 45. In step S15, the server 40 transmits information related to the supply point and the movement route to the beverage supply robot 10.

[0099] When the beverage supply robot 10 receives information about the supply point and the travel route from the server 40, it moves along the travel route to the supply point. As a result, the beverage supply robot 10 that has arrived at the supply point can supply a beverage to the user who issued the supply instruction.

[0100] Returning to step S11, if the supply instruction has not been received (NO in S11), the server 40 enters a standby state for receiving a supply instruction. In step S16, the server 40 (the inference data acquisition unit 400) communicates with the facility management server 50 to acquire the activity schedule 410 of each user in the facility.

[0101] Next, in step S17, the server 40 (inference unit 402) inputs each user's behavior schedule 410 into the trained model 753, thereby inferring each user's usage schedule 420 of the beverage supply robot 10.

[0102] In step S18, the server 40 (inference unit 402) outputs the inferred usage schedule 420 for each user.

[0103] In step S19, the server 40 (movement route determination unit 404) determines a movement route for the beverage supply robot 10 based on the output usage schedule 420 of each user. In step S20, the server 40 (movement route determination unit 404) transmits information about the determined movement route and information about the planned beverage supply point and planned beverage supply time included in the usage schedule 420 to the beverage supply robot 10.

[0104] When the beverage supply robot 10 receives information regarding the planned supply point, planned supply time, and travel route from the server 40, the beverage supply robot 10 departs from the station 20 so as to arrive at the planned supply point at the planned supply time, and travels along the travel route to the planned supply point. In this way, the beverage supply robot 10 that has arrived at the planned supply point can supply a beverage to a user who requests a beverage.

[0105] As described above, according to the first embodiment, beverage supply robot 10 moves autonomously to the user without waiting for a supply instruction from the user, in accordance with usage schedule 420 inferred from user behavior schedule 410. This makes it possible to realize beverage supply system 100 that can use beverage supply robot 10 to supply a beverage at a timing appropriate for the user.

[0106] [Embodiment 2] <Configuration of beverage supply system> Figure 12 is a block diagram showing the configuration of a beverage supply system according to embodiment 2. As shown in Figure 1, beverage supply system 100 includes beverage supply robot 10, station 20, a plurality of wireless communication devices 30, server 40, and sensor 90. Beverage supply system 100 according to embodiment 2 is configured by adding sensor 90 to beverage supply system 100 according to embodiment 1.

[0107] The sensor 90 detects biometric information of a user present in the facility. The biometric information includes information indicating the state of the living body and information indicating physical activity and movement. For example, the biometric information includes the user's eye movement (eye movement, blink frequency, pupil diameter, etc.), arm (particularly hand) movement, pulse rate, heart rate, brain waves, sweating, or temperature of peripheral body parts. Any of these pieces of biometric information can be detected using a well-known contact or non-contact sensor. The sensor 90 is typically a wearable device worn by the person or a camera.

[0108] One example of sensor 90 is a camera installed within the facility. The camera is positioned so that the user's eyes or arms (particularly, hands) are included in the capture range. The camera outputs captured video images to learning device 70 and server 40. The camera may also be installed in user terminal 60.

[0109] The movement of the user's eyes or arms can be detected by analyzing the video captured by the camera. Specifically, when the user is performing input work on the user terminal 60 (e.g., a personal computer), the movement of the user's eyes directed at the display of the user terminal 60 (e.g., eye movement) or the movement of the user's hands when operating the keyboard of the user terminal 60 (e.g., operation speed) can be measured by analyzing the captured video.

[0110] The user's pulse rate can be measured, for example, by photoplethysmography using a light-emitting diode and an optical sensor (such as a phototransistor).

[0111] A user's brain waves can be detected, for example, by near-infrared spectroscopy or an electroencephalograph. Near-infrared spectroscopy is a method of observing changes in the brain's blood volume using a light source and a light-receiving sensor. An electroencephalograph is a sensor that picks up minute electrical currents generated by brain activity using electrodes attached to the skull, amplifies them, and measures them as brain waves. Brain wave information includes data indicating basic rhythms, including frequency bands such as alpha waves and beta waves.

