WO2018174290A1 - Conversation control system, and robot control system - Google Patents

Conversation control system, and robot control system Download PDF

Info

Publication number
WO2018174290A1
WO2018174290A1 PCT/JP2018/011919 JP2018011919W WO2018174290A1 WO 2018174290 A1 WO2018174290 A1 WO 2018174290A1 JP 2018011919 W JP2018011919 W JP 2018011919W WO 2018174290 A1 WO2018174290 A1 WO 2018174290A1
Authority
WO
WIPO (PCT)
Prior art keywords
scenario
robot
dialogue
internal state
control unit
Prior art date
Application number
PCT/JP2018/011919
Other languages
French (fr)
Japanese (ja)
Inventor
雄介 柴田
智彦 大内
麻莉子 矢作
浩平 小川
雄一郎 吉川
石黒 浩
Original Assignee
株式会社 ゼンショーホールディングス
国立大学法人大阪大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社 ゼンショーホールディングス, 国立大学法人大阪大学 filed Critical 株式会社 ゼンショーホールディングス
Publication of WO2018174290A1 publication Critical patent/WO2018174290A1/en

Links

Images

Definitions

  • the present invention relates to a dialogue control device and a robot control system.
  • the present invention has been made in consideration of such points, and in the case where a dialogue between a user and a robot or the like is performed, a dialogue that creates a more realistic sensation of dialogue through the exchange of dialogue reflecting the internal state of the robot or the like. It is an object to provide a control device and a robot control system.
  • An interactive control device includes an information acquisition unit that acquires interactive scenario information for interacting with a customer who has visited a restaurant from a plurality of interactive scenario information, and progress of the interactive scenario corresponding to the acquired interactive scenario information
  • An internal state value acquisition unit that acquires a value obtained by quantifying the internal state that changes with the time, and the accumulation of values acquired by the internal state value acquisition unit as the conversation scenario corresponding to the acquired conversation scenario information progresses
  • a scenario control unit that corrects the dialogue scenario based on the value.
  • the scenario control unit may accumulate values obtained by quantifying the first internal state or the second internal state that are in conflict with each other as the dialogue scenario progresses.
  • the first internal state may be a positive emotion obtained for a customer preference or evaluation regarding a recommended menu
  • the second internal state may be a negative emotion obtained for customer preference or evaluation regarding the recommended menu.
  • the first internal state may be a positive feeling obtained for at least one of the customer's evaluation regarding the ordered menu and the customer's evaluation regarding the customer service
  • the second internal state may be a negative emotion obtained with respect to at least one of a customer evaluation related to the ordered menu and a customer evaluation related to the customer service.
  • the scenario control unit separates the ongoing dialogue scenario based on the customer selection information, the cumulative value of the first internal state and the cumulative value of the second internal state. It may be controlled whether to switch to the interactive scenario.
  • the internal state may include at least one of the emotion and intention of the robot that interacts with the customer.
  • a robot motion control unit that performs an action representing the internal state according to a cumulative value of values obtained by quantifying the internal state by the scenario control unit may be provided.
  • a robot control system includes a robot that performs an operation according to an instruction signal, and an operation terminal that generates an instruction signal and transmits the instruction signal to the robot.
  • An information acquisition unit that acquires dialogue scenario information for interacting with customers who have visited restaurants from multiple dialogue scenario information, and the internal state that changes with the progress of the dialogue scenario corresponding to the acquired dialogue scenario information is quantified
  • the robot motion control unit may change at least one of the voice color of the robot, the motion, the eye color, and the way the eyes glow according to the cumulative value of the values obtained by digitizing the internal state by the scenario control unit.
  • FIG. 1 is a block diagram showing a robot control system 1 according to an embodiment.
  • the flowchart which shows an example of the flow of a process in a scenario control part.
  • the sequence diagram which shows a part of progress example of the dialogue scenario in a basic scenario.
  • the figure which shows the scene corresponding to the dialogue scenario 1-1 typically.
  • the figure which shows the scene corresponding to the dialogue scenario 1-5 typically.
  • the block diagram which shows the robot control system which concerns on a 1st modification.
  • FIG. 1 is a block diagram illustrating a robot control system 1 according to an embodiment.
  • the robot control system 1 includes a robot 2, an operation terminal 3 that is an example of a dialog control device, a handy terminal 4, and a store system 5 such as a POS (point of sales) system.
  • the robot control system 1 is, for example, a system for a restaurant customer (hereinafter referred to as a user) to interact with the robot 2 via an operation terminal 3 that operates the robot 2.
  • the robot 2 in the example of FIG. 1 is a machine having a human-like appearance and interactive function, that is, a humanoid. Note that the robot 2 may have a dissimilar appearance to humans such as animals and characters. Further, the robot 2 may be a virtual robot based on an image displayed on the display unit 38.
  • the robot 2 may be a general-purpose robot that can be programmed in posture, motion, or the like, or may be a robot developed for the robot control system 1.
  • the robot 2 includes a robot drive unit 21 and a robot control unit 22 that is an example of a drive control unit.
  • the robot 2 is driven by electric power supplied from a commercial power source.
  • the robot 2 may be a battery-driven type such as a battery.
  • the robot drive unit 21 includes, for example, an actuator that drives a part of the robot 2 having a degree of freedom, such as a body part, an arm part, a neck part, an eyeball part, a buttocks part, or a mouth part, and a voice output that outputs the speech voice of the robot 2 And a lighting device for lighting the eyeball portion of the robot 2.
  • a degree of freedom such as a body part, an arm part, a neck part, an eyeball part, a buttocks part, or a mouth part
  • a voice output that outputs the speech voice of the robot 2
  • a lighting device for lighting the eyeball portion of the robot 2.
  • a robot control command which is an example of a command signal, is input from the operation terminal 3 to the robot control unit 22.
  • the robot control unit 22 generates the drive control signal described above based on the input robot control command, and outputs the generated drive control signal to the robot drive unit 21. That is, the robot 2 can operate according to the robot control command.
  • the robot control command includes a command for controlling the posture, position, and lighting state of the robot 2 and a command for causing the robot 2 to speak.
  • This robot control command includes robot side scenario data described later.
  • the operation terminal 3 is carried by the user and is, for example, a tablet terminal having a touch function.
  • the operation terminal 3 may be a smartphone, a desktop display type terminal, or the like.
  • the operation terminal 3 is driven by power supplied from a built-in battery, and includes an orientation sensor 31, an operation generation unit 32, a scenario DB (database) 33, an internal state value accumulation unit 34, An information acquisition unit 35, an internal state value acquisition unit 36, a scenario control unit 37, a display unit 38, an input unit 39 such as a touch panel, an audio output unit 40, and a robot operation control unit 41 are included.
  • the operation terminal 3 according to the present embodiment corresponds to a dialogue control device.
  • the azimuth sensor 31 outputs an azimuth detection signal indicating the azimuth of the operation terminal 3 to the motion generation unit 32.
  • the direction sensor 31 outputs a direction detection signal indicating the detected direction of the operation terminal 3 to the motion generation unit 32.
  • the motion generation unit 32 generates a robot control command for controlling the posture of the robot 2 so as to face the direction in which the operation terminal 3 exists based on the direction detection signal from the direction sensor 31.
  • the motion generation unit 32 generates a robot control command for controlling the motion of the robot 2 based on the output of the direction sensor 31.
  • the motion generation unit 32 transmits the generated robot control command to the robot control unit 22 via wireless communication such as Wi-Fi, for example.
  • the robot control unit 22 receives the robot control command from the motion generation unit 32, and outputs a drive control signal corresponding to the received robot control command to the robot drive unit 21, thereby controlling the operation of the robot 2. Do.
  • the scenario DB 33 stores a plurality of pieces of dialogue scenario information for carrying out a dialogue between the user and the robot 2.
  • the dialogue scenario information includes a plurality of scenario groups (each scenario group has a plurality of dialogue scenarios) and a plurality of dialogue scenarios.
  • One dialogue scenario is exchanged between the robot 2 and the user. This is a story of dialogue.
  • One dialogue scenario includes robot-side scenario data used by the robot 2 for dialogue, that is, speech and movement, and user-side scenario data used for dialogue, that is, selection by the operation terminal 3.
  • the user-side scenario data includes some dialogue scenario data that can be responded to the robot-side scenario data, and the user can select arbitrary data from these through the input unit 39.
  • the dialogue scenario data has a tree structure in which the robot side scenario data and the user side scenario data are alternately coupled as nodes.
  • a predetermined series of nodes ranging from the highest level to the lowest level is managed as a basic scenario used for a typical dialogue, for example.
  • Other node groups may be managed as a correction scenario for correcting the basic scenario.
  • dialogue scenario data for a restaurant may be divided into a plurality of scenario groups according to status data indicating a staying state of the user at the restaurant.
  • the status data in the example of FIG. 1 is information indicating that the user has performed the following actions: sitting, ordering, food and drink provision, meal termination, and accounting.
  • the dialogue scenario data is provided from the scenario group classified by these status data.
  • the internal state value accumulation unit 34 accumulates values obtained by quantifying the internal state that changes with the progress of the dialogue scenario corresponding to the dialogue scenario information.
  • the changing internal state includes a first internal state indicating positive emotion and a second internal state indicating negative emotion, and each is quantified.
  • the degree of positive emotion is indicated by a numerical value. That is, the higher the numerical value, the higher the positive feeling.
  • the degree of negative emotion is indicated by a numerical value. In other words, the higher the numerical value, the higher the negative emotion.
  • the neutral feeling is digitized by 0, for example.
  • the internal state includes not only emotions but also all elements that can normally exist in a person such as senses, intentions, and interests. Therefore, hereinafter, the emotion is mainly described as the internal state, but the applicable internal state is not limited to this.
  • the first internal state is a positive emotion obtained for the user's preference or evaluation regarding a recommended menu in a restaurant such as a restaurant
  • the second internal state is a user's preference regarding a recommended menu, for example.
  • negative emotions obtained for evaluation is obtained for evaluation.
  • the first internal state is a positive emotion obtained for at least one of a user's evaluation related to a menu ordered in a restaurant such as a restaurant and a user's evaluation related to a customer service
  • the second internal state is a negative emotion obtained for at least one of the user's evaluation regarding the ordered menu and the user's evaluation regarding the customer service.
  • the internal state value accumulation unit 34 stores numerical values indicating positive emotions and negative emotions in association with options of the user-side dialogue scenario. Thereby, the internal state value storage unit 34 can output a numerical value corresponding to the option selected by the user.
  • the information acquisition unit 35 acquires dialog scenario information for interacting with a user who has visited a restaurant from a plurality of dialog scenario information in the scenario DB 33. That is, the information acquisition unit 35 acquires dialogue scenario information for interacting with the user from the plurality of dialogue scenario information stored in the scenario DB 33 according to the control of the scenario control unit 37, and sends it to the scenario control unit 37. Output.
  • the internal state value acquisition unit 36 acquires a value obtained by quantifying the internal state that changes with the progress of the dialog scenario corresponding to the dialog scenario information. That is, the internal state value acquisition unit 36 acquires a value obtained by quantifying the internal state that changes with the progress of the dialogue scenario stored in the internal state value storage unit 34 according to the control of the scenario control unit 37. Output to the control unit 37.
  • the scenario control unit 37 advances the dialogue scenario based on the dialogue scenario information acquired by the information acquisition unit 35. Moreover, the scenario control part 37 updates the cumulative value of the value which digitized the internal state acquired via the internal state value acquisition part 36 with progress of a dialogue scenario. That is, as described above, the scenario control unit 37 determines at least one of the value obtained by quantifying the first internal state indicating positive emotion and the value obtained by quantifying the second internal state indicating negative emotion as the progress of the dialogue scenario. It accumulates with. Then, the scenario control unit 37 corrects the dialogue scenario based on the cumulative value of the values acquired by the internal state value acquisition unit 36. In addition, the scenario control unit 37 sends the cumulative value obtained by quantifying the first internal state indicating positive emotion and the cumulative value obtained by quantifying the second internal state indicating negative emotion via the motion generating unit 32. This is supplied to the robot operation control unit 41.
  • the scenario control unit 37 causes the display unit 38 to display user-side dialog scenario options based on the user-side scenario data included in the dialog scenario information.
  • the scenario control unit 37 causes the display unit 38 to display user-side dialog scenario options based on the user-side scenario data included in the dialog scenario information.
  • the motion generation unit 32 generates a robot control command for causing the robot 2 to speak and operate based on the robot side scenario data input from the scenario control unit 37. At this time, the motion generation unit 32 includes commands related to motions such as the tone of the utterance voice, the lighting state of the eyeball, and the posture of the robot 2 in the command content of the robot control command corresponding to the robot side scenario data.
  • the user can select a desired dialogue scenario using the input unit 39 from the choices of the dialogue scenario displayed on the display unit 38. That is, the input unit 39 outputs a dialogue scenario selection signal indicating a dialogue scenario selected by the user to the scenario control unit 37 and the voice output unit 40.
  • the voice output unit 40 outputs the selected dialogue scenario as a voice based on the dialogue scenario selection signal input from the input unit 39.
  • the robot motion control unit 41 causes the robot 2 to perform an action representing the internal state in accordance with a cumulative value of values obtained by quantifying the internal state sequentially received from the operation terminal 3 as the dialogue progresses.
  • the internal state is a state including at least one of emotion and intention.
  • the robot motion control unit 41 can cause the robot 2 to perform an action reflecting at least one of the emotion and intention of the robot 2.
  • the robot motion control unit 41 generates a robot control command that changes at least one of the voice color, motion, eye color, and eye glow of the robot 2 according to the accumulated value of the internal state.
  • the internal state of the robot 2 is digitized, and the robot 2 is intended to perform some action according to the accumulated value of the digitized values.
  • a control unit 41 is provided.
  • the robot motion control unit 41 may increase the voice of the robot 2, increase the voice, speed up the blinking of light, change the tone to a pleasant tone, You can change the color of your eyes to a warm color, change your face to a more enjoyable expression by raising your corners, raise your arms, or adjust your face upwards.
  • the robot motion control unit 41 may reduce the pitch of the voice of the robot 2, decrease the voice, slow the blinking of light, or change it to a sad tone as an action expressing negative emotion. You can change the color of the eyes to a cool color, change the face to a sad expression by lowering the corner of the mouth, whisper the face or body, lower the arm, or lower the shoulder.
  • the robot 2 by operating the robot 2 so as to include at least one of emotion and intention through the exchange of dialogue, the realism of the dialogue is suppressed, and the naturalness of the dialogue is suppressed by suppressing the sense of incongruity of the dialogue. Can be generated. Thereby, the user can improve the satisfaction by the dialogue with the robot 2.
  • the robot motion control unit 41 may generate a robot control command based on the robot control command input from the motion generation unit 32 and the accumulated value obtained by quantifying the internal state. For example, the robot motion control unit 41 may correct another robot control command input from the motion generation unit 32 and generate another robot control command.
  • the robot motion control unit 41 is input from the scenario control unit 37 with a cumulative value obtained by quantifying the internal state, but is not limited thereto, and the robot motion control unit 41 includes the internal state. A cumulative value obtained by quantifying the values may be generated.
  • FIG. 2 is a diagram showing a table of control operation examples of the scenario control unit 37, and the options A, B, and C are selected by the user from the options displayed on the display unit 38 as the dialogue scenario progresses. Information on options is shown.
  • Positive emotion ( ⁇ ) indicates a cumulative value obtained by quantifying the first internal state indicating the positive emotion when the user selects the target option
  • negative emotion ( ⁇ ) indicates the target option.
  • the cumulative value of the value which digitized the 2nd internal state which shows the negative emotion at the time of selecting is shown.
  • the scenario control unit 37 corrects the dialogue scenario based on the cumulative value acquired by the internal state value acquisition unit 36. For example, as shown in the first line in the table of FIG. 2, the scenario control unit 37 sets the dialogue scenario information when the cumulative value of the value obtained by quantifying the first internal state indicating positive emotion is equal to or greater than a predetermined threshold value 10. Is corrected in consideration of the positive emotion of the robot 2. On the other hand, as shown in the second row in the table of FIG. 2, the scenario control unit 37 interacts when the cumulative value of the value obtained by quantifying the second internal state indicating negative emotion is equal to or greater than the predetermined threshold value 10. The scenario information is corrected in consideration of the negative emotion of the robot 2.
