JP4815940B2 - Robot control system, robot apparatus, and robot control method - Google Patents

Description

  The present invention relates to a robot control system, a robot apparatus, and a robot control method.

  Patent Document 1 has sound detection means for detecting sound emitted from a sound source, directing the line of sight toward the calling person, recognizing the face of the person being faced, and actions according to some commands An autonomous behavior robot that can take the action is disclosed.

  In Patent Literature 2, a robot is connected to an information management base station that stores advertisement information via a wireless or wired connection, and a user status determination unit determines a user's status, and the determination result indicates a user status. A technique is disclosed in which suitable advertisement information is obtained from an information management base station by an advertisement information acquisition unit and presented to a user by advertisement information presenting means.

Patent Document 3 selects user information storage means for storing user information and operation information corresponding to user information stored in the user information storage means from operation information for causing the robot to execute an advertisement operation. The input voice is recognized by the selection means, the acquisition means for acquiring the operation information selected by the selection means from the information providing apparatus via the network, the voice recognition means for recognizing continuous input voice, and the voice recognition means. A robot control device is disclosed that includes an advertisement motion control means for controlling the robot to execute an advertisement motion based on the motion information acquired by the acquisition means when it is determined that the robot has been determined to be.
JP 2003-62777 A JP 2004-17200 A JP 2004-302328 A

  However, conventionally, since the robot has unilaterally explained to the user, it has not been left in the user's impression and could not be effectively explained.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a technique that allows a robot to make a description that will leave an impression.

According to the present invention,
A robot control system for controlling a robot that performs explanation,
A viewer status grasping unit for grasping the number of viewers in the vicinity of the robot;
An operation pattern storage unit that stores a plurality of operation patterns that the robot performs for the viewer when the robot performs explanation;
An operation pattern selection unit that selects any one of the operation patterns from the operation pattern storage unit based on the number of viewers grasped by the viewer status grasping unit;
An operation determining unit that determines an operation of the robot based on the operation pattern selected by the operation pattern selecting unit;
A robot control system is provided.

According to the present invention,
A robot control method for controlling a robot that performs explanation,
Determining the number of viewers in the vicinity of the robot;
Based on the number of viewers grasped in the step of grasping the number of viewers, an operation pattern storage unit that stores a plurality of operation patterns that the robot performs for the viewers when the robot performs explanation Selecting any one of the operation patterns;
Determining the motion of the robot based on the motion pattern selected in the step of selecting the motion pattern;
A robot control method is provided.

  Here, the robot can be described by voice output, image output, gesture or sign language operation, or a combination thereof. In addition, the explanation may be that the robot intentionally displays or transmits some information to the viewer, such as announcement, guidance, advertisement, presentation, education, communication with the viewer, and the like. The vicinity of the robot can be a range where the viewer can understand the explanation of the robot, such as a range where the voice output by the robot can be heard, a range where the image output and operation of the robot can be seen, etc. . In this way, when the robot makes an explanation according to the number of viewers in the vicinity of the robot, the robot can perform an appropriate operation for the viewer. As a result, the description of the robot can remain in the viewer's impression.

  Here, the robot can be a robot device or a virtual robot represented by a character. The robot can be an interactive robot having a voice recognition function.

  In the robot control system of the present invention, the movement pattern may be a line-of-sight movement pattern for controlling a line-of-sight of the robot to the viewer, and the viewer status grasping unit is a position of the viewer relative to the robot. The motion determination unit can determine the line-of-sight motion of the robot to the viewer based on the viewer position and the motion pattern selected by the motion pattern selection unit.

  Here, the line-of-sight movement pattern can include setting information such as an eye contact point that directs the robot's line of sight, a frequency at which the robot's line of sight is directed to each eye contact point, and a time that the robot's line of sight is directed to each eye contact point. . In this way, according to the number of viewers near or around the robot, the robot can perform a line-of-sight operation such as eye contact appropriate for the viewer when explaining the robot. . As a result, the description of the robot can remain in the viewer's impression. Note that the motion determination unit can be, for example, a gaze motion generation unit that generates a gaze motion of the robot.

  In the robot control system of the present invention, the motion determination unit may determine an eye contact point that directs the robot's line of sight based on the position of the viewer and the motion pattern selected by the motion pattern selection unit.

  In this way, the robot can direct the line of sight toward the viewer, and the robot can perform a line-of-sight operation such as eye contact appropriate for the viewer. Here, the motion determination unit can estimate the position of the viewer's eyes based on the position of the viewer, and determine the position of the viewer's eyes as an eye contact point that directs the line of sight of the robot.

In the robot control system of the present invention, the operation pattern may include a setting of a frequency of directing the line of sight of the robot to the eye contact point and a time for the robot to direct the line of sight to the eye contact point. Can determine a line-of-sight movement that directs the line of sight of the robot to the eye contact point according to the set frequency and the time based on the movement pattern selected by the movement pattern selection unit.
Thus, for example, when the number of viewers in the vicinity of the robot is small, the robot performs a careful eye contact operation that directs the eyes to each eye contact point for a long time, or when the number of viewers near the robot is large, However, the time for directing the line of sight to each eye contact point can be shortened so that the line of sight is directed evenly to a large number of viewers.

  In the robot control system of the present invention, the robot may be configured such that at least one of a neck, a torso, or an eye is movable, and the motion determination unit is configured so that the robot performs the line-of-sight motion. One or more movements of the neck, the torso, or the eye can be determined.

  In this way, the robot can perform a line-of-sight operation such as eye contact appropriate for the viewer.

  In the robot control system of the present invention, the motion pattern storage unit includes a first motion pattern in which the robot targets each of the viewers and a group in which the viewers are grouped. 2, and the operation pattern selection unit selects the first operation pattern when the number of viewers grasped by the viewer status grasping unit is a predetermined number or less. The second operation pattern can be selected when the number is larger than the predetermined number.

  Here, when the motion pattern is a line-of-sight motion pattern, the first motion pattern includes a setting in which the robot sets each viewer to an eye contact point that directs the viewer's line of sight, and the second motion pattern includes: It is possible to include a setting in which each group into which viewers are grouped is set as an eye contact point for directing the robot's line of sight.

  In the robot control system of the present invention, the viewer status grasping unit can detect a change in the number of viewers, and the operation pattern selecting unit is configured so that the viewer status grasping unit determines the number of viewers. When a change is detected, the motion pattern corresponding to a new number can be selected, and the motion determination unit can determine the motion of the robot based on the motion pattern selected by the motion pattern selection unit. it can.

  In this way, when the number of viewers in the vicinity of the robot changes while the robot is explaining, the operation pattern can be reselected according to the number of viewers. It can be left in the viewer's impression.

  The robot control system of the present invention may further include a setting storage unit that stores a setting as to whether or not a viewer who satisfies a predetermined condition is to be a special operation target, and the operation pattern storage unit includes the predetermined pattern A special motion pattern targeting a viewer who satisfies the condition can be stored in association with the predetermined condition, and the viewer status grasping unit satisfies the predetermined condition from the viewers A viewer can be detected, and the operation pattern selection unit is a case where the viewer condition grasping unit detects a viewer that satisfies the predetermined condition, and the setting storage unit satisfies the predetermined condition. When it is set that the viewer is set as the special operation target, the special operation pattern corresponding to the predetermined condition can be selected.