[0112] The peripheral part of the user is, for example, the wrist, finger, ear, nose, etc. The temperature of the peripheral part can be measured by a sensor attached to a part of the person's body.

[0113] The beverage supply system 100 of embodiment 2, like the beverage supply system 100 of embodiment 1, is configured to use a trained model to infer a usage schedule for the user's planned use of the beverage supply robot 10 from the user's state within the facility.

[0114] In the second embodiment, the user's biometric information detected by the sensor 90 is used as status data representing the user's condition within the facility. The user's biometric information changes along with the user's work efficiency. "Work efficiency" refers to the proportion of work that can be completed within a certain period of time. For example, when performing input work on a user terminal 60 (personal computer), the work efficiency corresponds to the ratio of the actual amount of work (e.g., the amount of characters entered) within a certain period of time to the standard amount of work that can be completed within that period of time.

[0115] For example, as the user's work efficiency decreases, the user's eye and / or hand movements become slower within a certain period of time. Furthermore, as the user's work efficiency decreases, the beta waves contained in the user's brain waves within a certain period of time decrease, and the user's level of alertness decreases. In this way, the user's biological information can be an index of the user's work efficiency.

[0116] <Configuration of Learning Device> The trained model used in the beverage supply system 100 according to the second embodiment is created by the learning device 70. Next, the configuration of the learning device 70 will be described with reference to Figs. 13 and 14 .

[0117] The learning device 70 according to the second embodiment uses biometric information of a user in a facility as input to create a trained model for inferring the user's planned time of use of the beverage dispensing robot 10. The hardware configuration of the learning device 70 is the same as that of the learning device 70 according to the first embodiment shown in FIG. 4, and therefore a description thereof will be omitted.

[0118] (Example of Functional Configuration of Learning Device) Fig. 13 is a schematic diagram showing an example of the functional configuration of the learning device 70. Each functional configuration shown in Fig. 13 is typically realized by the CPU 71 of the learning device 70 executing a learning program 751.

[0119] 13 , the learning device 70 includes, as a mechanical configuration, a learning data acquisition unit 700, a model generation unit 702, and a trained model storage unit 703. The learning device 70 according to the second embodiment differs from the learning device 70 according to the first embodiment in the learning data 752.

[0120] The learning data acquisition unit 700 acquires learning data 752 stored in the storage device 75. The learning data 752 includes biometric data 730 indicating changes in the biometric information of the user over time within the facility as status data representing the user's status within the facility. The learning data 752 further includes usage history time data 740 indicating the time when the user used the beverage dispensing robot 10.

[0121] Fig. 14 is a diagram showing an example of biometric data 730 and usage history time data 740. As shown in Fig. 14, biometric data 730 is data indicating changes over time in the biometric information of a user detected by sensor 90. Fig. 14 shows examples of the biometric information, such as the pulse rate, heart rate, facial image, and brain waves of the user. When multiple users are present in a facility and each user is able to use beverage dispensing robot 10, biometric data 730 is generated for each user.

[0122] 14 shows biometric data 730 of a user with a user ID of "U1." The biometric data 730 represents changes in the user's biometric information over time for each day. The biometric data 730 includes several days' worth of the user's biometric information.

[0123] 14 further shows actual usage time data 740 of the beverage dispensing robot 10 for a user with a user ID of "U1." The actual usage time data 740 includes information about the time each user used the beverage dispensing robot 10 on a given day. Specifically, the information includes information such as what time each user received a beverage from the beverage dispensing robot 10 on a given day. The actual usage time data 740 includes information about the actual usage time for the same day as the biometric data 730.

[0124] Biometric data 730 can be acquired by communicating with sensor 90. Historical usage time data 740 can be acquired by exchanging data with server 40. In the second embodiment, learning device 70 generates learning data 752 for each user by associating biometric data 730 with historical usage time data 740. That is, learning data 752 associates, for each user, information on how the user's biometric information has changed with information on when the user received a beverage.