  • the scenario control unit 37 accumulates values obtained by quantifying the first internal state or the second internal state, which are mutually contradictory, as the dialogue scenario progresses, and corrects the dialogue scenario based on these accumulated values. To do. That is, the scenario control unit 37 updates the accumulated value of the value acquired by the internal state value acquiring unit 36 as the dialog scenario corresponding to the dialog scenario information acquired via the information acquiring unit 35 progresses, Correct the dialogue scenario when the value reaches the threshold. Thereby, the scenario control unit 37 can correct the conversation scenario into a conversation scenario according to the internal state of the robot 2.
  • scenario correction refers to changing the order of dialogue scenarios in ongoing dialogue scenario information, changing dialogue scenario information to other dialogue scenario information, and controlling the operation of the robot 2. It means changing the dialogue scenario.
  • the dialogue scenario information is corrected in consideration of the positive emotion of the robot 2
  • the dialogue scenario information is changed to a more familiar feeling, for example, a light feeling like the dialogue between friends.
  • the dialogue scenario information is changed to dialogue scenario information with a feeling of heavy honorific tone.
  • the scenario correction performed by the scenario control unit 37 can change the dialog scenario information, so that the dialog contents can also be changed.
  • the robot-side scenario data related to the robot control command is corrected, and the operation of the robot 2 is further increased.
  • the robot side scenario data related to the robot control command is corrected to make the operation of the robot 2 slower or reduce the movement amount.
  • the scenario control unit 37 corrects the dialogue scenario in accordance with the accumulated value of the value indicating the internal state of the robot 2 that is changed as the dialogue scenario progresses, so that the dialogue between the robot 2 and the user is exchanged. It is possible to create a sense of reality.
  • the scenario control unit 37 corrects the scenario according to the accumulated value obtained by accumulating the digitized values, the value indicating the internal state generated so far can be reflected in the scenario correction.
  • the internal state value indicating the emotion and intention stored in the robot 2 can be reflected in the dialogue and operation of the robot 2.
  • the robot 2 it is possible to cause the robot 2 to perform a tone-like dialogue or action that a human performs when a dialogue that enhances negative emotions continues.
  • an emotional element generated in a dialogue between humans even in a dialogue between the robot 2 and the user.
  • the dialogue scenario when both the cumulative value of positive emotion and the cumulative value of negative emotion are less than 10, the dialogue scenario is advanced. Although omitted in FIG. 2, the dialogue scenario may be switched when the cumulative value of at least one of positive emotion and negative emotion reaches a predetermined value (for example, 20) greater than 10.
  • a predetermined value for example, 20
  • FIG. 3 is a flowchart showing an example of the process flow in the scenario control unit 37.
  • An example of the process flow in the scenario control unit 37 will be described with reference to FIG.
  • an example will be described in which the scenario control unit 37 reads out and advances a correction scenario corresponding to the status data input from the handy terminal 4 via the store system 5 from the scenario DB 33.
  • the scenario control unit 37 causes the display unit 38 to display the user-side dialogue scenario options based on the user-side scenario data included in the read dialogue scenario information. Subsequently, the user selects a specific option displayed on the display unit 38 (step S300). In this case, the robot-side scenario data included in the dialogue scenario information is output to the motion generation unit 32.
  • the scenario control unit 37 acquires the internal state value of the robot 2 from the internal state value accumulation unit 34 via the internal state value acquisition unit 36 based on the option selected by the user (step S302). That is, the scenario control unit 37 acquires a value obtained by quantifying the first internal state and a value obtained by quantifying the second internal state.
  • the scenario control unit 37 accumulates the value obtained by quantifying the first internal state and the value obtained by quantifying the second internal state, and updates the accumulated value (step S304). ). That is, the scenario control unit 37 updates the cumulative value as the positive emotion ( ⁇ ) and the cumulative value as the negative emotion ( ⁇ ), respectively.
  • the scenario control unit 37 determines whether any of the cumulative value as the positive emotion ( ⁇ ) and the cumulative value as the negative emotion ( ⁇ ) exceeds a threshold, for example, 10 (step S306). When either the cumulative value as the positive emotion ( ⁇ ) or the cumulative value as the negative emotion ( ⁇ ) is equal to or greater than the threshold (step S306: YES), the scenario control unit 37 corrects the dialogue scenario (step S308). ). On the other hand, when both the cumulative value as positive emotion ( ⁇ ) and the cumulative value as negative emotion ( ⁇ ) are less than the threshold value (step S306: NO), the scenario control unit 37 selects the currently executed dialogue scenario. Continue to proceed (step S310).
  • a threshold for example, 10
  • step S312 determines whether or not the dialogue scenario has ended.
  • step S312 determines that the dialogue scenario has ended.
  • step S312 determines that the conversation scenario has not ended.
  • step S312 repeats the processing from (step S300).
  • the scenario control unit 37 accumulates a cumulative value as a positive emotion ( ⁇ ) and a cumulative value as a negative emotion ( ⁇ ), which are opposite to each other, as the dialogue scenario progresses. If any of them is equal to or greater than the threshold value, the dialogue scenario is corrected. On the other hand, the scenario control unit 37 advances the dialogue scenario when any of these accumulated values is less than the threshold value. Thereby, when the accumulated value is equal to or greater than the threshold value, the scenario control unit 37 can correct the conversation scenario to a conversation scenario according to the internal state of the robot 2 and can select the conversation scenario according to the internal state of the robot 2. is there.
  • FIG. 4 is a sequence diagram showing a part of the progress example of the dialogue scenario.
  • the progress example of the scenario will be described with reference to FIG.
  • Dialog scenarios 1-1 to 1-4 in FIG. 4 are partial scenarios included in the dialog scenario 1.
  • the initial value of the cumulative value of positive emotion is assumed to be ⁇
  • the initial value of the cumulative value of negative emotion is assumed to be ⁇ .
  • the dialogue scenario 1 starts from the dialogue scenario 1-1.
  • the user's options in the dialogue scenario 1-1 are option 1 and option 2.
  • the user selects option 1 the first internal state value indicating positive emotion is 0, and the second internal state value indicating negative emotion is 1. That is, for the robot 2, option 1 indicates that it is not a preferable selection. For this reason, the cumulative value of the negative emotion of the robot 2 increases to ⁇ + 1.
  • the first internal state value indicating positive emotion is 1, and the second internal state value indicating negative emotion is 0. That is, for the robot 2, option 2 is a preferable selection. For this reason, the cumulative value of the positive emotion of the robot 2 increases to ⁇ + 1.
  • the dialogue scenario 1-3 proceeds.
  • the user's options in the dialogue scenario 1-3 are options 3-7.
  • option 3 is selected.
  • the first internal state value indicating positive emotion is 1
  • the second internal state value indicating negative emotion is 1. That is, for the robot 2, the option 3 indicates that the positive emotion is increased and the negative emotion is decreased. For this reason, the cumulative value of the positive emotion of the robot 2 increases to ⁇ + 2, and the cumulative value of the negative emotion increases to ⁇ + 1.
  • the dialogue scenario 1-4 proceeds according to the selection of the option 3.
  • the user's options in the dialogue scenario 1-4 are option 8 and option 9.
  • the first internal state value indicating positive emotion is 3, and the second internal state value indicating negative emotion is 0. That is, for the robot 2, option 8 indicates that it is a very preferable choice.
  • the cumulative value of the positive emotion of the robot 2 increases to ⁇ + 5.
  • the cumulative value of positive emotion becomes equal to or greater than the threshold value.
  • the scenario control unit 37 obtains a scenario correction in which the dialogue scenario 1 is corrected in consideration of the positive emotion of the robot 2.
  • the scenario correction scenario proceeds in accordance with the selection of the option 8.
  • the first internal state value indicating positive emotion is 0, and the second internal state value indicating negative emotion is 2. That is, for the robot 2, the option 9 is not a preferable choice. For this reason, the cumulative value of the negative emotion of the robot 2 increases to ⁇ + 3.
  • the first internal state value and the second internal state value are associated with each option, and the cumulative value and negative emotion of the value obtained by quantifying the first internal state indicating positive emotion according to the selection of the option are displayed. The accumulated value of the value obtained by quantifying the second internal state shown is updated.
  • FIG. 5 is a diagram schematically showing a scene corresponding to the dialogue scenario 1-1.
  • the operation terminal 3 is, for example, a tablet-type portable terminal that includes a display unit 38 and a transparent touch panel type input unit 39 installed together with the display unit 38.
  • the scenario control unit 37 Based on the robot-side scenario data of the dialogue scenario 1-1 acquired by the information acquisition unit 35, the utterance “What to order?” Is executed from the robot 2.
  • the scenario control unit 37 causes the display unit 38 to display user-side dialog scenario options based on the user-side scenario data included in the dialog scenario 1-1.
  • the options here are option 500 and option 502.
  • the cumulative negative emotion value of the robot 2 increases by 1.
  • the positive emotion cumulative value of the robot 2 increases by one.
  • the body 2a of the robot 2 is supported on the pedestal 2b so as to be rotatable about a vertical axis, and the body 2a of the robot 2 may rotate according to a dialogue scenario.
  • FIG. 6 is a diagram schematically showing a scene corresponding to the dialogue scenario 1-3.
  • the scenario control unit 37 displays the user-side dialog scenario options on the display unit 38 based on the user-side scenario data included in the dialog scenario 1-3 acquired by the information acquisition unit 35. It is displayed.
  • the options here are options 600, 602, 604, 606, and 608.
  • the options 600, 602, and 608 are items related to user preferences regarding the recommended menu
  • the options 604 and 606 are items related to user evaluation regarding the recommended menu.
  • the first internal state according to the present embodiment is a positive emotion obtained for at least one of the user's evaluation regarding the ordered menu and the user's evaluation regarding the customer service
  • the second internal state Is a negative emotion obtained for at least one of the user's evaluation regarding the ordered menu and the user's evaluation regarding the customer service.
  • the cumulative value of the positive emotion of the robot 2 is increased by 1, and the cumulative value of the negative emotion is increased by 1.
  • the cumulative value of the positive emotion of the robot 2 increases by 2, and the cumulative value of the negative emotion does not change.
  • the option 604 is selected, the cumulative value of the positive emotion of the robot 2 does not change, and the cumulative value of the negative emotion increases by one.
  • the option 606 is selected, the cumulative values of the positive emotion and negative emotion of the robot 2 are not changed.
  • the option 608 the cumulative value of the positive emotion of the robot 2 increases by 3, but the cumulative value of the negative emotion does not change.
  • FIG. 7 is a diagram schematically showing a scene corresponding to a dialogue scenario in which scenario correction considering positive emotion is performed.
  • the robot 2 speaks according to the robot side scenario data included in the dialogue scenario.
  • the scenario correction is performed in consideration of the positive emotion
  • the operation of the robot 2 is corrected to the operation indicating the positive emotion.
  • the scenario is corrected so that the facial expression of the robot 2 is a smile and the voice is high.
  • the scenario control unit 37 causes the display unit 38 to display user-side dialog scenario options based on the user-side scenario data included in the correction scenario.
  • the options here are option 700 and option 702.
  • the option 700 is selected, the cumulative negative emotion value of the robot 2 increases by 1.
  • the option 702 is selected, the positive emotion cumulative value of the robot 2 increases by one.
  • FIG. 8 is a diagram schematically showing a scene corresponding to the dialogue scenario 1-5.
  • the robot 2 says “Did you like it?” According to the robot-side scenario data included in the dialogue scenario 1-5.
  • the robot 2 since the second internal state value indicating negative emotion is increased by 2 by selection 9 (FIG. 4), the robot 2 is performing an operation indicating negative emotion.
  • the operation of the robot 2 is controlled so that the expression becomes darker and the voice becomes lower than the expression of the robot 2 in the previous dialogue scenario 1-4.
  • FIG. 9 is a diagram schematically illustrating a scene relating to user evaluation of a menu.
  • FIG. 10 is a diagram schematically showing a scene related to user evaluation for customer service.
  • the robot 2 makes a statement “Is the meal delicious?”
  • the options 900 and 902 are items relating to user evaluation regarding the ordered menu. For example, when the option 900 is selected, the cumulative value of the positive emotion of the robot 2 is increased by 3, and the cumulative value of the negative emotion of the robot 2 is not changed. On the other hand, when the option 902 is selected, the cumulative value of the negative emotion of the robot 2 increases by 1, and the cumulative value of the positive emotion of the robot 2 does not change.
  • the robot 2 remarks “Has you enjoyed it” according to the robot side scenario data included in the post-meal dialogue scenario?
  • the options 1000 and 1002 are items relating to user evaluation regarding the ordered menu. That is, the first internal state according to the present embodiment includes positive emotions obtained for at least one of user evaluation regarding the ordered menu and user evaluation regarding the customer service, and the second internal state is the order Negative feelings obtained for at least one of the user's evaluation regarding the selected menu and the user's evaluation regarding the customer service.
  • the option 1000 the cumulative value of the positive emotion of the robot 2 is increased by 3, and the cumulative value of the negative emotion of the robot 2 is not changed.
  • the option 1002 when the option 1002 is selected, the cumulative value of the negative emotion of the robot 2 increases by 1, and the cumulative value of the positive emotion of the robot 2 does not change.
  • a user who visits a restaurant can perform an appropriate dialogue with the robot 2 installed in the restaurant and the operation terminal 3. That is, in the present embodiment, the scenario control unit 37 corrects the dialogue scenario based on the accumulated value of the value indicating the internal state whose value is changed as the dialogue scenario progresses. Thereby, the scenario control unit 37 can correct the dialogue scenario into a dialogue scenario according to the internal state of the robot 2, and can give a sense of dialogue and a sense of reality to the exchange of dialogue with the robot 2. Thereby, the unnaturalness of the dialogue is eliminated, and the user's satisfaction can be increased.
  • FIG. 11 is a block diagram showing the robot control system 1 according to the first modification.
  • the scenario DB 33 and the internal state value storage unit 34 are installed in the operation terminal 3, but are not limited thereto.
  • the scenario DB 33 and the internal state value storage unit 34 may be outside the operation terminal 3.
  • the scenario control unit 37 is connected to the scenario DB 33 and the internal state value storage unit 34 via a network.
  • the network and the operation terminal 3 or the scenario DB 33 are connected via wired or wireless communication.
  • the network and the operation terminal 3 or the internal state value storage unit 34 are connected via wired or wireless communication.
  • the robot motion control unit 41 is installed in the operation terminal 3, but is not limited thereto.
  • the robot motion control unit 41 may be in the robot 2.
  • the scenario DB 33 by having the scenario DB 33 outside the operation terminal 3, it is not necessary to manage the scenario DB at the operation terminal 3, and the scenario data can be centrally managed. Furthermore, by setting the scenario DB 33 on the cloud so that a specific number of editors can edit or add scenario data, it is possible to select more varied dialogue contents. It becomes possible.
  • FIG. 12 is a diagram illustrating a configuration of the robot control system 1 according to the second modification.
  • the motion generation unit 32, the information acquisition unit 35, the internal state value acquisition unit 36, and the scenario control unit 37 may be provided in the robot 2 instead of the operation terminal 3. That is, the main control function may be provided in the robot 2 instead of the operation terminal 3.
  • the operations of the operation generation unit 32, the information acquisition unit 35, the internal state value acquisition unit 36, and the scenario control unit 37 are not significantly different from the operations in the above-described embodiment, and the flowchart shown in FIG. The dialogue process is executed according to the above.
  • the configuration of the operation terminal 3 can be simplified, and the amount of data transmitted from the operation terminal 3 to the robot 2 can be reduced, so that the performance of the communication line between the operation terminal 3 and the robot 2 can be reduced. Even when the value is low, dialogue with the robot 2 can be performed without any trouble.
  • the motion generation unit 32 may be provided outside the operation terminal 3 and the robot 2.
  • the motion generation unit 32 may be built in the store system 5 or provided in a communication device separate from the store system 5. May be.
  • the dialogue is described as a state in which the user or the robot 2 outputs some information, but is not limited thereto.
  • a silent scenario may be prepared in advance and the silent scenario may be output. .
  • this silent scenario is also included as scenario data.
  • the robot control system 1 according to all the embodiments and the modifications described above can be suitably applied to the eating and drinking service as described above, but may be applied to scenes other than the eating and drinking service.
  • the motion generation unit 32 may be provided outside the operation terminal 3 and the robot 2.
  • the motion generation unit 32 may be built in the store system 5 or provided in a communication device separate from the store system 5. May be.
  • the operation terminal 3 may have a function of ordering food and drink through communication with the store system 5.
  • the robot control system 1 can be suitably applied to a food and drink service, but may be applied to scenes other than the food and drink service.