  Here, the predetermined condition is, for example, when a predetermined number or more of new viewers newly added during explanation by the robot are grasped, when a long-time viewer who is in the vicinity of the robot for a predetermined time or more is grasped, or For example, an active viewer who is determined to listen to the explanation of the robot eagerly can be recognized. In addition, when the motion pattern is a line-of-sight motion pattern, the special motion pattern includes a setting that sets a viewer who satisfies a predetermined condition as an eye contact point that directs the robot's line of sight to distinguish it from other viewers. can do. The robot can also utter a summary of explanation to a viewer who satisfies a predetermined condition.

  The robot control system of the present invention includes an echo acquisition unit that acquires an echo of the viewer with respect to the motion pattern executed by the robot in association with information indicating the motion pattern, and the echo acquired by the echo acquisition unit. And an echo totaling unit that counts the motion patterns for each motion pattern, wherein the motion pattern selection unit compares the echoes of the plurality of motion patterns, and the robot executes the motion pattern having a high echo. The operation pattern can be preferentially selected.

  Here, the reverberation can be, for example, customer collection efficiency or sales efficiency. In addition, the level of understanding of the viewer can be increased. In this way, according to the number of viewers in the vicinity of the robot, it is possible to appropriately select the operation of the robot, and it is possible to select the operation with high reverberation. Can be left in the viewer's impression.

  In the robot control system of the present invention, the echo acquisition unit can acquire the viewer's echo for the motion pattern executed by each of the plurality of robots in association with information indicating the motion pattern. These responses can be totaled. The results counted by the echo tabulation unit can be shared by a plurality of robots. As a result, it is possible to share an action pattern having a high response among a plurality of robots or to feed back to a plurality of robots.

  The robot control system according to the present invention includes a content storage unit that stores an utterance file and a motion file to be reproduced by the robot, and reads the designated utterance file and the motion file from the content storage unit, and causes the robot to reproduce them. A content playback unit, wherein the motion determination unit is configured to determine whether the motion determined based on the motion pattern is performed by playback of the motion file read by the content playback unit. When there is a contradiction, it is possible to cause the robot to preferentially perform the operation of the robot that is executed by reproducing the motion file.

  In this way, the robot motion can be determined based on the motion pattern so as not to contradict the preset robot motion.

  For example, when an explanation is given about a robot, the action pattern storage unit has the same content as the explanation of the robot, but the explanation rules that change the behavior of the robot at that time, how to talk about the robot, etc. Can be stored.

According to the present invention,
A robot control system according to any of the above,
An operation execution unit for executing the operation determined by the operation determination unit;
The robot apparatus characterized by including is provided.

  Here, the robot apparatus can be autonomously moving or interactive. Further, the robot apparatus can have a voice output function.

  In the robot control method of the present invention, the movement pattern is a line-of-sight movement pattern for controlling a line of sight of the robot to the viewer, and the robot control method includes a step of grasping a position of the viewer with respect to the robot. In the step of determining the movement of the robot, a line-of-sight movement of the robot toward the viewer may be performed based on the movement pattern selected in the step of selecting the position of the viewer and the movement pattern. Can be determined.

  In the robot control method of the present invention, the step of determining the motion of the robot includes an eye contact point that directs the robot's line of sight based on the motion pattern selected in the step of selecting the viewer position and the motion pattern. Can be included.

  In the robot control method of the present invention, the operation pattern may include a setting of a frequency of directing the robot's line of sight to the eye contact point and a time for the robot to direct the line of sight to the eye contact point. Determining a line-of-sight movement that directs the robot's line of sight to the eye contact point according to the set frequency and the time based on the movement pattern selected in the step of selecting the movement pattern. Can do.

  In the robot control method of the present invention, the robot may be configured such that at least one of a neck, a torso, or an eye is movable, and the robot performs the line-of-sight motion in the step of determining the motion of the robot. In addition, one or more movements of the neck, the body, or the eyes of the robot can be determined.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, etc. are also effective as an aspect of the present invention.

  According to the present invention, it is possible for the robot to make an explanation that will leave an impression.

  Next, the best mode for carrying out the invention will be described in detail with reference to the drawings. In the following drawings, configurations of parts not related to the essence of the present invention are omitted. Moreover, the same code | symbol is attached | subjected to the same component and description is abbreviate | omitted suitably.

  In the following embodiments of the present invention, a robot will explain a product in an indoor or outdoor commercial space, for example. The robot control system according to the present embodiment controls the operation of the robot so that the robot conveys the content to the user in an impressive manner.

(First embodiment)
In the present embodiment, the robot control system can output a voice and controls the robot to be explained by the voice output. The robot control system allows the robot to explain to the viewer along with the voice output when the robot explains by voice output according to the number and positions of people in the vicinity of the robot (hereinafter referred to as viewers). Select the motion pattern to perform and determine the robot motion. In the present embodiment, the movement pattern can be a line-of-sight movement pattern for controlling the line of sight to the viewer of the robot.

  For example, if there are a small number of viewers, such as four viewers, the robot will explain to all people while making eye contact. For example, if there are 10 or more viewers, the viewers are divided into left and right groups, and the robot makes an explanation while alternately making eye contact with each group. Further, for example, when there is a new viewer newly added, the robot makes intensive eye contact with the new viewer or explains the summary. As a result, the explanation of the robot can be made deeper to the viewer. In addition, when the robot makes eye contact, the viewer can concentrate and listen to the explanation of the robot with a certain tension.

FIG. 1 is a block diagram showing an example of the configuration of the robot control system in the present embodiment.
The robot control system 100 includes a viewer status acquisition unit 101, a viewer status grasping unit 102, an operation pattern selection unit 104, a scenario reading unit 106, a gaze motion generation unit 108 (motion determination unit), a content playback unit 110, and an operation pattern storage. Unit 112, scenario storage unit 114, viewer status storage unit 116, and content storage unit 118.

  Here, the robot controlled by the robot control system 100 can be a robot device capable of autonomous movement, or can be a virtual robot indicated by a character. In any case, the robot is configured so that at least one of the neck, the torso, and the eyes can move. When the robot is a robot apparatus, the robot control system 100 may be incorporated in the robot apparatus, provided outside the robot apparatus, and controlled from the outside. Further, only a part of the robot control system 100 may be incorporated in the robot apparatus. When the robot control system 100 is provided outside the robot apparatus, a control signal is transmitted to the robot apparatus via various networks such as wireless communication. In addition, when only a part of the robot control system 100 is incorporated in the robot apparatus and the rest is provided outside the robot apparatus, data exchange between them is also performed via various networks such as wireless communication. Can do. For example, in the present embodiment, only the viewer status acquisition unit 101 may be provided outside the robot apparatus, and other configurations may be incorporated in the robot apparatus. The configuration of the robot apparatus will be described later.

  The viewer status acquisition unit 101 acquires information on viewers near or around the robot. The viewer status acquisition unit 101 can be configured by, for example, a human body detection sensor such as a camera or a temperature sensor, an ultrasonic tag sensor, a wireless tag sensor, or the like.