[0125] 13 , the model generation unit 702 learns the user's planned use time of the beverage dispensing robot 10 based on the learning data 752 from the learning data acquisition unit 700. The biometric data 730 is used as input data, and the actual use time data 740 is used as training data (correct answer data). That is, the model generation unit 702 generates a trained model 753 for inferring the user's planned use time of the beverage dispensing robot 10 from the user's biometric data 730 and actual use time data 740. The inferred planned use time corresponds to the planned time a beverage is to be dispensed to the user.

[0126] As described in the first embodiment, the learning algorithm used by the model generation unit 702 can be a known algorithm such as supervised learning, unsupervised learning, reinforcement learning, etc. In the second embodiment, the neural network learns the usage time of each user of the beverage dispensing robot 10 by so-called supervised learning in accordance with the learning data 752 created based on a combination of the biometric data 730 and the actual usage time data 740 of each user acquired by the learning data acquisition unit 700.

[0127] The model generation unit 702 performs the above-described learning to generate a trained model 753. The trained model storage unit 703 stores the trained model 753 generated by the model generation unit 702.

[0128] <Functional Configuration of Server> The server 40 is configured to infer the planned time of use of the beverage dispensing robot 10 for each user, using the trained model 753 generated by training by the training device 70. Next, the configuration of the server 40 will be described with reference to Figures 15 and 16. Note that the hardware configuration of the server 40 is the same as the hardware configuration of the server 40 according to the first embodiment shown in Figure 8, and therefore description thereof will be omitted.

[0129] 15 is a schematic diagram showing an example of the functional configuration of the server 40. The functional configuration shown in FIG. 15 is typically realized by the CPU 41 of the server 40 executing a program 451.

[0130] As shown in FIG. 15, the server 40 includes, as its mechanical configuration, an inference data acquisition unit 400, an inference unit 402, a trained model storage unit 403, and a travel route determination unit 404.

[0131] The inference data acquisition unit 400 acquires biometric data 430 of each user by communicating with the sensor 90. The biometric data 430 is data that indicates changes in the biometric information of the user over time within the facility. The biometric data 430 is generated for each user who is allowed to use the beverage dispensing robot 10.

[0132] The inference unit 402 infers the planned use time 440 of the beverage supply robot 10 for each user using the trained model 753 stored in the trained model storage unit 403. Specifically, the inference unit 402 inputs the biometric data 430 of each user acquired by the data for inference acquisition unit 400 to the trained model 753. The inference unit 402 acquires the planned use time 440 of the beverage supply robot 10 for each user from the trained model 753. This planned use time 440 represents the planned time at which a beverage is to be supplied to the user. The planned use time 440 is generated for each user in association with the biometric data 430. The planned use time 440 of each user is provided to the movement route determination unit 404.

[0133] The movement route determination unit 404 determines the movement route of the beverage supply robot 10 so that the beverage supply robot 10 can move to each user and supply beverages to them, based on the scheduled usage time 440 of each user provided by the inference unit 402 and the user's location information 450 within the facility obtained through communication with the facility management server 50.

[0134] For example, assume that it is inferred from biometric data 430 of a user with user ID "U1" that the user plans to use beverage supply robot 10. In this case, movement route determination unit 404 determines a planned point at which a beverage is to be supplied to the user from user position information 450. Then, movement route determination unit 404 determines a movement route for beverage supply robot 10 using the determined planned supply point and information stored in storage device 45 about routes that beverage supply robot 10 can travel.

[0135] Server 40 transmits to beverage supply robot 10 information regarding the planned supply point and planned supply time (planned use time 440) and information regarding the movement route of beverage supply robot 10. Upon receiving the information from server 40, control unit 18 of beverage supply robot 10 controls drive unit 13 and beverage supply unit 12 (temperature adjustment unit 124) so ​​that a beverage at an appropriate temperature according to the set supply temperature can be supplied to the user at the planned supply time.

[0136] Specifically, the control unit 18 sets the time (movement start time) at which the beverage supply robot 10 departs the station 20 so that the beverage supply robot 10 arrives at the scheduled supply point at the scheduled supply time. Then, the control unit 18 sets the movement start time to a time that is earlier than the scheduled supply time by the travel time.