Landscapes

  • Manipulator (AREA)

Abstract

The purpose of the present invention is to further increase the realism of a conversation, through the back and forth of a conversation reflecting an inner state of a robot or the like, in a situation in which a conversation is being carried on between a user and the robot or the like. A conversation control device according to one embodiment is provided with an information acquisition unit, an inner state value acquisition unit, and a scenario control unit. The information acquisition unit acquires, from multiple pieces of conversation scenario information, conversation scenario information for having a conversation with a customer visiting a restaurant. The inner state value acquisition unit acquires a value obtained by digitizing an inner state which changes according to the progress of a conversation scenario corresponding to the acquired conversation scenario information. The scenario control unit, on the basis of a cumulative value of the values acquired by the inner state value acquisition unit, corrects a conversation scenario according to the progress of the conversation scenario corresponding to the acquired conversation scenario information.

Description

対話制御装置およびロボット制御システムDialog control device and robot control system
 本発明は、対話制御装置およびロボット制御システムに関する。 The present invention relates to a dialogue control device and a robot control system.
 近年、レストランその他公共施設において、来場客の案内役としてヒューマノイド型ロボットを活用する事例がみられる(特開2015-66623号公報参照)。このようなロボットとのコミュニケーション(対話)を専用の操作端末を用いて行う技術が検討されており、この技術によれば、ユーザは、操作端末を介してロボットとの対話を楽しんだり、案内情報を入手したりすることができる。 In recent years, there are cases where humanoid robots are used as a guide for visitors in restaurants and other public facilities (see Japanese Patent Laid-Open No. 2015-66623). A technology for performing communication (dialogue) with such a robot using a dedicated operation terminal has been studied. According to this technology, a user can enjoy a dialogue with a robot via an operation terminal, or guide information can be obtained. Can be obtained.
 しかしながら、タブレット等の操作端末を介してロボットと対話を行う場合、人間同士の対話で発生する感情的な要素がないため、対話に対する臨場感がなく、操作端末の操作者は対話を十分に楽しめないという問題があった。 However, when interacting with a robot via an operation terminal such as a tablet, there is no emotional element that occurs in human interaction, so there is no sense of realism in the interaction, and the operator of the operation terminal can fully enjoy the interaction. There was no problem.
 本発明は、このような点を考慮してなされたものであり、ユーザとロボット等とで対話を行う場合において、ロボット等の内部状態を反映した対話のやり取りを通じてより対話の臨場感が生まれる対話制御装置およびロボット制御システムを提供することを目的とする。 The present invention has been made in consideration of such points, and in the case where a dialogue between a user and a robot or the like is performed, a dialogue that creates a more realistic sensation of dialogue through the exchange of dialogue reflecting the internal state of the robot or the like. It is an object to provide a control device and a robot control system.
 一実施形態に係る対話制御装置は、飲食店に来店した顧客と対話するための対話シナリオ情報を複数の対話シナリオ情報から取得する情報取得部と、取得した対話シナリオ情報に対応する対話シナリオの進行に伴って変化する内部状態を数値化した値を取得する内部状態値取得部と、取得した対話シナリオ情報に対応する対話シナリオの進行に伴って、内部状態値取得部で取得された値の累積値に基づいて対話シナリオを補正するシナリオ制御部と、を備える。 An interactive control device according to an embodiment includes an information acquisition unit that acquires interactive scenario information for interacting with a customer who has visited a restaurant from a plurality of interactive scenario information, and progress of the interactive scenario corresponding to the acquired interactive scenario information An internal state value acquisition unit that acquires a value obtained by quantifying the internal state that changes with the time, and the accumulation of values acquired by the internal state value acquisition unit as the conversation scenario corresponding to the acquired conversation scenario information progresses A scenario control unit that corrects the dialogue scenario based on the value.
 シナリオ制御部は、対話シナリオの進行に伴って、互いに相反する第1内部状態または第2内部状態を数値化した値をそれぞれ累積してもよい。 The scenario control unit may accumulate values obtained by quantifying the first internal state or the second internal state that are in conflict with each other as the dialogue scenario progresses.
 第1内部状態は、推奨メニューに関する顧客の嗜好または評価に対して得られる正感情であってもよく、
 第2内部状態は、推奨メニューに関する顧客の嗜好または評価に対して得られる負感情であってもよい。
The first internal state may be a positive emotion obtained for a customer preference or evaluation regarding a recommended menu,
The second internal state may be a negative emotion obtained for customer preference or evaluation regarding the recommended menu.
 第1内部状態は、注文したメニューに関する顧客の評価、および接客サービスに関する顧客の評価、の少なくともいずれかに対して得られる正感情であってもよく、
 第2内部状態は、注文したメニューに関する顧客の評価、および接客サービスに関する顧客の評価、の少なくともいずれかに対して得られる負感情であってもよい。
The first internal state may be a positive feeling obtained for at least one of the customer's evaluation regarding the ordered menu and the customer's evaluation regarding the customer service,
The second internal state may be a negative emotion obtained with respect to at least one of a customer evaluation related to the ordered menu and a customer evaluation related to the customer service.
 シナリオ制御部は、顧客の選択情報と、第1内部状態を数値化した値の累積値と、第2内部状態を数値化した値の累積値と、に基づいて、進行中の対話シナリオを別の対話シナリオに切り替えるか否かを制御してもよい。 The scenario control unit separates the ongoing dialogue scenario based on the customer selection information, the cumulative value of the first internal state and the cumulative value of the second internal state. It may be controlled whether to switch to the interactive scenario.
 内部状態は、顧客と対話を行うロボットの感情および意図の少なくとも一方を含でもよい。 The internal state may include at least one of the emotion and intention of the robot that interacts with the customer.
 シナリオ制御部によって内部状態を数値化した値の累積値に応じて、内部状態を表す行動を行うロボット動作制御部を備えてもよい。 A robot motion control unit that performs an action representing the internal state according to a cumulative value of values obtained by quantifying the internal state by the scenario control unit may be provided.
 一実施形態に係るロボット制御システムは、指示信号に応じた動作を行うロボットと、指示信号を生成してロボットに送信する操作端末と、を備え、操作端末は、
 飲食店に来店した顧客と対話するための対話シナリオ情報を複数の対話シナリオ情報から取得する情報取得部と、取得した対話シナリオ情報に対応する対話シナリオの進行に伴って変化する内部状態を数値化した値を取得する内部状態値取得部と、取得した対話シナリオ情報に対応する対話シナリオの進行に伴って、内部状態値取得部で取得された値の累積値に基づいて対話シナリオを補正するシナリオ制御部と、を備える。
A robot control system according to an embodiment includes a robot that performs an operation according to an instruction signal, and an operation terminal that generates an instruction signal and transmits the instruction signal to the robot.
An information acquisition unit that acquires dialogue scenario information for interacting with customers who have visited restaurants from multiple dialogue scenario information, and the internal state that changes with the progress of the dialogue scenario corresponding to the acquired dialogue scenario information is quantified A scenario that corrects a dialogue scenario based on the cumulative value of values acquired by the internal state value acquisition unit as the dialogue scenario corresponding to the acquired dialogue scenario information progresses. A control unit.
 ロボット動作制御部は、シナリオ制御部によって内部状態を数値化した値の累積値に応じて、ロボットの声色、動作、目の色および目の光り方の少なくとも一つを変化させてもよい。 The robot motion control unit may change at least one of the voice color of the robot, the motion, the eye color, and the way the eyes glow according to the cumulative value of the values obtained by digitizing the internal state by the scenario control unit.
 本発明によれば、ユーザとロボット等とで対話を行う場合において、ロボット等の内部状態を反映した対話のやり取りを通じてより対話の臨場感を引き出すことができる。 According to the present invention, when a dialogue is performed between a user and a robot or the like, it is possible to bring out a sense of realism through dialogue exchange that reflects the internal state of the robot or the like.
一実施形態によるロボット制御システム1を示すブロック図。1 is a block diagram showing a robot control system 1 according to an embodiment. シナリオ制御部における制御動作例の表を示す図。The figure which shows the table | surface of the example of control operation in a scenario control part. シナリオ制御部における処理の流れの一例を示すフローチャート。The flowchart which shows an example of the flow of a process in a scenario control part. 基本シナリオ内の対話シナリオの進行例の一部を示すシーケンス図。The sequence diagram which shows a part of progress example of the dialogue scenario in a basic scenario. 対話シナリオ1-1に対応するシーンを模式的に示す図。The figure which shows the scene corresponding to the dialogue scenario 1-1 typically. 対話シナリオ1-3に対応するシーンを模式的に示す図。The figure which shows typically the scene corresponding to the dialogue scenario 1-3. 正感情を考慮したシナリオ補正がかかった対話シナリオに対応するシーンを模式的に示す図。The figure which shows typically the scene corresponding to the dialogue scenario in which the scenario correction | amendment which considered the positive emotion was applied. 対話シナリオ1-5に対応するシーンを模式的に示す図。The figure which shows the scene corresponding to the dialogue scenario 1-5 typically. メニューに対する顧客の評価に関するシーンを模式的に示す図。The figure which shows typically the scene regarding the customer's evaluation with respect to a menu. 客サービスに対する顧客の評価に関するシーンを模式的に示す図。The figure which shows typically the scene regarding the customer's evaluation with respect to customer service. 第1変形例に係るロボット制御システムを示すブロック図。The block diagram which shows the robot control system which concerns on a 1st modification. 第2変形例に係るロボット制御システムの構成を示す図。The figure which shows the structure of the robot control system which concerns on a 2nd modification.
 以下、本発明の実施形態に係るロボット制御システムおよび対話制御装置の構成及び作用について、図面を参照しながら詳細に説明する。なお、以下に示す実施形態は本発明の実施形態の一例であって、本発明はこれらの実施形態に限定して解釈されるものではない。また、本実施形態で参照する図面において、同一部分または同様な機能を有する部分には同一の符号または類似の符号を付し、その繰り返しの説明は省略する場合がある。また、図面の寸法比率は説明の都合上実際の比率とは異なる場合や、構成の一部が図面から省略される場合がある。 Hereinafter, the configuration and operation of the robot control system and the dialogue control apparatus according to the embodiment of the present invention will be described in detail with reference to the drawings. In addition, embodiment shown below is an example of embodiment of this invention, This invention is limited to these embodiment, and is not interpreted. In the drawings referred to in this embodiment, the same portions or portions having similar functions are denoted by the same reference symbols or similar reference symbols, and repeated description thereof may be omitted. In addition, the dimensional ratio in the drawing may be different from the actual ratio for convenience of explanation, or a part of the configuration may be omitted from the drawing.
(一実施形態)
 図1は、一実施形態によるロボット制御システム1を示すブロック図である。図1に示すように、ロボット制御システム1は、ロボット2と、対話制御装置の一例である操作端末3と、ハンディ端末4と、POS(point of sales)システムなどの店舗システム5とを備える。ロボット制御システム1は、例えば、飲食店の来店客(以下、ユーザと呼ぶ)が、ロボット2を操作する操作端末3を介してロボット2と対話するためのシステムである。図1の例におけるロボット2は、人間に似た外観および対話機能を有する機械、すなわちヒューマノイドである。なお、ロボット2は、動物やキャラクタなどの人間と非類似の外観を有していてもよい。また、ロボット2は、表示部38に表示された画像による仮想的なロボットであってもよい。ロボット2は、姿勢や動作等をプログラム可能な汎用のロボットでもよいし、本ロボット制御システム1用に開発されたロボットでもよい。
(One embodiment)
FIG. 1 is a block diagram illustrating a robot control system 1 according to an embodiment. As shown in FIG. 1, the robot control system 1 includes a robot 2, an operation terminal 3 that is an example of a dialog control device, a handy terminal 4, and a store system 5 such as a POS (point of sales) system. The robot control system 1 is, for example, a system for a restaurant customer (hereinafter referred to as a user) to interact with the robot 2 via an operation terminal 3 that operates the robot 2. The robot 2 in the example of FIG. 1 is a machine having a human-like appearance and interactive function, that is, a humanoid. Note that the robot 2 may have a dissimilar appearance to humans such as animals and characters. Further, the robot 2 may be a virtual robot based on an image displayed on the display unit 38. The robot 2 may be a general-purpose robot that can be programmed in posture, motion, or the like, or may be a robot developed for the robot control system 1.
(ロボット2)
 図1に示すように、ロボット2は、ロボット駆動部21と、駆動制御部の一例であるロボット制御部22とを有する。ロボット2は、商用電源から供給される電力によって駆動される。なお、ロボット2は、電池等のバッテリ駆動型であってもよい。
(Robot 2)
As shown in FIG. 1, the robot 2 includes a robot drive unit 21 and a robot control unit 22 that is an example of a drive control unit. The robot 2 is driven by electric power supplied from a commercial power source. The robot 2 may be a battery-driven type such as a battery.
 ロボット駆動部21は、例えば、胴体部、腕部、首部、眼球部、瞼部または口部等の自由度を有するロボット2の部位を駆動するアクチュエータと、ロボット2の発話音声を出力する音声出力装置と、ロボット2の眼球部を点灯させる点灯装置とを有する。ロボット駆動部21を必要に応じて駆動することで、ユーザとの対話を行うために、ロボット2に発話させたり、ロボット2の姿勢を変化させたり、ロボット2の眼球部を所定の点灯色で点灯させたりすることができる。ロボット駆動部21には、ロボット制御部22から、ロボット駆動部21の駆動を制御する駆動制御信号が入力される。ロボット駆動部21は、駆動制御信号に応じて駆動される。 The robot drive unit 21 includes, for example, an actuator that drives a part of the robot 2 having a degree of freedom, such as a body part, an arm part, a neck part, an eyeball part, a buttocks part, or a mouth part, and a voice output that outputs the speech voice of the robot 2 And a lighting device for lighting the eyeball portion of the robot 2. By driving the robot drive unit 21 as necessary, the robot 2 is uttered, the posture of the robot 2 is changed, and the eyeball portion of the robot 2 is illuminated with a predetermined lighting color in order to interact with the user. It can be lit. A drive control signal for controlling the drive of the robot drive unit 21 is input from the robot control unit 22 to the robot drive unit 21. The robot drive unit 21 is driven according to the drive control signal.
 ロボット制御部22には、指令信号の一例であるロボット制御コマンドが操作端末3から入力される。このロボット制御部22は、入力されたロボット制御コマンドに基づいて既述した駆動制御信号を生成し、生成された駆動制御信号をロボット駆動部21に出力する。すなわち、ロボット2は、ロボット制御コマンドに応じて動作することができる。なお、ロボット制御コマンドは、これによって駆動を制御しようとするロボット駆動部21の種類や制御内容に応じて多種多様である。例えば、ロボット制御コマンドには、ロボット2の姿勢や位置や点灯状態を制御するためのコマンドや、ロボット2を発話させるためのコマンドがある。このロボット制御コマンドには、後述するロボット側シナリオデータが含まれる。 A robot control command, which is an example of a command signal, is input from the operation terminal 3 to the robot control unit 22. The robot control unit 22 generates the drive control signal described above based on the input robot control command, and outputs the generated drive control signal to the robot drive unit 21. That is, the robot 2 can operate according to the robot control command. There are various types of robot control commands depending on the type and control content of the robot drive unit 21 whose drive is to be controlled. For example, the robot control command includes a command for controlling the posture, position, and lighting state of the robot 2 and a command for causing the robot 2 to speak. This robot control command includes robot side scenario data described later.
(操作端末3)
 操作端末3は、ユーザが携帯するものであり、例えばタッチ機能を備えたタブレット端末である。この他、操作端末3は、スマートフォンや卓上のディスプレイ型の端末などでもよい。図1に示すように、操作端末3は、内蔵電池から供給される電力によって駆動され、方位センサ31と、動作生成部32と、シナリオDB(データベース)33と、内部状態値蓄積部34と、情報取得部35と、内部状態値取得部36と、シナリオ制御部37と、表示部38と、タッチパネル等の入力部39と、音声出力部40と、ロボット動作制御部41とを有する。なお、本実施形態に係る操作端末3が対話制御装置に対応する。
(Operation terminal 3)
The operation terminal 3 is carried by the user and is, for example, a tablet terminal having a touch function. In addition, the operation terminal 3 may be a smartphone, a desktop display type terminal, or the like. As shown in FIG. 1, the operation terminal 3 is driven by power supplied from a built-in battery, and includes an orientation sensor 31, an operation generation unit 32, a scenario DB (database) 33, an internal state value accumulation unit 34, An information acquisition unit 35, an internal state value acquisition unit 36, a scenario control unit 37, a display unit 38, an input unit 39 such as a touch panel, an audio output unit 40, and a robot operation control unit 41 are included. Note that the operation terminal 3 according to the present embodiment corresponds to a dialogue control device.
 方位センサ31は、操作端末3の方位を示す方位検出信号を動作生成部32へと出力する。方位センサ31は、検出された操作端末3の方位を示す方位検出信号を動作生成部32へと出力する。動作生成部32は、例えば、方位センサ31からの方位検出信号に基づいて、操作端末3が存在する方向を向くように、ロボット2の姿勢を制御するためのロボット制御コマンドを生成する。 The azimuth sensor 31 outputs an azimuth detection signal indicating the azimuth of the operation terminal 3 to the motion generation unit 32. The direction sensor 31 outputs a direction detection signal indicating the detected direction of the operation terminal 3 to the motion generation unit 32. For example, the motion generation unit 32 generates a robot control command for controlling the posture of the robot 2 so as to face the direction in which the operation terminal 3 exists based on the direction detection signal from the direction sensor 31.