  The viewer status grasping unit 102 analyzes the viewer information acquired by the viewer status obtaining unit 101 and grasps the number of viewers in the vicinity of the robot and the positions of those viewers with respect to the robot.

  Hereinafter, with reference to FIG. 2, processing in which the viewer status grasping unit 102 grasps the number and positions of viewers by image recognition will be described. Here, the viewer status acquisition unit 101 can be a camera capable of photographing the front of the robot.

  As shown in FIG. 2A, the viewer situation grasping unit 102 performs face detection by extracting an area identified as a face from the image obtained by the viewer situation obtaining unit 101. Subsequently, the viewer status grasping unit 102 determines the number of viewers in the vicinity of the robot, each viewer based on the size and shape of the detected face image, the position of the viewer status acquisition unit 101, and the position of the robot. The position (angle and distance) of the robot with respect to the robot is estimated. As a result, as shown in FIG. 2B, for example, a position estimation result of each viewer with respect to the robot is obtained. Here, four viewers A, B, C, and D are grasped.

  Here, the viewer status grasping unit 102 can also estimate the sex, age, etc. of the user based on the image acquired by the viewer status acquisition unit 101. In addition, correspondence data between the viewer's face image and its attribute information may be acquired in advance, and the viewer status grasping unit 102 may perform face recognition and perform personal authentication.

  Returning to FIG. 1, the viewer status acquisition unit 101 may be a tag reader. For example, a tag in which identification information and attribute information are stored in advance can be given to the viewer. In this case, the viewer status grasping unit 102 can estimate the number of viewers in the vicinity of the robot and the position of each viewer with respect to the robot by reading information from the tag with the viewer status acquisition unit 101. . Here, the tag can be an ultrasonic tag, for example, and the viewer status acquisition unit 101 can be a tag reader that reads information from the ultrasonic tag. In this case, by setting the position of the tag to be attached to each viewer at a predetermined location such as a chest, for example, the viewer status grasping unit 102 can determine from the tag position acquired by the viewer status acquiring unit 101 The position of the human eye can be estimated. Also, by storing attribute information such as the age of each viewer in the tag, for example, the distance from the chest to the eyes can be estimated. For example, when the viewer is a child (age 6 to 12 years old, etc.), it can be estimated that the eye position is about 30 cm above the tag position. Also in this case, the viewer's attribute information can be read and personal authentication can be performed.

  In addition, the viewer status acquisition unit 101 can be configured to use both a tag reader and a camera.

  The viewer status grasping unit 102 stores the viewer status such as the grasped number and position of viewers in the viewer status storage unit 116. The viewer status grasping unit 102 also detects a change in the viewer status based on the information acquired by the viewer status acquisition unit 101 and the information stored in the viewer status storage unit 116.

  The scenario storage unit 114 stores a scenario described about the behavior of the robot. The scenario can be, for example, a program in which a motion file or an utterance file to be reproduced by the robot and a reproduction means thereof are described. Here, the motion file indicates the motion of the robot. The utterance file indicates the utterance of the robot.

  In the present embodiment, the scenario also includes a setting for turning on / off the line-of-sight control. In the present embodiment, line-of-sight control refers to processing that causes the robot to perform eye contact with the viewer when explaining the voice output, etc., separately from the motion file operation. is there. When the motion file includes a line-of-sight motion, the line-of-sight control is set to be off. The line-of-sight motion is an operation of directing the robot's line of sight in a predetermined direction. For example, when a robot explains a product, there may be a motion file in which the robot points to the product and points to the product. In such a case, since the line-of-sight motion “facing the product” is included in the motion file, the line-of-sight control is set to be off. In this way, it is possible to prevent the line-of-sight control from contradicting the operation specified in the motion file.

  FIG. 3 is a diagram illustrating an example of a scenario stored in the scenario storage unit 114. Here, the scenario storage unit 114 stores a script language for designating a motion file and an utterance file and playing them in order.

  The scenario name for this scenario is “greeting”. In this scenario playback, the operation pattern “001” is selected when the number of viewers in the vicinity of the robot is 2 or less, and the operation pattern “002” is selected when the number of viewers is 3 or more and 20 or less. The operation pattern “003” is set to be selected when the number of viewers is 21 or more.

  Returning to FIG. 1, the scenario reading unit 106 selects a scenario from the scenario storage unit 114 and reads it in accordance with the setting or user instruction. The scenario reading unit 106 instructs the content reproduction unit 110 and the line-of-sight motion generation unit 108 according to the scenario read from the scenario storage unit 114. In the present embodiment, the scenario reading unit 106 notifies the gaze motion generation unit 108 when the gaze control is set to ON in the scenario read from the scenario storage unit 114.

  The content storage unit 118 stores motion files and utterance files. The content reproduction unit 110 receives an instruction from the scenario reading unit 106, reads out a corresponding motion file or utterance file from the content storage unit 118, issues a mechanical drive command for robot motion reproduction or a robot utterance command, and generates a scenario. Perform playback.

  The motion pattern selection unit 104 selects a robot motion pattern from the motion pattern storage unit 112 based on the number of viewers grasped by the viewer state grasping unit 102 and the scenario read by the scenario reading unit 106.

  The operation pattern storage unit 112 stores an operation pattern that enables effective eye contact when the robot performs an explanation. In the present embodiment, the movement pattern is a line-of-sight movement pattern for controlling the line of sight to the viewer of the robot. For example, a rule is stored such that when the number of viewers near the robot is small, eye contact is carefully performed, and when the number of viewers increases, eye contact is made to a large number. Here, for example, when the number of viewers is equal to or less than a predetermined number, the robot can perform eye contact using each viewer as an eye contact point. On the other hand, when the number of viewers is larger than the predetermined number, the robot can divide the viewers into several groups and perform eye contact using each group as one eye contact point.

  The motion pattern storage unit 112 stores an eye contact point at which the line of sight of the robot is directed, how often the line of sight is directed at the point, and how long the line of sight is directed at the point. In addition, the operation pattern storage unit 112 also stores, for example, whether or not to perform a preferential operation for a new viewer, a long-time viewer, an active viewer, or the like. In the present embodiment, the operation pattern storage unit 112 also functions as a setting storage unit that stores a setting as to whether or not a viewer who satisfies a predetermined condition is a special operation target. In another example, the setting storage unit may be provided separately from the operation pattern storage unit 112. Here, the new viewer is a viewer who has been grasped by the viewer status grasping unit 102 that he / she has newly come near the robot. The long-time viewer is a viewer who has been recognized by the viewer status grasping unit 102 to be in the vicinity of the robot for a certain amount of time. The active viewer is a viewer who is judged to be actively listening to the explanation of the robot, such as nodding, by the viewer status grasping unit 102.

  For example, when the viewer status acquisition unit 101 is a camera, whether or not the viewer is an active viewer is determined by confirming the nod based on the change in the position of the person's face by image recognition by the viewer status grasping unit 102. be able to. The viewer status grasping unit 102 recognizes, for example, a person who nods, a person who has a lot of eye contact times and time, or a viewer who makes a positive voice among the viewers near the robot. It is possible to determine whether or not the viewer is an active viewer.