[0137] When the movement start time arrives, beverage supply robot 10 departs station 20 and moves along the movement path to the planned supply point. During movement, control unit 18 executes intermittent operation of temperature adjustment unit 124 to maintain the liquid temperature at the planned supply temperature. When beverage supply robot 10 arrives at the planned supply point at the planned supply time, control unit 18 stops operation of drive unit 13. Even after arrival at the planned supply point, control unit 18 executes intermittent operation of temperature adjustment unit 124 to maintain the liquid temperature at the planned supply temperature.

[0138] When the supply of the beverage to the user is completed, the beverage supply robot 10 operates the driving unit 13 again and moves from the supply point to the station 20. When the beverage supply robot 10 returns to the station 20, the control unit 18 stops the operation of the driving unit 13. The beverage supply robot 10 receives power again from the station 20 and charges the battery 16.

[0139] In this way, the beverage supply robot 10 moves to the user independently without waiting for a supply instruction from the user, according to the expected use time 440 inferred from the user's biometric data 430. This allows the beverage supply robot 10 to be positioned near the user in a situation where it is estimated that the user will desire a beverage. This allows the user to receive a beverage promptly from the beverage supply robot 10.

[0140] Server 40 can also determine the movement route of beverage supply robot 10 so that beverage supply robot 10 visits multiple users based on the users' scheduled usage times 440. By having beverage supply robot 10 move along the movement route to visit multiple users, beverages can be supplied to each user at an appropriate time.

[0141] Fig. 16 is a flowchart showing an example of a beverage supply process executed by the server 40. Each step of this flowchart is typically executed by the CPU 41 of the server 40. The flowchart shown in Fig. 16 is obtained by replacing S16 to S20 of the flowchart shown in Fig. 11 with S20 to S26.

[0142] As shown in Figure 16, if a supply instruction has not been received from the user terminal 60 (NO in S11), the server 40 (inference data acquisition unit 400) acquires biometric data 430 of each user within the facility by communicating with the sensor 90 in step S20.

[0143] Next, in step S21, the server 40 (inference unit 402) inputs each user's biometric data 430 into the trained model 753 to infer the planned usage time 440 of each user's beverage supply robot 10.

[0144] In step S22, the server 40 (inference unit 402) outputs the inferred scheduled use time 440 of each user.

[0145] In step S23, the server 40 (movement route determination unit 404) acquires the location information 450 of each user by exchanging data with the facility management server 50. In step S24, the server 40 (movement route determination unit 404) determines a planned point where the beverage will be dispensed by the beverage dispenser robot 10 based on the acquired location information 450 of each user.

[0146] Next, in step S25, server 40 (movement route determination unit 404) determines a movement route for beverage supply robot 10 based on each user's planned use time 440 (planned supply time) and the planned beverage supply point. In step S26, server 40 (movement route determination unit 404) transmits information about the determined movement route and information about the planned beverage supply point and planned supply time (planned use time 440) to beverage supply robot 10.

[0147] When the beverage supply robot 10 receives information regarding the planned supply point, planned supply time, and travel route from the server 40, the beverage supply robot 10 departs from the station 20 so as to arrive at the planned supply point at the planned supply time, and travels along the travel route to the planned supply point. In this way, the beverage supply robot 10 that has arrived at the planned supply point can supply a beverage to a user who requests a beverage.

[0148] As described above, according to the second embodiment, beverage supply robot 10 moves autonomously to the user without waiting for a supply instruction from the user, in accordance with scheduled use time 440 inferred from user biometric data 430. This makes it possible to realize beverage supply system 100 that can use beverage supply robot 10 to supply a beverage at a timing appropriate for the user.

[0149] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The technical scope of the present disclosure is defined by the claims, not by the description of the above embodiments, and is intended to include all modifications within the meaning and scope of the claims.