 動作生成部32は、方位センサ31の出力に基づいてロボット2の動作を制御するロボット制御コマンドを生成する。動作生成部32は、生成されたロボット制御コマンドを、例えば、Wi-Fiなどの無線通信を介してロボット制御部22へと送信する。ロボット制御部22は、動作生成部32からのロボット制御コマンドを受信し、受信されたロボット制御コマンドに応じた駆動制御信号をロボット駆動部21へと出力することにより、ロボット2の動作の制御を行う。 The motion generation unit 32 generates a robot control command for controlling the motion of the robot 2 based on the output of the direction sensor 31. The motion generation unit 32 transmits the generated robot control command to the robot control unit 22 via wireless communication such as Wi-Fi, for example. The robot control unit 22 receives the robot control command from the motion generation unit 32, and outputs a drive control signal corresponding to the received robot control command to the robot drive unit 21, thereby controlling the operation of the robot 2. Do.
 図1に示すように、シナリオDB33には、ユーザとロボット2との間での対話を行うための複数の対話シナリオ情報が蓄積されている。対話シナリオ情報には、複数のシナリオ群(各シナリオ群は複数の対話シナリオを有する)や、複数の対話シナリオが含まれており、一つの対話シナリオは、ロボット2とユーザとの間でやり取りされる対話のストーリである。一つの対話シナリオには、ロボット2が対話、すなわち発話、動作などに用いるロボット側シナリオデータと、ユーザが対話、すなわち、操作端末3での選択に用いるユーザ側シナリオデータとで構成されている。ユーザ側シナリオデータは、ロボット側シナリオデータに対して応答可能な幾つかの対話シナリオデータを含んでおり、これらの中から任意のデータをユーザは入力部39にて選択することができる。そして、ユーザ側シナリオデータに示される対話シナリオの選択肢のそれぞれに、互いに異なる新たな下位のユーザ側シナリオデータが対応付けられている。すなわち、対話シナリオデータは、ロボット側シナリオデータとユーザ側シナリオデータが交互にノードとして結合されたツリー構造を有する。ツリー構造のうち、最上位から最下位にわたる所定の一連のノード群は、例えば典型的な対話に用いられる基本シナリオとして管理されている。また、他のノード群は、基本シナリオを補正する補正シナリオとして管理されていてもよい。 As shown in FIG. 1, the scenario DB 33 stores a plurality of pieces of dialogue scenario information for carrying out a dialogue between the user and the robot 2. The dialogue scenario information includes a plurality of scenario groups (each scenario group has a plurality of dialogue scenarios) and a plurality of dialogue scenarios. One dialogue scenario is exchanged between the robot 2 and the user. This is a story of dialogue. One dialogue scenario includes robot-side scenario data used by the robot 2 for dialogue, that is, speech and movement, and user-side scenario data used for dialogue, that is, selection by the operation terminal 3. The user-side scenario data includes some dialogue scenario data that can be responded to the robot-side scenario data, and the user can select arbitrary data from these through the input unit 39. Then, different lower-order user-side scenario data different from each other are associated with each of the dialogue scenario options shown in the user-side scenario data. That is, the dialogue scenario data has a tree structure in which the robot side scenario data and the user side scenario data are alternately coupled as nodes. In the tree structure, a predetermined series of nodes ranging from the highest level to the lowest level is managed as a basic scenario used for a typical dialogue, for example. Other node groups may be managed as a correction scenario for correcting the basic scenario.
 例えば、飲食店向けの対話シナリオデータは、飲食店におけるユーザの滞在状態を示すステータスデータに応じて複数のシナリオ群に区分されていてもよい。図1の例におけるステータスデータは、着席、注文、飲食物提供、食事終了、及び、会計の各行為をしたことを示す情報である。対話シナリオデータはこれらのステータスデータにより区分されるシナリオ群の中から提供される。 For example, dialogue scenario data for a restaurant may be divided into a plurality of scenario groups according to status data indicating a staying state of the user at the restaurant. The status data in the example of FIG. 1 is information indicating that the user has performed the following actions: sitting, ordering, food and drink provision, meal termination, and accounting. The dialogue scenario data is provided from the scenario group classified by these status data.
 内部状態値蓄積部34は、対話シナリオ情報に対応する対話シナリオの進行に伴って変化する内部状態を数値化した値を蓄積している。変化する内部状態には、正感情を示す第1内部状態と、負感情を示す第2内部状態とがあり、それぞれが数値化されている。例えば、正感情の度合いを数値の大きさで示している。つまり、数値が大きくなるほど、正感情の高ぶりが大きくなることを示している。同様に、負感情の度合いを数値の大きさで示している。つまり、数値が大きくなるほど、負感情の高ぶりが大きくなることを示している。また、ニュートラルな感情は、例えば0で数値化されている。なお、内部状態には、感情だけでなく、感覚、意図、興味など人の中に通常内在しうるあらゆる要素が含まれる。そのため、以降では内部状態として感情を中心に記載するが、適用できる内部状態はこれに限られない。 The internal state value accumulation unit 34 accumulates values obtained by quantifying the internal state that changes with the progress of the dialogue scenario corresponding to the dialogue scenario information. The changing internal state includes a first internal state indicating positive emotion and a second internal state indicating negative emotion, and each is quantified. For example, the degree of positive emotion is indicated by a numerical value. That is, the higher the numerical value, the higher the positive feeling. Similarly, the degree of negative emotion is indicated by a numerical value. In other words, the higher the numerical value, the higher the negative emotion. Moreover, the neutral feeling is digitized by 0, for example. It should be noted that the internal state includes not only emotions but also all elements that can normally exist in a person such as senses, intentions, and interests. Therefore, hereinafter, the emotion is mainly described as the internal state, but the applicable internal state is not limited to this.
 具体的には、第1内部状態は、例えばレストランなどの飲食店内での推奨メニューに関するユーザの嗜好または評価に対して得られる正感情であり、第2内部状態は、例えば推奨メニューに関するユーザの嗜好または評価に対して得られる負感情である。また、他の例では、第1内部状態は、例えばレストランなどの飲食店内で注文したメニューに関するユーザの評価、および接客サービスに関するユーザの評価、の少なくともいずれかに対して得られる正感情であり、第2内部状態は、注文したメニューに関するユーザの評価、および接客サービスに関するユーザの評価、の少なくともいずれかに対して得られる負感情である。本実施形態に係る内部状態値蓄積部34は、ユーザ側対話シナリオの選択肢に関連付けて、正感情及び負感情のそれぞれを示す数値を記憶している。これにより、内部状態値蓄積部34は、ユーザが選択した選択肢に対応する数値を出力可能である。 Specifically, the first internal state is a positive emotion obtained for the user's preference or evaluation regarding a recommended menu in a restaurant such as a restaurant, and the second internal state is a user's preference regarding a recommended menu, for example. Or negative emotions obtained for evaluation. In another example, the first internal state is a positive emotion obtained for at least one of a user's evaluation related to a menu ordered in a restaurant such as a restaurant and a user's evaluation related to a customer service, The second internal state is a negative emotion obtained for at least one of the user's evaluation regarding the ordered menu and the user's evaluation regarding the customer service. The internal state value accumulation unit 34 according to the present embodiment stores numerical values indicating positive emotions and negative emotions in association with options of the user-side dialogue scenario. Thereby, the internal state value storage unit 34 can output a numerical value corresponding to the option selected by the user.
 情報取得部35は、飲食店に来店したユーザと対話するための対話シナリオ情報をシナリオDB33内の複数の対話シナリオ情報の中から取得する。すなわち、この情報取得部35は、シナリオ制御部37の制御に従い、シナリオDB33に記憶される複数の対話シナリオ情報の中から、ユーザと対話するための対話シナリオ情報を取得し、シナリオ制御部37に出力する。 The information acquisition unit 35 acquires dialog scenario information for interacting with a user who has visited a restaurant from a plurality of dialog scenario information in the scenario DB 33. That is, the information acquisition unit 35 acquires dialogue scenario information for interacting with the user from the plurality of dialogue scenario information stored in the scenario DB 33 according to the control of the scenario control unit 37, and sends it to the scenario control unit 37. Output.
 内部状態値取得部36は、対話シナリオ情報に対応する対話シナリオの進行に伴って変化する内部状態を数値化した値を取得する。すなわち、この内部状態値取得部36は、シナリオ制御部37の制御に従い、内部状態値蓄積部34に記憶される対話シナリオの進行に伴って変化する内部状態を数値化した値を取得し、シナリオ制御部37に出力する。 The internal state value acquisition unit 36 acquires a value obtained by quantifying the internal state that changes with the progress of the dialog scenario corresponding to the dialog scenario information. That is, the internal state value acquisition unit 36 acquires a value obtained by quantifying the internal state that changes with the progress of the dialogue scenario stored in the internal state value storage unit 34 according to the control of the scenario control unit 37. Output to the control unit 37.
 シナリオ制御部37は、情報取得部35が取得した対話シナリオ情報に基づいて、対話シナリオを進行させる。また、シナリオ制御部37は、対話シナリオの進行に伴って、内部状態値取得部36を介して取得した内部状態を数値化した値の累積値を更新する。すなわち、シナリオ制御部37は、上述したように正感情を示す第1内部状態を数値化した値と、負感情を示す第2内部状態を数値化した値との少なくとも一方を、対話シナリオの進行に伴って累積する。そして、シナリオ制御部37は、内部状態値取得部36で取得された値の累積値に基づいて対話シナリオを補正する。また、シナリオ制御部37は、正感情を示す第1内部状態を数値化した値の累積値、及び負感情を示す第2内部状態を数値化した累積値を、動作生成部32を介して、ロボット動作制御部41に供給する。 The scenario control unit 37 advances the dialogue scenario based on the dialogue scenario information acquired by the information acquisition unit 35. Moreover, the scenario control part 37 updates the cumulative value of the value which digitized the internal state acquired via the internal state value acquisition part 36 with progress of a dialogue scenario. That is, as described above, the scenario control unit 37 determines at least one of the value obtained by quantifying the first internal state indicating positive emotion and the value obtained by quantifying the second internal state indicating negative emotion as the progress of the dialogue scenario. It accumulates with. Then, the scenario control unit 37 corrects the dialogue scenario based on the cumulative value of the values acquired by the internal state value acquisition unit 36. In addition, the scenario control unit 37 sends the cumulative value obtained by quantifying the first internal state indicating positive emotion and the cumulative value obtained by quantifying the second internal state indicating negative emotion via the motion generating unit 32. This is supplied to the robot operation control unit 41.
 一方で、シナリオ制御部37は、対話シナリオ情報に含まれるユーザ側シナリオデータに基づいて、ユーザ側の対話シナリオの選択肢を表示部38に表示させる。なお、対話シナリオの選択肢は、基本的に複数であるが、対話の段階によっては1つとなるときがあってもよい。また、シナリオ制御部37の詳細な制御動作例は、後述する。 On the other hand, the scenario control unit 37 causes the display unit 38 to display user-side dialog scenario options based on the user-side scenario data included in the dialog scenario information. Although there are basically a plurality of dialogue scenario options, there may be one choice depending on the dialogue stage. A detailed control operation example of the scenario control unit 37 will be described later.
 動作生成部32は、シナリオ制御部37から入力されたロボット側シナリオデータに基づいて、ロボット2に発話及び動作させるためのロボット制御コマンドを生成する。このとき、動作生成部32は、ロボット側シナリオデータに対応して、ロボット制御コマンドの指令内容に、発話音声のトーンや、眼球の点灯状態やロボット2の姿勢等の動作に関する指令を含める。 The motion generation unit 32 generates a robot control command for causing the robot 2 to speak and operate based on the robot side scenario data input from the scenario control unit 37. At this time, the motion generation unit 32 includes commands related to motions such as the tone of the utterance voice, the lighting state of the eyeball, and the posture of the robot 2 in the command content of the robot control command corresponding to the robot side scenario data.
 ユーザは、表示部38で表示された対話シナリオの選択肢の中から、入力部39を用いて所望の対話シナリオを選択することができる。すなわち、入力部39は、ユーザによって選択された対話シナリオを示す対話シナリオ選択信号を、シナリオ制御部37および音声出力部40に出力する。音声出力部40は、入力部39から入力された対話シナリオ選択信号に基づいて、選択された対話シナリオを音声出力する。 The user can select a desired dialogue scenario using the input unit 39 from the choices of the dialogue scenario displayed on the display unit 38. That is, the input unit 39 outputs a dialogue scenario selection signal indicating a dialogue scenario selected by the user to the scenario control unit 37 and the voice output unit 40. The voice output unit 40 outputs the selected dialogue scenario as a voice based on the dialogue scenario selection signal input from the input unit 39.
 ロボット動作制御部41は、対話の進行に伴い操作端末3から順に受信する内部状態を数値化した値の累積値に応じて、内部状態を表す行動をロボット2に行わせる。内部状態とは、感情および意図の少なくとも一方を含む状態である。換言すると、ロボット動作制御部41は、ロボット2の感情および意図の少なくとも一方を反映した行動をロボット2に行わせることが可能である。例えば、ロボット動作制御部41は、内部状態を数値化した値の累積値に応じて、ロボット2の声色、動作、目の色および目の光り方の少なくとも一つを変化させるロボット制御コマンドを生成する。本実施形態では、ロボット2の内部状態を数値化して、数値化した値の累積値に応じて、ロボット2に何らかの行動を行わせることを意図しており、その意図を達成するためにロボット動作制御部41を設けている。 The robot motion control unit 41 causes the robot 2 to perform an action representing the internal state in accordance with a cumulative value of values obtained by quantifying the internal state sequentially received from the operation terminal 3 as the dialogue progresses. The internal state is a state including at least one of emotion and intention. In other words, the robot motion control unit 41 can cause the robot 2 to perform an action reflecting at least one of the emotion and intention of the robot 2. For example, the robot motion control unit 41 generates a robot control command that changes at least one of the voice color, motion, eye color, and eye glow of the robot 2 according to the accumulated value of the internal state. To do. In the present embodiment, the internal state of the robot 2 is digitized, and the robot 2 is intended to perform some action according to the accumulated value of the digitized values. A control unit 41 is provided.
 例えば、ロボット動作制御部41は、正感情を表す動作として、ロボット2の、声の高さを上げたり、声を大きくしたり、光の点滅を早くしたり、楽しそうな口調に変えたり、目の色を暖色系に変えたり、口角を上げることで楽しそうな表情の顔に変えたり、腕を上げたり、顔を上向き加減にしたりしてもよい。一方で、ロボット動作制御部41は、負感情を表す動作として、ロボット2の、声の高さを下げたり、声を小さくしたり、光の点滅を遅くしたり、悲しそうな口調に変えたり、目の色を寒色系に変えたり、口角を下げることで悲しそうな表情の顔に変えたり、顔や体を俯き加減にしたり、腕を下げたり、肩を下げたりしてもよい。このように、対話のやり取りを通じて、ロボット2を、感情および意図の少なくとも一方を含むように動作させることで、対話の臨場感を高めるとともに、対話の違和感を抑制して、対話の自然さなどを生じさせることができる。これにより、ユーザは、ロボット2との対話による満足感を向上させることができる。 For example, the robot motion control unit 41 may increase the voice of the robot 2, increase the voice, speed up the blinking of light, change the tone to a pleasant tone, You can change the color of your eyes to a warm color, change your face to a more enjoyable expression by raising your corners, raise your arms, or adjust your face upwards. On the other hand, the robot motion control unit 41 may reduce the pitch of the voice of the robot 2, decrease the voice, slow the blinking of light, or change it to a sad tone as an action expressing negative emotion. You can change the color of the eyes to a cool color, change the face to a sad expression by lowering the corner of the mouth, whisper the face or body, lower the arm, or lower the shoulder. As described above, by operating the robot 2 so as to include at least one of emotion and intention through the exchange of dialogue, the realism of the dialogue is suppressed, and the naturalness of the dialogue is suppressed by suppressing the sense of incongruity of the dialogue. Can be generated. Thereby, the user can improve the satisfaction by the dialogue with the robot 2.
 さらにまた、ロボット動作制御部41は、動作生成部32から入力されたロボット制御コマンドと、内部状態を数値化した値の累積値とに基づき、ロボット制御コマンドを生成してもよい。例えば、ロボット動作制御部41は、動作生成部32から入力されたロボット制御コマンドを補正して別のロボット制御コマンドを生成してもよい。 Furthermore, the robot motion control unit 41 may generate a robot control command based on the robot control command input from the motion generation unit 32 and the accumulated value obtained by quantifying the internal state. For example, the robot motion control unit 41 may correct another robot control command input from the motion generation unit 32 and generate another robot control command.
 なお、本実施形態に係るロボット動作制御部41は、内部状態を数値化した値の累積値をシナリオ制御部37から入力されているが、これに限らず、ロボット動作制御部41内で内部状態を数値化した値の累積値を生成してもよい。 Note that the robot motion control unit 41 according to the present embodiment is input from the scenario control unit 37 with a cumulative value obtained by quantifying the internal state, but is not limited thereto, and the robot motion control unit 41 includes the internal state. A cumulative value obtained by quantifying the values may be generated.
 次に、図2に基づきシナリオ制御部37の詳細な制御動作例を説明する。図2は、シナリオ制御部37の制御動作例の表を示す図であり、選択肢A、B、Cは、対話シナリオの進行に伴って表示部38に表示される選択肢の中からユーザが選択した選択肢の情報を示している。正感情(α)は、対象となる選択肢をユーザが選択した場合の正感情を示す第1内部状態を数値化した値の累積値を示し、負感情(β)は、対象となる選択肢をユーザが選択した場合の負感情を示す第2内部状態を数値化した値の累積値を示している。 Next, a detailed control operation example of the scenario control unit 37 will be described with reference to FIG. FIG. 2 is a diagram showing a table of control operation examples of the scenario control unit 37, and the options A, B, and C are selected by the user from the options displayed on the display unit 38 as the dialogue scenario progresses. Information on options is shown. Positive emotion (α) indicates a cumulative value obtained by quantifying the first internal state indicating the positive emotion when the user selects the target option, and negative emotion (β) indicates the target option. The cumulative value of the value which digitized the 2nd internal state which shows the negative emotion at the time of selecting is shown.