FIG. 4 is a diagram illustrating an example of the internal configuration of the operation pattern storage unit 112.
The operation pattern storage unit 112 includes an operation pattern column, an eye contact point column, a movement order / frequency column, an eye contact time (seconds) column, a new viewer preferential column, a long viewer preferential column, and an active viewer preferential column. Including.

  An operation pattern name is stored in the operation pattern column. In the eye contact point column, the number of eye contact points with which the robot makes eye contact, the definition of the type of viewer that becomes the eye contact point, and the like are stored. In the movement order / frequency column, the order in which the robot turns its line of sight to each eye contact point and the frequency are stored. The eye contact time column stores the time (seconds) for the robot to turn its line of sight to each eye contact point.

  In addition, the new viewer preferential column, the long-time viewer preferential column, and the active viewer preferential column store whether or not preferential treatment is given to the new viewer, the long-time viewer, and the active viewer, respectively. The

  For example, in the operation pattern “001”, the eye contact point is “n”. Here, “n” is the number of viewers grasped by the viewer situation grasping unit 102. In the operation pattern “001”, each viewer is set as an eye contact point. In addition, each eye contact point is set so that eye contact is sequentially performed on all eye contact points in one place for 2 seconds. Further, the new viewer preferential treatment and the active viewer preferential treatment are set not to be performed. On the other hand, it is set so that long-time viewer preferential treatment is performed. Here, when the viewer who is in the vicinity of the robot for 2 minutes or longer is a long-time viewer and there is such a long-time viewer, the operation pattern is set to shift from “001” to “L001”.

  For example, in the operation pattern “002”, the eye contact point is “n”. In the operation pattern “002”, each viewer is set as an eye contact point. In addition, “random (1/3)” is stored in the movement order / frequency column. Here, each eye contact point is set so that eye contact is performed at a frequency of 1/3 at random in one place for 1.3 seconds.

  For example, in the operation pattern “003”, the eye contact point is “n / 5”. That is, when the number of viewers is n, these viewers are classified into five groups, and each group is set as one eye contact point. In addition, each eye contact point is set so that eye contact is sequentially performed on all eye contact points in one place for 1 second. Here, it is set so that none of the new viewer preferential treatment, the long-time viewer preferential treatment, and the active viewer preferential treatment are performed.

  For example, in the operation pattern “L001”, the eye contact point is “n (L) + n (F)”. Here, n (L) is the number of long-time viewers grasped by the viewer situation grasping unit 102. N (F) is the number obtained by subtracting the number n (L) of long-time viewers from the number of viewers grasped by the viewer situation grasping unit 102. In the operation pattern “L001”, each viewer is set as an eye contact point. The eye contact point is set by distinguishing the eye contact point L corresponding to the long-time viewer and the eye contact point F corresponding to the other viewers. Each eye contact point is set so that eye contact is made in order to all eye contact points. Here, one eye contact is set to be performed for 3 seconds for the eye contact point L and for 1 second for the eye contact point F. Further, when this operation pattern is selected, it is set to continue this operation pattern as long as there is a viewer in the vicinity of the robot for two minutes or more.

  In the operation pattern “N001”, the eye contact point is “n (N) + n (F)”. Here, n (N) is the number of new viewers grasped by the viewer situation grasping unit 102. N (F) is the number obtained by subtracting the number n (N) of new viewers from the number of viewers grasped by the viewer situation grasping unit 102. Further, when this operation pattern is selected, it is also set to continue this operation pattern for 2 minutes.

  In the operation pattern “A001”, the eye contact point is “n (A) + n (F)”. Here, n (A) is the number of active viewers grasped by the viewer situation grasping unit 102. N (F) is a number obtained by subtracting the number n (A) of active viewers from the number of viewers grasped by the viewer situation grasping unit 102. In addition, when this operation pattern is selected, it is set to continue this operation pattern as long as the active viewer is in the vicinity of the robot.

  Returning to FIG. 1, the line-of-sight motion generation unit 108 receives an instruction from the scenario reading unit 106 and generates a line-of-sight motion based on the operation pattern selected by the operation pattern selection unit 104. Specifically, the line-of-sight motion generation unit 108 generates a line-of-sight motion file for moving one or more of the robot's neck, torso, or eyes (eyeball) so that the robot performs an appropriate line-of-sight movement. The line-of-sight motion generation unit 108 can determine an eye contact point at which the line of sight of the robot is directed based on the viewer position stored in the viewer state storage unit 116 and the motion pattern selected by the motion pattern selection unit 104. . Further, the robot control system 100 can include a storage unit that stores the position of the robot's eyes, and the line-of-sight motion generation unit 108 considers the position of the robot's eyes so that the robot's eyes face the eye contact point. Can be. For example, when the eye contact points are on the left and right when viewed from the robot, and the operation pattern for performing eye contact in the order is set, the line-of-sight motion generation unit 108 causes the line-of-sight motion in which the robot turns the line of sight alternately. Is generated.

  In addition, the motion pattern storage unit 112 can include a setting of the frequency of directing the robot's line of sight to the eye contact point and the time for the robot to direct the line of sight to the eye contact point. Based on the operation pattern selected by the user 104, it is possible to generate a line-of-sight motion that directs the robot's line of sight toward the eye contact point according to a set frequency and time. Further, the line-of-sight motion generation unit 108 stores to which eye contact point the eye contact has been performed, and can be used when creating the next line-of-sight motion, such as when the operation pattern is changed during the process.

  In addition, the gaze motion generation unit 108, when impossible or unnatural movement occurs between the motion reproduced by the content reproduction unit 110 before the gaze motion generation and the gaze motion to be generated, After generating a line-of-sight motion that returns the face to the center position, an appropriate line-of-sight motion is generated, such as generating a line-of-sight motion based on the motion pattern storage unit 112.

  The content reproduction unit 110 issues a mechanical drive command for reproducing the visual axis motion of the robot based on the visual axis motion generated by the visual axis motion generation unit 108.

  Note that the viewer status acquisition unit 101 periodically acquires the status of surrounding viewers even while the content playback unit 110 issues a command. The viewer situation grasping unit 102 analyzes the data obtained by the viewer situation obtaining unit 101, and detects the number of viewers in the vicinity of the robot and the change in the position of those viewers with respect to the robot. When there is a change in the number or position of viewers in the vicinity of the robot due to movement of the robot or movement of the viewer, the motion pattern selection unit 104 operates from the motion pattern storage unit 112 according to a new situation. Reselect the pattern. Also, when the viewer status grasping unit 102 detects an active viewer, the viewer status grasping unit 102 refers to the operation pattern storage unit 112 to determine whether or not the active viewer preferential setting is set in the operation pattern selected at that time. to decide. When the active viewer preferential setting is set, the operation pattern selection unit 104 can reselect an active viewer preferential operation pattern.