[0150] 10 beverage supply robot, 11, 24, 30 wireless communication device, 12 beverage supply unit, 13 drive unit, 14 camera, 15 operation unit, 18, 26 control unit, 20 station, 22 charger, 40 server, 41, 71 CPU, 42, 72 RAM, 43, 73 ROM, 44, 74 I / F device, 45, 75 storage device, 50 facility management server, 60 user terminal, 70 learning device, 76 drive, 80 storage medium, 90 sensor, 100 beverage supply system, 120 liquid storage tank, 122 liquid supply path, 124 temperature adjustment unit, 126 temperature sensor, 128 stirring unit, 130 discharge outlet, 132 opening and closing mechanism, 400 inference data acquisition unit, 402 inference unit, 403, 703 learned model storage unit, 410 Behavior schedule, 420 Usage schedule, 430, 730 Biometric data, 440 Scheduled supply time, 450 User location information, 451 Program, 700 Learning data acquisition unit, 702 Model generation unit, 710 Behavior record data, 720 Usage record data, 740 Usage record time data, 751 Learning program, 752 Learning data, 753 Learned model, NW Communication network.

Claims

1. A learning device that generates a trained model used to control a beverage supply robot that moves autonomously within a facility and supplies beverages to users, the learning device comprising: a data acquisition unit that acquires training data including status data that represents the status of a user within the facility and usage history data that indicates the user's usage history of the beverage supply robot; and a model generation unit that uses the training data to generate a trained model that takes the user's status data as input and outputs a usage schedule that indicates the user's plans to use the beverage supply robot.

2. The learning device described in claim 1, wherein the status data includes behavior history data indicating the user's behavior and location within the facility, and the usage history data includes information indicating the location and time at which the beverage supply robot supplied the user with a beverage.

3. The learning device described in claim 1, wherein the status data includes time-varying data of the user's biometric information, and the usage history data includes information indicating the time when the beverage supply robot supplied a beverage to the user.

4. A learning method for generating a trained model used to control a beverage supply robot that moves autonomously within a facility and supplies beverages to users, comprising the steps of: acquiring, by a computer, training data including status data that represents the status of a user within the facility and usage history data that indicates the user's usage history of the beverage supply robot; and using the training data, generating, by the computer, a trained model that takes the user's status data as input and outputs a usage schedule that indicates the user's planned use of the beverage supply robot.

5. A beverage supply system comprising: a beverage supply robot that moves autonomously within a facility to supply beverages to users; and a control device that controls the beverage supply robot, wherein the control device includes: a data acquisition unit that acquires status data representing the status of a user within the facility; an inference unit that infers a usage schedule of a user from the status data of the user input from the data acquisition unit using a trained model for inferring a usage schedule indicating when the beverage supply robot will be used from the status data; and a movement path determination unit that determines a movement path of the beverage supply robot in accordance with the usage schedule output from the inference unit.

6. The beverage supply system described in claim 5, wherein the status data includes an activity schedule indicating the user's planned activities and locations within the facility, the inference unit outputs a planned supply point and planned supply time for supplying beverages to the user as the usage schedule, and the movement route determination unit determines the movement route of the beverage supply robot based on the planned supply point and planned supply time output from the inference unit.

7. A beverage supply system as described in claim 5, further comprising a sensor that detects biometric information of a user within the facility, wherein the status data includes time-varying data of the user's biometric information detected by the sensor, wherein the inference unit outputs a scheduled supply time for supplying a beverage to the user as the usage schedule, and wherein the movement route determination unit determines a scheduled supply point for supplying a beverage to the user based on location information of the user, and determines the movement route of the beverage supply robot based on the determined scheduled supply point and the scheduled supply time output from the inference unit.

8. A beverage supply system described in any one of claims 5 to 7, wherein the movement route determination unit determines the movement route of the beverage supply robot so as to visit multiple users based on the usage schedules of the multiple users.

9. A method for controlling a beverage supply robot that moves autonomously within a facility and supplies beverages to users, comprising the steps of: acquiring status data representing the status of a user within the facility by a computer; inferring a usage schedule of a user from the input status data of the user by the computer using a trained model for inferring a usage schedule indicating the planned use of the beverage supply robot from the status data; and determining a movement path of the beverage supply robot by the computer in accordance with the inferred usage schedule.

Images