 この図2に示すように、シナリオ制御部37は、内部状態値取得部36で取得された値の累積値に基づいて対話シナリオを補正する。例えば図2における表内の1行目に示すように、シナリオ制御部37は、正感情を示す第1内部状態を数値化した値の累積値が所定の閾値10以上の場合に、対話シナリオ情報をロボット2の正感情を考慮に入れて補正する。一方で、図2における表内の2行目に示すように、シナリオ制御部37は、負感情を示す第2内部状態を数値化した値の累積値が所定の閾値10以上の場合に、対話シナリオ情報をロボット2の負感情を考慮に入れて補正する。このように、シナリオ制御部37は、対話シナリオの進行に伴って、互いに相反する第1内部状態または第2内部状態を数値化した値をそれぞれ累積し、これら累積値に基づいて対話シナリオを補正する。すなわち、シナリオ制御部37は、情報取得部35を介して取得した対話シナリオ情報に対応する対話シナリオの進行に伴って、内部状態値取得部36で取得された値の累積値を更新し、累積値が閾値に達したときに対話シナリオを補正する。これにより、シナリオ制御部37は、対話シナリオをロボット2の内部状態に応じた対話シナリオに補正可能である。 As shown in FIG. 2, the scenario control unit 37 corrects the dialogue scenario based on the cumulative value acquired by the internal state value acquisition unit 36. For example, as shown in the first line in the table of FIG. 2, the scenario control unit 37 sets the dialogue scenario information when the cumulative value of the value obtained by quantifying the first internal state indicating positive emotion is equal to or greater than a predetermined threshold value 10. Is corrected in consideration of the positive emotion of the robot 2. On the other hand, as shown in the second row in the table of FIG. 2, the scenario control unit 37 interacts when the cumulative value of the value obtained by quantifying the second internal state indicating negative emotion is equal to or greater than the predetermined threshold value 10. The scenario information is corrected in consideration of the negative emotion of the robot 2. As described above, the scenario control unit 37 accumulates values obtained by quantifying the first internal state or the second internal state, which are mutually contradictory, as the dialogue scenario progresses, and corrects the dialogue scenario based on these accumulated values. To do. That is, the scenario control unit 37 updates the accumulated value of the value acquired by the internal state value acquiring unit 36 as the dialog scenario corresponding to the dialog scenario information acquired via the information acquiring unit 35 progresses, Correct the dialogue scenario when the value reaches the threshold. Thereby, the scenario control unit 37 can correct the conversation scenario into a conversation scenario according to the internal state of the robot 2.
 ここで、本実施形態に係るシナリオ補正とは、進行中の対話シナリオ情報内の対話シナリオの順番を変更したり、対話シナリオ情報を他の対話シナリオ情報に変更したり、ロボット2の動作制御に関する対話シナリオを変更したりすることを意味する。例えば、対話シナリオ情報をロボット2の正感情を考慮に入れて補正する場合には、より親しみをました感じの、例えば友人間で対話しているような軽い感じの対話シナリオ情報に変更する。一方で、対話シナリオ情報をロボット2の負感情を考慮に入れて補正する場合には、例えば、敬語調の重い感じの対話シナリオ情報に変更する。このように、シナリオ制御部37が行うシナリオ補正は、対話シナリオ情報を変更可能であるので、対話内容も変更できる。 Here, scenario correction according to the present embodiment refers to changing the order of dialogue scenarios in ongoing dialogue scenario information, changing dialogue scenario information to other dialogue scenario information, and controlling the operation of the robot 2. It means changing the dialogue scenario. For example, when the dialogue scenario information is corrected in consideration of the positive emotion of the robot 2, the dialogue scenario information is changed to a more familiar feeling, for example, a light feeling like the dialogue between friends. On the other hand, when correcting the dialogue scenario information in consideration of the negative emotion of the robot 2, for example, the dialogue scenario information is changed to dialogue scenario information with a feeling of heavy honorific tone. As described above, the scenario correction performed by the scenario control unit 37 can change the dialog scenario information, so that the dialog contents can also be changed.
 また、例えば、対話シナリオ情報をロボット2の正感情を考慮に入れて補正する場合には、ロボット制御コマンドに関するロボット側シナリオデータを補正し、ロボット2の動作をより大きくする。一方で、対話シナリオ情報をロボット2の負感情を考慮に入れて補正する場合には、ロボット制御コマンドに関するロボット側シナリオデータを補正し、ロボット2の動作をより緩慢にしたり、動作量を小さくしたりする。このように、シナリオ制御部37が、対話シナリオの進行に伴って変更されるロボット2の内部状態を示す値の累積値に応じて対話シナリオを補正するので、ロボット2とユーザとの対話のやりとりに臨場感を生じさせることが可能である。 Also, for example, when the dialogue scenario information is corrected in consideration of the positive emotion of the robot 2, the robot-side scenario data related to the robot control command is corrected, and the operation of the robot 2 is further increased. On the other hand, when the dialogue scenario information is corrected in consideration of the negative emotion of the robot 2, the robot side scenario data related to the robot control command is corrected to make the operation of the robot 2 slower or reduce the movement amount. Or As described above, the scenario control unit 37 corrects the dialogue scenario in accordance with the accumulated value of the value indicating the internal state of the robot 2 that is changed as the dialogue scenario progresses, so that the dialogue between the robot 2 and the user is exchanged. It is possible to create a sense of reality.
 さらにまた、シナリオ制御部37は、数値化した値を累積した累積値に応じて、シナリオを補正するので、これまでに生じた内部状態を示す値をシナリオ補正に反映可能である。換言すると、ロボット2内に蓄積された感情や意図を示す内部状態値をロボット2の対話や動作に反映可能である。例えば、負感情を高めるような対話が続いたときに人間が行うような口調の対話や動作をロボット2に行わせることが可能である。逆に、正感情を高めるような対話が続いたときに人が行うような口調の対話や動作をロボット2に行わせることも可能である。これにより、人間同士の対話で発生する感情的な要素をロボット2とユーザとの対話でも生じさせることが可能となる。 Furthermore, since the scenario control unit 37 corrects the scenario according to the accumulated value obtained by accumulating the digitized values, the value indicating the internal state generated so far can be reflected in the scenario correction. In other words, the internal state value indicating the emotion and intention stored in the robot 2 can be reflected in the dialogue and operation of the robot 2. For example, it is possible to cause the robot 2 to perform a tone-like dialogue or action that a human performs when a dialogue that enhances negative emotions continues. Conversely, it is also possible to cause the robot 2 to perform a tone-like dialogue or action that a person performs when a dialogue that enhances positive emotions continues. Thereby, it is possible to cause an emotional element generated in a dialogue between humans even in a dialogue between the robot 2 and the user.
 図2の例では、正感情の累積値と負感情の累積値がともに10未満の場合は、対話シナリオを進行させるようにしている。また、図2では省略しているが、正感情と負感情の少なくとも一方の累積値が10より大きい所定値(例えば20)に達した場合には、対話シナリオを切り替えるようにしてもよい。 In the example of FIG. 2, when both the cumulative value of positive emotion and the cumulative value of negative emotion are less than 10, the dialogue scenario is advanced. Although omitted in FIG. 2, the dialogue scenario may be switched when the cumulative value of at least one of positive emotion and negative emotion reaches a predetermined value (for example, 20) greater than 10.
 図3は、シナリオ制御部37における処理の流れの一例を示すフローチャートであり、この図3に基づきシナリオ制御部37における処理の流れの一例を説明する。ここでは、店舗システム5を介してハンディ端末4から入力されたステータスデータに応じた補正シナリオをシナリオ制御部37がシナリオDB33から読み出して進行させる例について説明する。 FIG. 3 is a flowchart showing an example of the process flow in the scenario control unit 37. An example of the process flow in the scenario control unit 37 will be described with reference to FIG. Here, an example will be described in which the scenario control unit 37 reads out and advances a correction scenario corresponding to the status data input from the handy terminal 4 via the store system 5 from the scenario DB 33.
 シナリオ制御部37は、読み出された対話シナリオ情報に含まれるユーザ側シナリオデータに基づいて、ユーザ側の対話シナリオの選択肢を表示部38に表示させる。続いて、ユーザは、表示部38に表示された特定の選択肢を選択する(ステップS300)。この場合、対話シナリオ情報に含まれるロボット側シナリオデータは、動作生成部32に出力される。 The scenario control unit 37 causes the display unit 38 to display the user-side dialogue scenario options based on the user-side scenario data included in the read dialogue scenario information. Subsequently, the user selects a specific option displayed on the display unit 38 (step S300). In this case, the robot-side scenario data included in the dialogue scenario information is output to the motion generation unit 32.
 次に、シナリオ制御部37は、ユーザが選択した選択肢に基づき、内部状態値取得部36を介して内部状態値蓄積部34からロボット2の内部状態値を取得する(ステップS302)。すなわち、シナリオ制御部37は、第1内部状態を数値化した値及び第2内部状態を数値化した値を取得する。 Next, the scenario control unit 37 acquires the internal state value of the robot 2 from the internal state value accumulation unit 34 via the internal state value acquisition unit 36 based on the option selected by the user (step S302). That is, the scenario control unit 37 acquires a value obtained by quantifying the first internal state and a value obtained by quantifying the second internal state.
 次に、シナリオ制御部37は、対話シナリオの進行に伴って、第1内部状態を数値化した値及び第2内部状態を数値化した値のそれぞれを累積し、累積値を更新する(ステップS304)。すなわち、シナリオ制御部37は、正感情(α)としての累積値、及び負感情(β)としての累積値をそれぞれ更新する。 Next, as the dialogue scenario progresses, the scenario control unit 37 accumulates the value obtained by quantifying the first internal state and the value obtained by quantifying the second internal state, and updates the accumulated value (step S304). ). That is, the scenario control unit 37 updates the cumulative value as the positive emotion (α) and the cumulative value as the negative emotion (β), respectively.
 次に、シナリオ制御部37は、正感情(α)としての累積値、及び負感情(β)としての累積値のいずれかが閾値、例えば10を超えたか否かを判断する(ステップS306)。正感情(α)としての累積値、及び負感情(β)としての累積値のいずれかが閾値以上になる場合(ステップS306:YES)、シナリオ制御部37は、対話シナリオを補正する(ステップS308)。一方で、正感情(α)としての累積値、及び負感情(β)としての累積値のいずれも閾値未満の場合(ステップS306:NO)、シナリオ制御部37は、現在実行中の対話シナリオを継続して進行させる(ステップS310)。 Next, the scenario control unit 37 determines whether any of the cumulative value as the positive emotion (α) and the cumulative value as the negative emotion (β) exceeds a threshold, for example, 10 (step S306). When either the cumulative value as the positive emotion (α) or the cumulative value as the negative emotion (β) is equal to or greater than the threshold (step S306: YES), the scenario control unit 37 corrects the dialogue scenario (step S308). ). On the other hand, when both the cumulative value as positive emotion (α) and the cumulative value as negative emotion (β) are less than the threshold value (step S306: NO), the scenario control unit 37 selects the currently executed dialogue scenario. Continue to proceed (step S310).
 次に、シナリオ制御部37は、対話シナリオが終了したか否かを判断する(ステップS312)。シナリオ制御部37は、対話シナリオが終了したと判断した場合(ステップS312:YES)、シナリオ制御部37は、対話シナリオを終了する。一方で、シナリオ制御部37は、対話シナリオが終了していないと判断した場合(ステップS312:NO)、シナリオ制御部37は、(ステップS300)からの処理を繰り返す。 Next, the scenario control unit 37 determines whether or not the dialogue scenario has ended (step S312). When the scenario control unit 37 determines that the dialogue scenario has ended (step S312: YES), the scenario control unit 37 ends the dialogue scenario. On the other hand, when the scenario control unit 37 determines that the conversation scenario has not ended (step S312: NO), the scenario control unit 37 repeats the processing from (step S300).
 このように、シナリオ制御部37は、対話シナリオの進行に伴って、互いに相反する正感情(α)としての累積値、及び負感情(β)としての累積値をそれぞれ累積し、これら累積値のいずれかが閾値以上の場合に対話シナリオを補正する。一方で、シナリオ制御部37は、これら累積値のいずれもが閾値未満の場合には対話シナリオを進行させる。これにより、累積値が閾値以上になると、シナリオ制御部37は、対話シナリオをロボット2の内部状態に応じた対話シナリオに補正可能であり、ロボット2の内部状態に応じた対話シナリオを選択可能である。 As described above, the scenario control unit 37 accumulates a cumulative value as a positive emotion (α) and a cumulative value as a negative emotion (β), which are opposite to each other, as the dialogue scenario progresses. If any of them is equal to or greater than the threshold value, the dialogue scenario is corrected. On the other hand, the scenario control unit 37 advances the dialogue scenario when any of these accumulated values is less than the threshold value. Thereby, when the accumulated value is equal to or greater than the threshold value, the scenario control unit 37 can correct the conversation scenario to a conversation scenario according to the internal state of the robot 2 and can select the conversation scenario according to the internal state of the robot 2. is there.
 図4は、対話シナリオの進行例の一部を示すシーケンス図であり、この図4に基づきシナリオの進行例を説明する。図4の対話シナリオ1-1~1-4は、対話シナリオ1に含まれる部分的なシナリオである。
 図4では、正感情の累積値の初期値をαとし、負感情の累積値の初期値をβとして説明する。この図4に示すように、対話シナリオ1は、対話シナリオ1-1から開始される。対話シナリオ1-1におけるユーザの選択肢は、選択肢1及び選択肢2である。ユーザが選択肢1を選択した場合には、正感情を示す第1内部状態値は0であり、負感情を示す第2内部状態値は1である。つまり、ロボット2にとっては、選択肢1は、好ましい選択ではないことを示している。このため、ロボット2の負感情の累積値がβ+1に上がる。
FIG. 4 is a sequence diagram showing a part of the progress example of the dialogue scenario. The progress example of the scenario will be described with reference to FIG. Dialog scenarios 1-1 to 1-4 in FIG. 4 are partial scenarios included in the dialog scenario 1.
In FIG. 4, the initial value of the cumulative value of positive emotion is assumed to be α, and the initial value of the cumulative value of negative emotion is assumed to be β. As shown in FIG. 4, the dialogue scenario 1 starts from the dialogue scenario 1-1. The user's options in the dialogue scenario 1-1 are option 1 and option 2. When the user selects option 1, the first internal state value indicating positive emotion is 0, and the second internal state value indicating negative emotion is 1. That is, for the robot 2, option 1 indicates that it is not a preferable selection. For this reason, the cumulative value of the negative emotion of the robot 2 increases to β + 1.
 一方で、ユーザが選択肢2を選択した場合には、正感情を示す第1内部状態値は1であり、負感情を示す第2内部状態値は0である。つまり、ロボット2にとっては、選択肢2は、好ましい選択であることを示している。このため、ロボット2の正感情の累積値がα+1に上がる。 On the other hand, when the user selects option 2, the first internal state value indicating positive emotion is 1, and the second internal state value indicating negative emotion is 0. That is, for the robot 2, option 2 is a preferable selection. For this reason, the cumulative value of the positive emotion of the robot 2 increases to α + 1.
 次に、選択肢2の選択に従い、対話シナリオ1-3が進行する。対話シナリオ1-3におけるユーザの選択肢は、選択肢3~7である。ここでは、選択肢3が選択されている。選択肢3では、正感情を示す第1内部状態値は1であり、負感情を示す第2内部状態値は1である。つまり、ロボット2にとっては、選択肢3は、正感情を高め、負感情を下げる選択であることを示している。このため、ロボット2の正感情の累積値がα+2に上がり、負感情の累積値がβ+1に上がる。 Next, according to the selection of option 2, the dialogue scenario 1-3 proceeds. The user's options in the dialogue scenario 1-3 are options 3-7. Here, option 3 is selected. In option 3, the first internal state value indicating positive emotion is 1, and the second internal state value indicating negative emotion is 1. That is, for the robot 2, the option 3 indicates that the positive emotion is increased and the negative emotion is decreased. For this reason, the cumulative value of the positive emotion of the robot 2 increases to α + 2, and the cumulative value of the negative emotion increases to β + 1.
 次に、選択肢3の選択に従い、対話シナリオ1-4が進行する。対話シナリオ1-4におけるユーザの選択肢は、選択肢8と選択肢9である。ユーザが選択肢8を選択した場合には、正感情を示す第1内部状態値は3であり、負感情を示す第2内部状態値は0である。つまり、ロボット2にとっては、選択肢8は、非常に好ましい選択であることを示している。このため、ロボット2の正感情の累積値がα+5まで上がる。これにより、正感情の累積値が閾値以上になる。このため、シナリオ制御部37は、対話シナリオ1をロボット2の正感情を考慮に入れて補正したシナリオ補正を得る。このように、選択肢8の選択に従い、シナリオ補正のシナリオが続いて進行する。 Next, the dialogue scenario 1-4 proceeds according to the selection of the option 3. The user's options in the dialogue scenario 1-4 are option 8 and option 9. When the user selects option 8, the first internal state value indicating positive emotion is 3, and the second internal state value indicating negative emotion is 0. That is, for the robot 2, option 8 indicates that it is a very preferable choice. For this reason, the cumulative value of the positive emotion of the robot 2 increases to α + 5. As a result, the cumulative value of positive emotion becomes equal to or greater than the threshold value. For this reason, the scenario control unit 37 obtains a scenario correction in which the dialogue scenario 1 is corrected in consideration of the positive emotion of the robot 2. As described above, the scenario correction scenario proceeds in accordance with the selection of the option 8.
 一方で、ユーザが選択肢9を選択した場合には、正感情を示す第1内部状態値は0であり、負感情を示す第2内部状態値は2である。つまり、ロボット2にとっては、選択肢9は、あまり好ましい選択ではないことを示している。このため、ロボット2の負感情の累積値がβ+3に上がる。このように、選択肢毎に第1内部状態値、及び第2内部状態値が対応付けられ、選択肢の選択に応じて正感情を示す第1内部状態を数値化した値の累積値、負感情を示す第2内部状態を数値化した値の累積値が更新されていく。 On the other hand, when the user selects option 9, the first internal state value indicating positive emotion is 0, and the second internal state value indicating negative emotion is 2. That is, for the robot 2, the option 9 is not a preferable choice. For this reason, the cumulative value of the negative emotion of the robot 2 increases to β + 3. In this way, the first internal state value and the second internal state value are associated with each option, and the cumulative value and negative emotion of the value obtained by quantifying the first internal state indicating positive emotion according to the selection of the option are displayed. The accumulated value of the value obtained by quantifying the second internal state shown is updated.
 次に、図4における対話シナリオの具体的な適用例について説明する。図5は、対話シナリオ1-1に対応するシーンを模式的に示す図である。この図5に示すように、対話は、ロボット2と操作端末3を用いて行われる。操作端末3は、例えば、表示部38と、表示部38とともに設置された透明なタッチパネル式の入力部39と、を備えるタブレット型の携帯端末である。ロボット2からは、情報取得部35により取得された対話シナリオ1-1のロボット側シナリオデータに基づいて、「なにを注文する?」という発話が実行される。一方で、シナリオ制御部37は、対話シナリオ1-1に含まれるユーザ側シナリオデータに基づいて、ユーザ側の対話シナリオの選択肢を表示部38に表示させている。ここでの選択肢は選択肢500と選択肢502である。選択肢500が選択されると、ロボット2の負感情の累積値が1上がり、選択肢502が選択されると、ロボット2の正感情の累積値が1上がる。なお、ロボット2の胴体部2aは、台座2b上に、鉛直方向の軸回りに回転可能に支持されており、対話シナリオに従ってロボット2の胴体部2aが回転する場合もある。 Next, a specific application example of the dialogue scenario in FIG. 4 will be described. FIG. 5 is a diagram schematically showing a scene corresponding to the dialogue scenario 1-1. As shown in FIG. 5, the dialogue is performed using the robot 2 and the operation terminal 3. The operation terminal 3 is, for example, a tablet-type portable terminal that includes a display unit 38 and a transparent touch panel type input unit 39 installed together with the display unit 38. Based on the robot-side scenario data of the dialogue scenario 1-1 acquired by the information acquisition unit 35, the utterance “What to order?” Is executed from the robot 2. On the other hand, the scenario control unit 37 causes the display unit 38 to display user-side dialog scenario options based on the user-side scenario data included in the dialog scenario 1-1. The options here are option 500 and option 502. When the option 500 is selected, the cumulative negative emotion value of the robot 2 increases by 1. When the option 502 is selected, the positive emotion cumulative value of the robot 2 increases by one. Note that the body 2a of the robot 2 is supported on the pedestal 2b so as to be rotatable about a vertical axis, and the body 2a of the robot 2 may rotate according to a dialogue scenario.