FIG. 5 is a diagram illustrating an example of the internal configuration of the viewer status storage unit 116.
The viewer status storage unit 116 includes a time field, an action pattern field, a neighborhood field, and a change field.
For example, it is known that the viewer a and the viewer b are in the vicinity of the robot at 12:35. At this time, the operation pattern “001” is selected. Subsequently, at 12:36, it is recognized that there are a viewer a, a viewer b, and a viewer c near the robot. Here, N (c) is stored in the change column as the new viewer N from the previous record. At this time, the operation pattern remains “001” and does not change. Next, at 12:37, it is recognized that there are viewers b and viewers c near the robot. Here, since the viewer b is in the vicinity of the robot for 2 minutes or longer, it is recognized that the viewer b is the viewer L for a long time, and L (b) is stored in the change column. Here, referring to FIG. 4, when the motion pattern “001” is selected, the motion pattern is set to shift to “L001” when there is a long-time viewer who is in the vicinity of the robot for 2 minutes or more. ing. Therefore, the operation pattern shifts to “L001”.

  In the above embodiment, an example is described in which a scenario stored in the scenario storage unit 114 is described in association with a motion pattern number and a range of the number of viewers near the robot. . In another example, instead of describing the scenario, the motion pattern storage unit 112 stores the motion pattern performed on the viewers in the vicinity of the robot in association with the range of the number of viewers in the vicinity. You can also

  In addition, the motion pattern storage unit 112 is not limited to the number and position of viewers in the vicinity of the robot. The operation pattern can be stored. For example, even if the number of viewers near the robot is the same, different operation patterns can be stored in association with places, time zones, and the like.

FIG. 6 is a flowchart showing a processing procedure of the robot control system 100 in the present embodiment.
First, when the robot is initialized and the program is started, the scenario reading unit 106 reads a scenario from the scenario storage unit 114 (S100). Which scenario is read is determined based on settings or user instructions.

  Subsequently, the viewer status acquisition unit 101 acquires the viewer status (S102). The viewer status grasping unit 102 grasps the number and positions of viewers in the vicinity of the robot based on the viewer status acquired by the viewer status acquiring unit 101 (S104). The viewer status grasping unit 102 stores the grasped information in the viewer status storage unit 116. Further, the viewer status grasping unit 102 refers to the viewer status storage unit 116 and determines whether or not there is a change in the number or position of viewers (S106). When there is a change in the number or position of viewers (YES in S106), the viewer status grasping unit 102 stores the grasped number and location of viewers in the viewer status storage unit 116 in association with time (see FIG. S108). The operation pattern selection unit 104 selects an operation pattern from the operation pattern storage unit 112 based on the number of viewers grasped by the viewer state grasping unit 102 and the scenario read by the scenario reading unit 106 (S110).

  Subsequently, the scenario reading unit 106 interprets the scenario command (S112) and determines whether or not the line-of-sight control is on (S114). If there is no change in step S106 (NO in S106), the process directly proceeds to step S112. When the line-of-sight control is on in step S114 (YES in S114), the scenario reading unit 106 notifies the line-of-sight motion generation unit 108. The line-of-sight motion generation unit 108 refers to the viewer status storage unit 116 and acquires the number of viewers, positions, and the like. Subsequently, the line-of-sight motion generation unit 108 generates a line-of-sight motion based on the motion pattern selected by the motion pattern selection unit 104, the number of viewers, and the position (S116). The content reproduction unit 110 reads out and reproduces the corresponding content data from the content storage unit 118 based on an instruction from the scenario reading unit 106, and also reproduces the line-of-sight motion when the line-of-sight motion generation unit 108 generates the line-of-sight motion. (S118). If the line-of-sight control is not on in step S114 (NO in S114), the process proceeds directly to step S118.

  A control unit (not shown) determines whether or not the scenario read by the scenario reading unit 106 has ended (S120). If the scenario has not ended, the process returns to step S102 to acquire the viewer's situation. If it is determined in step S120 that the scenario has been completed (YES in S120), it is determined whether or not the next scenario is to be read (S122). If the next scenario is to be read (YES in S122), the process returns to step S100. Load a new scenario. On the other hand, if the next scenario is not read in step S122 (NO in S122), the process ends.

  As described above, according to the robot control system 100 according to the present embodiment, when the robot makes an explanation according to the number of viewers near or around the robot, the robot makes appropriate eye contact with the viewer. Can be done. As a result, the description of the robot can remain in the viewer's impression.

(Second Embodiment)
In the present embodiment, as in the first embodiment, the robot control system, along with the voice output, when the robot performs explanation by voice output according to the number and positions of viewers in the vicinity of the robot, The operation pattern that the robot performs for the viewer is selected. In addition to this, the robot control system preferentially selects an operation pattern with high reverberation in consideration of the viewer's reverberation with respect to the operation pattern executed by the robot. Also in the present embodiment, the motion pattern can be a line-of-sight motion pattern.

FIG. 7 is a block diagram showing the configuration of the robot control system 100 in the present embodiment.
In the present embodiment, the robot control system 100 further includes an echo acquisition unit 132, an echo totaling unit 134, and an echo storage unit 142 in addition to the configuration described in the first embodiment.

  The echo acquisition unit 132 acquires the viewer's echo for the motion pattern executed by the robot in association with information indicating the motion pattern. Here, the reverberation can be, for example, customer collection efficiency, sales efficiency, and the level of understanding of the viewer. The echo tabulation unit 134 tabulates the echoes acquired by the echo acquisition unit 132 for each operation pattern. The echo acquired by the echo acquisition unit 132 and the result totaled by the echo totaling unit 134 are stored in the echo storage unit 142.

  In the present embodiment, the motion pattern selection unit 104 compares the echoes of a plurality of motion patterns based on the results tabulated by the echo tabulation unit 134, and selects a motion pattern with a high echo as the motion pattern to be executed by the robot.

  The echo acquisition unit 132 acquires the echo of a viewer near the robot, for example, by voice recognition. The response can be, for example, a viewer's level of understanding. For example, by causing a viewer in the vicinity of the robot to ask a question and recognizing an answer to the question by voice, information such as how much the viewer understands the explanation of the robot is acquired.

  The echo acquisition unit 132 also acquires information such as how many viewers gathered when the robot explained the operation pattern based on the viewer situation grasped by the viewer situation grasping unit 102. Further, for example, the robot can make a promise such as “Please go to the shop of XX” for the viewers around. The robot can promise to the viewer by shaking hands, fingering, dialogue, etc. The echo acquisition unit 132 also acquires information such as how many viewers the robot has promised and how many of the viewers who have promised the robot have executed the promise.

  Further, when making a promise, the robot can identify the promised viewer by some method. For example, when the robot is arranged in a theme park, the viewer can be presented with an admission ticket, and the viewer can be identified by the reference number of the admission ticket. Further, for example, a card or the like with identification information can be given to the promised viewer and the identification information can be stored. Furthermore, a viewer's face image can also be acquired.

  The echo acquisition unit 132 can be configured to be able to communicate with a network, and can acquire information from other terminals via the network. For example, information of a person who purchased a product can be transmitted to the echo acquisition unit 132 from a terminal installed in a store. Thereby, the echo acquisition part 132 can grasp | ascertain it, for example, when the viewer who promised actually purchased goods. The echo acquisition unit 132 stores the acquired information in the echo storage unit 142.