 図6は、対話シナリオ1-3に対応するシーンを模式的に示す図である。この図6に示すように、シナリオ制御部37は、情報取得部35により取得された対話シナリオ1-3に含まれるユーザ側シナリオデータに基づいて、ユーザ側の対話シナリオの選択肢を表示部38に表示させている。ここでの選択肢は、選択肢600、602、604、606、608である。例えば選択肢600、602、608は、推奨メニューに関するユーザの嗜好に関する項目であり、選択肢604、606は、推奨メニューに関するユーザの評価に関する項目である。これから分かるように、本実施形態に係る第1内部状態は、注文したメニューに関するユーザの評価、および接客サービスに関するユーザの評価、の少なくともいずれかに対して得られる正感情であり、第2内部状態は、注文したメニューに関するユーザの評価、および接客サービスに関するユーザの評価、の少なくともいずれかに対して得られる負感情である。 FIG. 6 is a diagram schematically showing a scene corresponding to the dialogue scenario 1-3. As shown in FIG. 6, the scenario control unit 37 displays the user-side dialog scenario options on the display unit 38 based on the user-side scenario data included in the dialog scenario 1-3 acquired by the information acquisition unit 35. It is displayed. The options here are options 600, 602, 604, 606, and 608. For example, the options 600, 602, and 608 are items related to user preferences regarding the recommended menu, and the options 604 and 606 are items related to user evaluation regarding the recommended menu. As can be seen from this, the first internal state according to the present embodiment is a positive emotion obtained for at least one of the user's evaluation regarding the ordered menu and the user's evaluation regarding the customer service, and the second internal state Is a negative emotion obtained for at least one of the user's evaluation regarding the ordered menu and the user's evaluation regarding the customer service.
 具体的には、選択肢600が選択されると、ロボット2の正感情の累積値が1上がり、負感情の累積値が1上がる。選択肢602が選択されると、ロボット2の正感情の累積値が2上がり、負感情の累積値は変わらない。選択肢604が選択されると、ロボット2の正感情の累積値は変わらず、負感情の累積値が1上がる。選択肢606が選択されても、ロボット2の正感情及び負感情の累積値は共に変わらない。選択肢608が選択されると、ロボット2の正感情の累積値が3上がるが、負感情の累積値は変わらない。 Specifically, when the option 600 is selected, the cumulative value of the positive emotion of the robot 2 is increased by 1, and the cumulative value of the negative emotion is increased by 1. When the option 602 is selected, the cumulative value of the positive emotion of the robot 2 increases by 2, and the cumulative value of the negative emotion does not change. When the option 604 is selected, the cumulative value of the positive emotion of the robot 2 does not change, and the cumulative value of the negative emotion increases by one. Even if the option 606 is selected, the cumulative values of the positive emotion and negative emotion of the robot 2 are not changed. When the option 608 is selected, the cumulative value of the positive emotion of the robot 2 increases by 3, but the cumulative value of the negative emotion does not change.
 図7は、正感情を考慮したシナリオ補正を行った対話シナリオに対応するシーンを模式的に示す図である。この図7に示すように、ロボット2は、対話シナリオに含まれるロボット側シナリオデータに従い発言をしている。この場合、正感情を考慮したシナリオ補正を行っているので、ロボット2の動作が正感情を示す動作に補正されている。ここでは、ロボット2の表情が笑顔になり、声が高くなるようにシナリオ補正されている。 FIG. 7 is a diagram schematically showing a scene corresponding to a dialogue scenario in which scenario correction considering positive emotion is performed. As shown in FIG. 7, the robot 2 speaks according to the robot side scenario data included in the dialogue scenario. In this case, since the scenario correction is performed in consideration of the positive emotion, the operation of the robot 2 is corrected to the operation indicating the positive emotion. Here, the scenario is corrected so that the facial expression of the robot 2 is a smile and the voice is high.
 また、シナリオ制御部37は、補正シナリオに含まれるユーザ側シナリオデータに基づいて、ユーザ側の対話シナリオの選択肢を表示部38に表示させている。ここでの選択肢は選択肢700と選択肢702である。選択肢700が選択されると、ロボット2の負感情の累積値が1上がり、選択肢702が選択されると、ロボット2の正感情の累積値が1上がる。 Also, the scenario control unit 37 causes the display unit 38 to display user-side dialog scenario options based on the user-side scenario data included in the correction scenario. The options here are option 700 and option 702. When the option 700 is selected, the cumulative negative emotion value of the robot 2 increases by 1. When the option 702 is selected, the positive emotion cumulative value of the robot 2 increases by one.
 図8は、対話シナリオ1-5に対応するシーンを模式的に示す図である。この図8に示すように、ロボット2は、対話シナリオ1-5に含まれるロボット側シナリオデータに従い「気に入らなかった?」と発言している。また、選択9(図4)により、負感情を示す第2内部状態値は2上がるので、ロボット2は、負感情を示す動作を行っている。ここでは、一つ前の対話シナリオ1-4のロボット2の表情と比較し、表情が暗くなり、声が低くなるようにロボット2の動作が制御されている。 FIG. 8 is a diagram schematically showing a scene corresponding to the dialogue scenario 1-5. As shown in FIG. 8, the robot 2 says “Did you like it?” According to the robot-side scenario data included in the dialogue scenario 1-5. In addition, since the second internal state value indicating negative emotion is increased by 2 by selection 9 (FIG. 4), the robot 2 is performing an operation indicating negative emotion. Here, the operation of the robot 2 is controlled so that the expression becomes darker and the voice becomes lower than the expression of the robot 2 in the previous dialogue scenario 1-4.
 ここで、図9及び図10に基づき、食後の対話シナリオの一例を説明する。図9は、メニューに対するユーザの評価に関するシーンを模式的に示す図である。図10は、客サービスに対するユーザの評価に関するシーンを模式的に示す図である。この図9に示すように、ロボット2は、食後の対話シナリオに含まれるロボット側シナリオデータに従い「食事おいしかった?」と発言している。これから分かるように、例えば選択肢900、902は、注文したメニューに関するユーザの評価に関する項目である。例えば、選択肢900が選択されると、ロボット2の正感情の累積値が3上がり、ロボット2の負感情の累積値は変わらない。一方で、選択肢902が選択されると、ロボット2の負感情の累積値が1上がり、ロボット2の正感情の累積値は変わらない。 Here, an example of a post-meal dialogue scenario will be described based on FIG. 9 and FIG. FIG. 9 is a diagram schematically illustrating a scene relating to user evaluation of a menu. FIG. 10 is a diagram schematically showing a scene related to user evaluation for customer service. As shown in FIG. 9, the robot 2 makes a statement “Is the meal delicious?” According to the robot-side scenario data included in the post-meal dialogue scenario. As can be seen, for example, the options 900 and 902 are items relating to user evaluation regarding the ordered menu. For example, when the option 900 is selected, the cumulative value of the positive emotion of the robot 2 is increased by 3, and the cumulative value of the negative emotion of the robot 2 is not changed. On the other hand, when the option 902 is selected, the cumulative value of the negative emotion of the robot 2 increases by 1, and the cumulative value of the positive emotion of the robot 2 does not change.
 また、図10に示すように、ロボット2は、食後の対話シナリオに含まれるロボット側シナリオデータに従い「楽しめた?」と発言している。これから分かるように、例えば選択肢1000、1002は、注文したメニューに関するユーザの評価に関する項目である。すなわち、本実施形態に係る第1内部状態は、注文したメニューに関するユーザの評価、および接客サービスに関するユーザの評価、の少なくともいずれかに対して得られる正感情を含み、第2内部状態は、注文したメニューに関するユーザの評価、および接客サービスに関するユーザの評価、の少なくともいずれかに対して得られる負感情を含む。例えば、選択肢1000が選択されると、ロボット2の正感情の累積値が3上がり、ロボット2の負感情の累積値は変わらない。一方で、選択肢1002が選択されると、ロボット2の負感情の累積値が1上がり、ロボット2の正感情の累積値は変わらない。 Further, as shown in FIG. 10, the robot 2 remarks “Has you enjoyed it” according to the robot side scenario data included in the post-meal dialogue scenario? As can be seen, for example, the options 1000 and 1002 are items relating to user evaluation regarding the ordered menu. That is, the first internal state according to the present embodiment includes positive emotions obtained for at least one of user evaluation regarding the ordered menu and user evaluation regarding the customer service, and the second internal state is the order Negative feelings obtained for at least one of the user's evaluation regarding the selected menu and the user's evaluation regarding the customer service. For example, when the option 1000 is selected, the cumulative value of the positive emotion of the robot 2 is increased by 3, and the cumulative value of the negative emotion of the robot 2 is not changed. On the other hand, when the option 1002 is selected, the cumulative value of the negative emotion of the robot 2 increases by 1, and the cumulative value of the positive emotion of the robot 2 does not change.
 以上のように、本実施形態によれば、飲食店に来店したユーザが飲食店内に設置されたロボット2と、操作端末3を介して適切な対話を行うことができる。すなわち、本実施形態では、シナリオ制御部37が、対話シナリオの進行に伴って値が変更される内部状態を示す値の累積値に基づいて対話シナリオを補正することとした。これにより、シナリオ制御部37は、対話シナリオをロボット2の内部状態に応じた対話シナリオに補正可能であり、ロボット2との対話のやりとりに対話感および臨場感を持たせることが可能である。これにより、対話の不自然さが解消されて、ユーザの満足度を高めることが可能となる。 As described above, according to the present embodiment, a user who visits a restaurant can perform an appropriate dialogue with the robot 2 installed in the restaurant and the operation terminal 3. That is, in the present embodiment, the scenario control unit 37 corrects the dialogue scenario based on the accumulated value of the value indicating the internal state whose value is changed as the dialogue scenario progresses. Thereby, the scenario control unit 37 can correct the dialogue scenario into a dialogue scenario according to the internal state of the robot 2, and can give a sense of dialogue and a sense of reality to the exchange of dialogue with the robot 2. Thereby, the unnaturalness of the dialogue is eliminated, and the user's satisfaction can be increased.
(第1変形例)
 図11は、第1変形例に係るロボット制御システム1を示すブロック図である。上述した実施形態においては、シナリオDB33及び内部状態値蓄積部34は、操作端末3内に設置されているものであったが、これには限られない。例えば、図11に示すように、シナリオDB33及び内部状態値蓄積部34は、操作端末3の外部にあるものでもよい。そして、シナリオ制御部37は、ネットワークを介してシナリオDB33及び内部状態値蓄積部34と接続される。ネットワークと、操作端末3又はシナリオDB33は、有線又は無線の通信を介して接続される。同様に、ネットワークと、操作端末3又は内部状態値蓄積部34は、有線又は無線の通信を介して接続される。
(First modification)
FIG. 11 is a block diagram showing the robot control system 1 according to the first modification. In the above-described embodiment, the scenario DB 33 and the internal state value storage unit 34 are installed in the operation terminal 3, but are not limited thereto. For example, as shown in FIG. 11, the scenario DB 33 and the internal state value storage unit 34 may be outside the operation terminal 3. The scenario control unit 37 is connected to the scenario DB 33 and the internal state value storage unit 34 via a network. The network and the operation terminal 3 or the scenario DB 33 are connected via wired or wireless communication. Similarly, the network and the operation terminal 3 or the internal state value storage unit 34 are connected via wired or wireless communication.
 また、上述した実施形態においては、ロボット動作制御部41は、操作端末3内に設置されているものであったが、これには限られない。例えば、図11に示すように、ロボット動作制御部41は、ロボット2内にあるものでもよい。このような構成にしても、上述した実施形態における動作と大きく異なることはなく、本変形例においても、図3に示すフローチャートにしたがって対話の処理が実行される。 In the above-described embodiment, the robot motion control unit 41 is installed in the operation terminal 3, but is not limited thereto. For example, as shown in FIG. 11, the robot motion control unit 41 may be in the robot 2. Even with such a configuration, there is no significant difference from the operation in the above-described embodiment, and in this modification as well, the dialogue process is executed according to the flowchart shown in FIG.
 本変形例によれば、シナリオDB33を操作端末3内ではなく外部に有することにより、操作端末3にてシナリオDBの管理を行う必要がなくなるとともに、シナリオデータの一元管理をすることも可能となる。さらに、シナリオDB33がクラウド上に存在し、特定多数の編集者がシナリオデータを編集、あるいは、追加することが可能であるように設定することにより、よりバリエーションに富んだ対話内容を選択することも可能となる。 According to the present modification, by having the scenario DB 33 outside the operation terminal 3, it is not necessary to manage the scenario DB at the operation terminal 3, and the scenario data can be centrally managed. . Furthermore, by setting the scenario DB 33 on the cloud so that a specific number of editors can edit or add scenario data, it is possible to select more varied dialogue contents. It becomes possible.
(第2変形例)
 図12は、第2変形例に係るロボット制御システム1の構成を示す図である。この図12に示すように、動作生成部32、情報取得部35、内部状態値取得部36、及びシナリオ制御部37は、操作端末3ではなく、ロボット2内に設けられていてもよい。すなわち、主要な制御機能を操作端末3ではなく、ロボット2に設けてもよい。動作生成部32、情報取得部35、内部状態値取得部36、及びシナリオ制御部37の動作は、上述した実施形態における動作と大きく異なることはなく、本変形例においても、図3に示すフローチャートにしたがって対話の処理が実行される。
(Second modification)
FIG. 12 is a diagram illustrating a configuration of the robot control system 1 according to the second modification. As illustrated in FIG. 12, the motion generation unit 32, the information acquisition unit 35, the internal state value acquisition unit 36, and the scenario control unit 37 may be provided in the robot 2 instead of the operation terminal 3. That is, the main control function may be provided in the robot 2 instead of the operation terminal 3. The operations of the operation generation unit 32, the information acquisition unit 35, the internal state value acquisition unit 36, and the scenario control unit 37 are not significantly different from the operations in the above-described embodiment, and the flowchart shown in FIG. The dialogue process is executed according to the above.
 本変形例によれば、操作端末3の構成を簡略化でき、また、操作端末3からロボット2へと送信するデータ量を削減できるため、操作端末3とロボット2との間の通信回線の性能が低い場合でも、支障なくロボット2との対話を行うことができる。なお、動作生成部32は、操作端末3及びロボット2の外部に設けられていてもよく、例えば、店舗システム5に内蔵されていてもよいし、店舗システム5とは別個の通信機器内に設けてもよい。 According to this modification, the configuration of the operation terminal 3 can be simplified, and the amount of data transmitted from the operation terminal 3 to the robot 2 can be reduced, so that the performance of the communication line between the operation terminal 3 and the robot 2 can be reduced. Even when the value is low, dialogue with the robot 2 can be performed without any trouble. The motion generation unit 32 may be provided outside the operation terminal 3 and the robot 2. For example, the motion generation unit 32 may be built in the store system 5 or provided in a communication device separate from the store system 5. May be.
 なお、上述した全ての実施形態及び変形例においては、対話は、ユーザ、又は、ロボット2が何かしらの情報を出力している状態として説明したが、これには限られない。例えば、食事中であれば、ロボット2との対話の必要性は低くなるので、このような場合においては、無言のシナリオをあらかじめ準備しておき、この無言のシナリオを出力するようにしてもよい。上述した実施形態及び変形例の説明においては、この無言のシナリオも、シナリオデータとして含まれるものとする。 In all the embodiments and modifications described above, the dialogue is described as a state in which the user or the robot 2 outputs some information, but is not limited thereto. For example, since it is less necessary to interact with the robot 2 during a meal, a silent scenario may be prepared in advance and the silent scenario may be output. . In the above description of the embodiment and the modification, it is assumed that this silent scenario is also included as scenario data.
 上述した全ての実施形態及び変形例に係るロボット制御システム1は、既述したように飲食サービスに好適に適用することができるが、飲食サービス以外の場面に適用されてもよい。 The robot control system 1 according to all the embodiments and the modifications described above can be suitably applied to the eating and drinking service as described above, but may be applied to scenes other than the eating and drinking service.
 なお、動作生成部32は、操作端末3およびロボット2の外部に設けられていてもよく、例えば、店舗システム5に内蔵されていてもよいし、店舗システム5とは別個の通信機器内に設けてもよい。また、操作端末3は、店舗システム5との通信によって飲食物を注文する機能を備えていてもよい。 The motion generation unit 32 may be provided outside the operation terminal 3 and the robot 2. For example, the motion generation unit 32 may be built in the store system 5 or provided in a communication device separate from the store system 5. May be. Further, the operation terminal 3 may have a function of ordering food and drink through communication with the store system 5.
 また、ロボット制御システム1は、既述したように飲食サービスに好適に適用することができるが、飲食サービス以外の場面に適用されてもよい。 Moreover, as described above, the robot control system 1 can be suitably applied to a food and drink service, but may be applied to scenes other than the food and drink service.
 本発明の態様は、上述した個々の実施形態に限定されるものではなく、当業者が想到しうる種々の変形も含むものであり、本発明の効果も上述した内容に限定されない。すなわち、特許請求の範囲に規定された内容およびその均等物から導き出される本発明の概念的な思想と趣旨を逸脱しない範囲で種々の追加、変更および部分的削除が可能である。 The aspects of the present invention are not limited to the individual embodiments described above, but include various modifications that can be conceived by those skilled in the art, and the effects of the present invention are not limited to the contents described above. That is, various additions, modifications, and partial deletions can be made without departing from the concept and spirit of the present invention derived from the contents defined in the claims and equivalents thereof.
1 ロボット制御システム、2 ロボット、3 操作端末、34 内部状態値蓄積部、36 内部状態値取得部、37 シナリオ制御部、41 ロボット動作制御部 1 robot control system, 2 robots, 3 operation terminals, 34 internal state value storage unit, 36 internal state value acquisition unit, 37 scenario control unit, 41 robot operation control unit