FIG. 8 is a diagram illustrating an internal configuration of the echo memory unit 142.
The echo storage unit 142 includes an operation pattern column, a surrounding column, a promise column, a purchase column, an echo rate column, and the like. Here, in the operation pattern column, identification information of the operation pattern selected by the operation pattern selection unit 104 is stored. The perimeter column stores how many viewers are gathered around the robot when the operation pattern is being explained. The promise column stores how many viewers have promised the robot when the operation pattern is being explained. Also, the purchase column stores how many viewers who have purchased the product among the viewers who have heard the explanation of the operation pattern.

  For example, when the explanation of the operation pattern “001” is performed, 30 viewers gather around, 10 viewers who have made a promise, and 6 viewers who have made a purchase are stored. In addition, when the operation pattern “002” is described, 10 viewers are gathered in the surrounding area, and one promised viewer is stored and 0 purchased viewers are stored.

  Based on these pieces of information, the echo tabulation unit 134 tabulates echoes and calculates an echo rate. Here, the echo rate of the motion pattern “001” is “◯”, which is higher than the echo rate “Δ” of the motion pattern “002”. That is, it is estimated that the reverberation rate is higher when the description is made with the operation pattern “001” than when the description is made with the operation pattern “002”. The operation pattern selection unit 104 preferentially selects an operation pattern with a higher echo rate based on the counting result of the echo counting unit 134.

  Also in the robot control system 100 according to the present embodiment, the same effects as the robot control system 100 according to the first embodiment can be obtained. Further, in the present embodiment, since the operation pattern is selected according to the viewer's response, the robot can be explained effectively.

(Third embodiment)
In the present embodiment, similarly to the second embodiment, the robot control system 100 preferentially selects an operation pattern with high reverberation in consideration of the viewer's reverberation with respect to the operation pattern executed by the robot. In the present embodiment, the robot control system 100 selects an operation pattern with a high response based on the viewer's response to an operation pattern executed by a plurality of robots. Thereby, a more effective operation pattern can be selected with high accuracy. Also in the present embodiment, the motion pattern can be a line-of-sight motion pattern.

FIG. 9 is a block diagram showing a configuration of the robot control system 100 in the present embodiment.
The robot control system 100 has a configuration in which a plurality of robot control devices 100 a and 100 b and an operation pattern selection device 300 are connected via a network 400. The robot control device 100a and the robot control device 100b are each incorporated in the robot device and control each robot device.

  The network 400 can be realized by various wireless environments such as a wireless LAN and a mobile phone, various wired environments such as a LAN, a WAN, and a telephone line, or a shared memory.

  Since the robot control device 100a and the robot control device 100b have the same configuration, only the robot control device 100a will be described here. The robot control apparatus 100a has substantially the same configuration as the robot control system 100 described with reference to FIG. 7 in the second embodiment.

  The robot control apparatus 100a further includes a transmission / reception unit 140. Moreover, the robot control apparatus 100a can be configured not to include the echo totaling unit 134. Here, the configuration in which the robot control device 100a includes the viewer status acquisition unit 101 is illustrated, but the robot control device 100a does not include the viewer status acquisition unit 101 and is provided outside the robot control device 100a. Information from the viewer status acquisition unit 101 can also be acquired via the transmission / reception unit 140.

  The motion pattern selection device 300 acquires the viewer's response to the motion pattern executed by each robot from a plurality of robot control devices 100a and 100b in association with information indicating the motion pattern. The motion pattern selection device 300 counts these, selects the motion pattern with the highest response, and feeds it back to the robot control devices 100a and 100b. In each of the robot control devices 100a and 100b, the robot can perform an effective explanation by causing the robot to execute the motion pattern selected by the motion pattern selection device 300.

  The operation pattern selection device 300 stores a transmission / reception unit 302, an echo acquisition unit 304, an echo totaling unit 306, an operation pattern selection unit 308, and an echo storage unit 310. The transmission / reception unit 302 transmits / receives data to / from other terminals via the network 400. The echo acquisition unit 304 acquires, via the transmission / reception unit 302, the viewer's echo for the motion pattern executed by each robot in association with information indicating the motion pattern. The echo tabulation unit 306 tabulates the echoes acquired by the echo acquisition unit 304 for each operation pattern. The echo acquired by the echo acquisition unit 304 and the result totaled by the echo totaling unit 306 are stored in the echo storage unit 310. The motion pattern selection unit 308 compares the echoes of a plurality of motion patterns based on the results counted by the echo tabulation unit 306, and selects a motion pattern having a high echo as a motion pattern to be executed by each robot. The operation pattern selection unit 308 notifies the selected operation pattern to the robot control devices 100a and 100b via the transmission / reception unit 302. In the robot control device 100a and the like, the echo acquisition unit 132 acquires the motion pattern selected from the motion pattern selection device 300 via the transmission / reception unit 140 and notifies the motion pattern selection unit 104 of the motion pattern. The operation pattern selection unit 104 can preferentially select an operation pattern with a high response selected by the operation pattern selection device 300.

  When the robot advertises, the store terminal 402 of the store selling the advertised product can also be connected to the network 400. The echo acquisition unit 304 acquires information such as whether or not the viewer has purchased the product from the store terminal 402 or the like via the transmission / reception unit 302. As another example, the echo acquisition unit 132 of the robot control apparatus 100a acquires information such as whether or not a viewer has purchased a product from the store terminal 402 or the like via the transmission / reception unit 302, They can be transmitted to the operation pattern selection apparatus 300.

  The motion pattern selection device 300 can be a server outside the robot device, but can also be configured to be incorporated in any of a plurality of robot devices, for example. For example, the robot control device 100a also has the function of the motion pattern selection device 300, and the viewer's response to the motion pattern executed by each robot is acquired from another robot control device 100b or the like in association with information indicating the motion pattern. You can also do it.

  In this way, it is possible to cause other robots to execute an action pattern that is effective in any of the robots, and it is possible to execute an effective explanation by the robot.

  The echo acquisition unit 304 can also acquire the time, position, and the like described by each robot in association with the echo and the motion pattern. The echo acquisition unit 304 stores these pieces of information in the echo storage unit 310. When selecting an operation pattern to be executed by each robot, the operation pattern selection unit 308 selects an operation pattern in consideration of the time and position to be explained by each robot. As a result, the operation pattern selection unit 308 can give each robot a substantially real-time instruction in accordance with the situation.

  Furthermore, the operation pattern selection unit 308 can select an operation pattern to be executed by each robot according to the arrangement relationship of the plurality of robots when the plurality of robots are explained. For example, a robot can be explained for a large number of people, and a robot that gives a careful explanation can be arranged around it.

  Also in the robot control system 100 according to the present embodiment, the same effect as the robot control system 100 according to the first and second embodiments can be obtained. Furthermore, in this embodiment, the robot can be explained more effectively.

(Robot device)
An example of the robot described in the above embodiment will be described. The robot control system 100 and the robot control apparatus 100a (and 100b) described in the above embodiment can be configured to be incorporated in the robot apparatus and control the robot apparatus. In another example, the robot control system 100 and the robot control device 100a (and 100b) described in the above embodiment may be provided outside the robot device to control each robot device. it can.