Claims (10)

  1.  飲食店に来店した顧客と対話するための対話シナリオ情報を複数の対話シナリオ情報から取得する情報取得部と、
     前記取得した対話シナリオ情報に対応する対話シナリオの進行に伴って変化する内部状態を数値化した値を取得する内部状態値取得部と、
     前記取得した対話シナリオ情報に対応する対話シナリオの進行に伴って、前記内部状態値取得部で取得された値の累積値に基づいて対話シナリオを補正するシナリオ制御部と、を備える対話制御装置。
    An information acquisition unit that acquires dialogue scenario information for interacting with a customer who has visited a restaurant from a plurality of dialogue scenario information;
    An internal state value acquisition unit that acquires a value obtained by quantifying the internal state that changes with the progress of the conversation scenario corresponding to the acquired conversation scenario information;
    A scenario control unit comprising: a scenario control unit that corrects a dialog scenario based on a cumulative value of values acquired by the internal state value acquisition unit as the dialog scenario corresponding to the acquired dialog scenario information progresses.
  2.  前記シナリオ制御部は、対話シナリオの進行に伴って、互いに相反する第1内部状態または第2内部状態を数値化した値をそれぞれ累積する、請求項1に記載の対話制御装置。 The dialogue control device according to claim 1, wherein the scenario control unit accumulates values obtained by quantifying the first internal state or the second internal state that are mutually contradictory with the progress of the dialogue scenario.
  3.  前記第1内部状態は、推奨メニューに関する前記顧客の嗜好または評価に対して得られる正感情であり、
     前記第2内部状態は、推奨メニューに関する前記顧客の嗜好または評価に対して得られる負感情である、請求項2に記載の対話制御装置。
    The first internal state is a positive emotion obtained for the customer's preference or evaluation regarding a recommended menu;
    The dialogue control apparatus according to claim 2, wherein the second internal state is a negative emotion obtained with respect to the preference or evaluation of the customer regarding a recommended menu.
  4.  前記第1内部状態は、注文したメニューに関する前記顧客の評価、および接客サービスに関する前記顧客の評価、の少なくともいずれかに対して得られる正感情であり、
     前記第2内部状態は、注文したメニューに関する前記顧客の評価、および接客サービスに関する前記顧客の評価、の少なくともいずれかに対して得られる負感情である、請求項2に記載の対話制御装置。
    The first internal state is a positive feeling obtained for at least one of the customer's evaluation regarding the ordered menu and the customer's evaluation regarding the customer service,
    The dialogue control apparatus according to claim 2, wherein the second internal state is a negative emotion obtained for at least one of the customer's evaluation regarding the ordered menu and the customer's evaluation regarding the customer service.
  5.  前記シナリオ制御部は、顧客の選択情報と、前記第1内部状態を数値化した値の累積値と、前記第2内部状態を数値化した値の累積値と、に基づいて、進行中の対話シナリオを別の対話シナリオに切り替えるか否かを制御する、請求項2乃至4のいずれか一項に記載の対話制御装置。 The scenario control unit is configured to perform an ongoing dialogue based on customer selection information, a cumulative value obtained by quantifying the first internal state, and a cumulative value obtained by quantifying the second internal state. The dialogue control device according to any one of claims 2 to 4, which controls whether or not to switch a scenario to another dialogue scenario.
  6.  前記内部状態は、前記顧客と対話を行うロボットの感情および意図の少なくとも一方を含む、請求項1乃至5のいずれか一項に記載の対話制御装置。 The dialogue control apparatus according to any one of claims 1 to 5, wherein the internal state includes at least one of emotion and intention of a robot that performs dialogue with the customer.
  7.  前記シナリオ制御部によって前記内部状態を数値化した値の累積値に応じて、前記内部状態を表す行動を行うロボット動作制御部を備える、請求項1乃至6のいずれか一項に記載の対話制御装置。 The dialogue control according to any one of claims 1 to 6, further comprising a robot operation control unit that performs an action representing the internal state in accordance with a cumulative value obtained by quantifying the internal state by the scenario control unit. apparatus.
  8.  指示信号に応じた動作を行うロボットと、
     前記指示信号を生成して前記ロボットに送信する操作端末と、を備え、
     前記操作端末は、
     飲食店に来店した顧客と対話するための対話シナリオ情報を複数の対話シナリオ情報から取得する情報取得部と、
     前記取得した対話シナリオ情報に対応する対話シナリオの進行に伴って変化する内部状態を数値化した値を取得する内部状態値取得部と、
     前記取得した対話シナリオ情報に対応する対話シナリオの進行に伴って、前記内部状態値取得部で取得された値の累積値に基づいて対話シナリオを補正するシナリオ制御部と、を備えるロボット制御システム。
    A robot that operates in response to the instruction signal;
    An operation terminal that generates the instruction signal and transmits the instruction signal to the robot,
    The operation terminal is
    An information acquisition unit that acquires dialogue scenario information for interacting with a customer who has visited a restaurant from a plurality of dialogue scenario information;
    An internal state value acquisition unit that acquires a value obtained by quantifying the internal state that changes with the progress of the conversation scenario corresponding to the acquired conversation scenario information;
    A robot control system comprising: a scenario control unit that corrects a dialogue scenario based on a cumulative value of values acquired by the internal state value acquisition unit as the dialogue scenario corresponding to the acquired dialogue scenario information progresses.
  9.  前記ロボットは、前記シナリオ制御部によって前記内部状態を数値化した値の累積値に応じて、前記内部状態を表す行動を行うロボット動作制御部を有する、請求項8に記載のロボット制御システム。 The robot control system according to claim 8, wherein the robot has a robot operation control unit that performs an action representing the internal state in accordance with a cumulative value obtained by quantifying the internal state by the scenario control unit.
  10.  前記ロボット動作制御部は、前記シナリオ制御部によって前記内部状態を数値化した値の累積値に応じて、前記ロボットの声色、動作、目の色および目の光り方の少なくとも一つを変化させる、請求項9に記載のロボット制御システム。 The robot motion control unit changes at least one of the voice color, motion, eye color, and eye shine of the robot according to a cumulative value of values obtained by quantifying the internal state by the scenario control unit. The robot control system according to claim 9.
PCT/JP2018/011919 2017-03-24 2018-03-23 Conversation control system, and robot control system WO2018174290A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2017-059953 2017-03-24
JP2017059953A JP2018161703A (en) 2017-03-24 2017-03-24 Dialogue control device and robot control system