FIG. 10 is a block diagram illustrating an example of an electrical configuration of the robot apparatus 200.
The robot apparatus 200 includes a body part 201 and a head part 202. The body unit 201 includes a controller 210 that controls the entire robot apparatus, a battery 211 that is a power source of the robot apparatus 200, a speaker 212, a microphone 213, an actuator 214A and an actuator 214B for moving two wheels, and a communication interface 224. Etc. are stored.

  The microphone 213 collects surrounding sounds and sends the obtained sound data to the controller 210.

  The controller 210 includes a CPU 210A and a memory 210B. The CPU 210A executes various processes by executing a control program stored in the memory 210B.

  The communication interface 224 performs communication with other terminals via a network such as the network 400. The transmission / reception unit 140 described in the third embodiment can be configured by the communication interface 224, for example.

  The head 202 contains a CCD camera 221A and a CCD camera 221B, and an actuator 222A and an actuator 222B for rotating the head 202. The CCD camera 221A, the CCD camera 221B, the actuator 222A, the actuator 222B, the battery 211, the speaker 212, the microphone 213, the actuator 214A, the actuator 214B, the communication interface 224, and the like function as an operation execution unit that executes the operation of the robot.

  The CCD camera 221 </ b> A and the CCD camera 221 </ b> B image the surrounding situation and send the obtained image data to the controller 210. The viewer status acquisition unit 101 described in the first to third embodiments can be configured by, for example, a CCD camera 221A or a CCD camera 221B.

  The actuator 222A and the actuator 222B rotate the head 202 of the robot apparatus 200 vertically and horizontally.

  The controller 210 appropriately reads information from the memory 210B based on audio data obtained via the microphone 213 and the communication interface 224 and image data obtained from the CCD camera 221A and the CCD camera 221B, and acquires the surrounding conditions. When the robot control system 100 and the robot control apparatus 100a (or 100b) are incorporated in the robot apparatus 200, each function of the robot control system 100 and the robot control apparatus 100a (or 100b) can be realized by the controller 210.

  The controller 210 controls the actuator 214A, the actuator 214B, the actuator 222A, the actuator 222B, and the like to cause the robot device 200 to execute the determined operation. In addition, the controller 210 generates a synthesized sound, supplies the synthesized sound to the speaker 212, and causes the robot apparatus 200 to output the determined utterance.

  Although not shown here, the robot apparatus 200 may include an arm unit and an actuator that drives the arm unit. The controller 210 controls the actuator that drives the arm unit to control the robot apparatus 200. It can also be used for finger cutting and shaking hands.

  The robot can be the robot apparatus 200 as shown in FIG. 10, but can also be a virtual robot shown by a character, for example. In this case, the virtual robot can be realized by displaying it on a monitor or the like. When the robot is a virtual robot, the direction of the line of sight of the virtual robot can be changed by drawing CG, for example.

  Each component of the robot control system 100, the robot control device 100a (or 100b), and the operation pattern selection device 300 described above realizes the components of this figure loaded in the CPU, memory, and memory of an arbitrary computer. And a storage unit such as a hard disk for storing the program, and a network connection interface. It will be understood by those skilled in the art that there are various modifications to the implementation method and apparatus.

  The embodiments of the present invention have been described above with reference to the drawings, but these are exemplifications of the present invention, and various configurations other than those described above can be adopted.

  In the above embodiment, the example in which the motion pattern is the line-of-sight motion pattern and the robot control system 100, the robot control device 100a, and the like have the line-of-sight motion generation unit 108 has been described. However, in other forms, the robot control system 100, the robot control apparatus 100a, etc., in addition to the line-of-sight motion generation unit 108, or instead of the line-of-sight motion generation unit 108, a motion generation unit that generates a motion other than the line-of-sight motion, It can be set as the structure which provided the control part etc. which control reproduction | regeneration of the speech file of a robot. For example, the motion pattern can be the speech rate of the robot, the loudness of the robot's voice, or the like. By controlling the reproduction of the utterance file of the robot by the control unit, it is possible to adjust the utterance speed of the robot and the loudness of the robot voice. For example, if the number of viewers in the vicinity of the robot is less than a predetermined number, the robot will explain slowly with a low voice, and if the number of viewers in the vicinity of the robot is larger than the predetermined number, the robot is large You can also give a quick explanation by voice.

The present invention can also include the following forms.
A robot control system for controlling a robot capable of outputting voice,
An operation pattern storage unit that stores a plurality of operation patterns performed by the robot on viewers in the vicinity of the robot;
An operation processing unit for causing the robot to execute any one of the operation patterns;
An echo acquisition unit that acquires an echo of the viewer with respect to an action pattern executed by the robot in association with information indicating the action pattern;
An echo totaling unit that tabulates the echoes acquired by the echo acquisition unit for each of the operation patterns;
An action pattern selection unit that compares the echoes of a plurality of the action patterns and selects the action pattern having a high echo as an action pattern to be executed by the robot;
Including
The motion processing unit is a robot control system that causes the robot to execute the motion pattern selected by the motion pattern selection unit.

FIG. 11 is a block diagram illustrating another example of the robot control system.
The robot control system 130 includes a scenario reading unit 106, a content reproduction unit 110, a scenario storage unit 114, a content storage unit 118, an echo acquisition unit 132, an echo totaling unit 134, an operation pattern selection unit 136, an echo storage unit 142, and an operation pattern storage. Unit 144 and motion generation unit 150.

  The robot control system 130 differs from the robot control system 100 described in the above embodiment in that it does not include the viewer status acquisition unit 101, the viewer status grasping unit 102, and the viewer status storage unit 116.

  The motion pattern selection unit 136 compares the echoes of a plurality of motion patterns on the basis of the results counted by the echo tabulation unit 134, regardless of the situation of the viewers in the vicinity of the robot, and the robot executes the motion patterns with high echoes. Select as operation pattern. The motion generation unit 150 generates a robot motion.

FIG. 12 is a diagram illustrating an example of an internal configuration of the operation pattern storage unit 144.
The operation pattern storage unit 144 includes an operation pattern list, a music view, a dance view, and a ramp view. Here, the dance table stores whether or not the robot is allowed to dance when the robot is explained. The music view stores whether or not to play music when the robot is explained. The lamp list stores whether or not the robot lamp is turned on when the robot is explained. For example, the operation pattern “001” is set to execute any of music, dance, and lamp.

  The motion generation unit 150 generates a robot motion according to the operation pattern selected by the operation pattern selection unit 136. Note that the motion generation unit 150 may include the function of the line-of-sight motion generation unit 108 described in the above embodiment. Further, the operation pattern storage unit 144 may include the same configuration as the operation pattern storage unit 112 described in the above embodiment.

  Also in such a form, as described with reference to FIG. 9 in the third embodiment, the echo totaling unit 134 counts the viewer's echoes for the motion patterns executed by the plurality of robots. Can be. Thereby, an effective operation pattern can be selected with high accuracy.

  INDUSTRIAL APPLICABILITY The present invention can be used for controlling a robot when making the robot explain presentation, guidance, advertisement, presentation, education, communication with viewers, and the like.