Publications (1)

Publication Number Publication Date
WO2018174290A1 true WO2018174290A1 (en) 2018-09-27

Family

ID=63585534

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2018/011919 WO2018174290A1 (en) 2017-03-24 2018-03-23 Conversation control system, and robot control system

Country Status (2)

Country Link
JP (1) JP2018161703A (en)
WO (1) WO2018174290A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10748644B2 (en) 2018-06-19 2020-08-18 Ellipsis Health, Inc. Systems and methods for mental health assessment
US11120895B2 (en) 2018-06-19 2021-09-14 Ellipsis Health, Inc. Systems and methods for mental health assessment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001300876A (en) * 2000-04-20 2001-10-30 Yamatake Corp Service robot and service system using it
JP2002283261A (en) * 2001-03-27 2002-10-03 Sony Corp Robot device and its control method and storage medium
JP2015090563A (en) * 2013-11-05 2015-05-11 Meet株式会社 Order management system, order management method, and program

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001300876A (en) * 2000-04-20 2001-10-30 Yamatake Corp Service robot and service system using it
JP2002283261A (en) * 2001-03-27 2002-10-03 Sony Corp Robot device and its control method and storage medium
JP2015090563A (en) * 2013-11-05 2015-05-11 Meet株式会社 Order management system, order management method, and program

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
KATO SHOHEI: "Affective computing and characterization for kensei communication robot", JOURNAL OF THE JAPANESE SOCIETY FOR ARTIFICIAL INTELLIGENCE, vol. 31, no. 5, 1 September 2016 (2016-09-01), pages 671 - 678 *
TAKEUCHI SHOUGO: "An emotion generation model based on the dialogist likability for sensitivity communication robot", JOURNAL OF THE ROBOTICS SOCIETY OF JAPAN, vol. 25, no. 7, 15 October 2007 (2007-10-15), pages 1125 - 1133 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10748644B2 (en) 2018-06-19 2020-08-18 Ellipsis Health, Inc. Systems and methods for mental health assessment
US11120895B2 (en) 2018-06-19 2021-09-14 Ellipsis Health, Inc. Systems and methods for mental health assessment
US11942194B2 (en) 2018-06-19 2024-03-26 Ellipsis Health, Inc. Systems and methods for mental health assessment

Also Published As

Publication number Publication date
JP2018161703A (en) 2018-10-18

Similar Documents

Publication Publication Date Title
US20220020360A1 (en) System and method for dialogue management
EP3381175B1 (en) Apparatus and method for operating personal agent
US20190206402A1 (en) System and Method for Artificial Intelligence Driven Automated Companion
JP7295136B2 (en) INTERACTIVE ANIME CHARACTER HEADSTEM AND METHOD
US11003860B2 (en) System and method for learning preferences in dialogue personalization
CN111086009B (en) Communication robot and computer-readable storage medium
KR102400398B1 (en) Animated Character Head Systems and Methods
WO2018174290A1 (en) Conversation control system, and robot control system
WO2018174289A1 (en) Conversation control system, and robot control system
CN115146048A (en) Multi-NPC dialogue text generation and display method, equipment and medium
KR100957858B1 (en) Avatar presenting method and computer readable medium processing the method
JP2018161712A (en) Dialogue control device and robot control system
JP2018161713A (en) Dialogue control device and robot control system
JP2018161702A (en) Robot control system and robot control device
JP2018161709A (en) Dialogue control system and dialogue control device
WO2024004609A1 (en) Information processing device, information processing method, and recording medium
JP2018161707A (en) Robot control system and robot control device
JP2018161710A (en) Dialogue control device and robot control system
JP2018161706A (en) Robot control system and robot control device
JP2018163504A (en) Interaction control device and interaction system
JP2018161708A (en) Robot control system and robot control device
WO2023089537A1 (en) Alternating perceived realities in a virtual world based on first person preferences and a relative coordinate system
JP2018161715A (en) Dialogue control device and dialogue system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18770393

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18770393

Country of ref document: EP

Kind code of ref document: A1