It is a block diagram which shows an example of a structure of the robot control system in embodiment of this invention. It is a figure which shows the process in which a viewer condition grasping | ascertainment part grasps | ascertains a viewer condition by image recognition. It is a figure which shows an example of the scenario memorize | stored in the scenario memory | storage part. It is a figure which shows an example of an internal structure of an operation pattern memory | storage part. It is a figure which shows an example of an internal structure of a viewer condition memory | storage part. It is a flowchart which shows the process sequence of the robot control system in embodiment of this invention. It is a block diagram which shows an example of a structure of the robot control system in embodiment of this invention. It is a figure which shows the internal structure of an echo memory | storage part. It is a block diagram which shows an example of a structure of the robot control system in embodiment of this invention. It is a block diagram which shows an example of the electrical structure of a robot. It is a block diagram which shows the other example of a robot control system. It is a figure which shows an example of an internal structure of an operation pattern memory | storage part.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Robot control system 100a Robot control apparatus 100b Robot control apparatus 101 Viewer condition acquisition part 102 Viewer condition grasping part 104 Action pattern selection part 106 Scenario reading part 108 Eye-gaze motion generation part 110 Content reproduction part 112 Action pattern storage part 114 Scenario storage Unit 116 viewer status storage unit 118 content storage unit 132 echo acquisition unit 134 echo totaling unit 136 motion pattern selection unit 140 transmission / reception unit 142 echo storage unit 144 motion pattern storage unit 150 motion generation unit 200 robot apparatus 201 trunk unit 202 head 203A , 203B Wheel 210 Controller 210A CPU
210B Memory 211 Battery 212 Speaker 213 Microphone 214A, 214B Actuator 221A, 221B CCD camera 222A, 222B Actuator 224 Communication interface 300 Operation pattern selection device 302 Transmission / reception unit 304 Echo acquisition unit 306 Echo totalization unit 308 Operation pattern selection unit 310 Echo storage unit 400 Network 402 Store terminal

Claims (20)

A robot control system for controlling a robot that performs explanation,
A viewer status grasping unit for grasping the number of viewers in the vicinity of the robot;
An operation pattern storage unit that stores a plurality of operation patterns that the robot performs for the viewer when the robot performs explanation;
An operation pattern selection unit that selects any one of the operation patterns from the operation pattern storage unit based on the number of viewers grasped by the viewer status grasping unit;
An operation determining unit that determines an operation of the robot based on the operation pattern selected by the operation pattern selecting unit;
Only including,
The viewer status grasping unit detects a change in the number of viewers,
When the viewer status grasping unit detects a change in the number of viewers, the operation pattern selection unit selects the operation pattern corresponding to a new number,
The operation determining unit is a robot control system that determines an operation of the robot based on the operation pattern selected by the operation pattern selecting unit .   A robot control system for controlling a robot that performs explanation,
  A viewer status grasping unit for grasping the number of viewers in the vicinity of the robot;
  An operation pattern storage unit that stores a plurality of operation patterns that the robot performs for the viewer when the robot performs explanation;
  An operation pattern selection unit that selects any one of the operation patterns from the operation pattern storage unit based on the number of viewers grasped by the viewer status grasping unit;
  An operation determining unit that determines an operation of the robot based on the operation pattern selected by the operation pattern selecting unit;
Including
  A setting storage unit for storing a setting as to whether or not a viewer who satisfies a predetermined condition is a special operation target;
  The motion pattern storage unit stores a special motion pattern targeted for a viewer who satisfies the predetermined condition, in association with the predetermined condition,
  The viewer status grasping unit detects a viewer that satisfies the predetermined condition from the viewers,
  The operation pattern selection unit is a case where the viewer condition grasping unit detects a viewer who satisfies the predetermined condition, and the viewer who satisfies the predetermined condition is set as the special operation target in the setting storage unit. If it is set to do, select the special operation pattern corresponding to the predetermined condition,
  The predetermined condition is a robot control system in which a predetermined number or more of new viewers are grasped or a long-time viewer in the vicinity of the robot is grasped for a predetermined time or more.   A robot control system for controlling a robot that performs explanation,
  A viewer status grasping unit for grasping the number of viewers in the vicinity of the robot;
  An operation pattern storage unit that stores a plurality of operation patterns that the robot performs for the viewer when the robot performs explanation;
  An operation pattern selection unit that selects any one of the operation patterns from the operation pattern storage unit based on the number of viewers grasped by the viewer status grasping unit;
  An operation determining unit that determines an operation of the robot based on the operation pattern selected by the operation pattern selecting unit;
Including
  The motion pattern storage unit is configured to group the viewers into a number less than the predetermined number when the robot has a first motion pattern in which each of the viewers is a motion target and the viewer has more than a predetermined number. Storing a second operation pattern for each divided group as an operation target;
  The operation pattern selection unit selects the first operation pattern when the number of the viewers grasped by the viewer status grasping unit is a predetermined number or less, and when the number is larger than the predetermined number, the first operation pattern is selected. A robot control system that selects two motion patterns. The robot control system according to any one of claims 1 to 3 ,
The movement pattern is a line-of-sight movement pattern for controlling the line of sight of the robot to the viewer,
The viewer status grasping unit grasps the position of the viewer with respect to the robot,
The motion determining unit is a robot control system that determines a line-of-sight motion of the robot to the viewer based on the position of the viewer and the motion pattern selected by the motion pattern selection unit. The robot control system according to claim 4 , wherein
The motion determination unit is a robot control system that determines an eye contact point for directing a line of sight of the robot based on the position of the viewer and the motion pattern selected by the motion pattern selection unit. The robot control system according to claim 5 , wherein
The operation pattern includes a setting of a frequency of directing the line of sight of the robot to the eye contact point and a time for the robot to direct the line of sight to the eye contact point,
The motion determining unit is a robot control system that determines a line-of-sight motion that directs the robot's line of sight toward the eye contact point according to the set frequency and the time based on the motion pattern selected by the motion pattern selection unit. The robot control system according to any one of claims 3 to 6 ,
The robot is configured such that at least one of a neck, a torso, or an eye is movable,
The movement determining unit is a robot control system that determines one or more movements of the neck, the trunk, or the eyes of the robot so that the robot performs the line-of-sight movement. The robot control system according to any one of claims 1 to 7 ,
An echo acquisition unit that acquires an echo of the viewer with respect to the motion pattern executed by the robot in association with information indicating the motion pattern;
An echo totaling unit that tabulates the echoes acquired by the echo acquisition unit for each of the operation patterns;
Further including
The motion pattern selection unit is a robot control system that compares echoes of a plurality of motion patterns and preferentially selects the motion pattern with high echoes as the motion pattern to be executed by the robot. The robot control system according to any one of claims 1 to 8 ,
A content storage unit for storing an utterance file and a motion file to be played by the robot;
A content playback unit that reads the specified utterance file and the motion file from the content storage unit and causes the robot to play back the content file,
Further including
The operation determining unit is executed by reproducing the motion file when the operation determined based on the operation pattern is inconsistent with the operation of the robot executed by reproducing the motion file read by the content reproducing unit. A robot control system for causing the robot to preferentially perform the operation of the robot. A robot control system according to any one of claims 1 to 9 ,
An operation execution unit for executing the operation determined by the operation determination unit;
The robot apparatus characterized by including. A robot control method for controlling a robot that performs explanation,
Determining the number of viewers in the vicinity of the robot;
Based on the number of viewers grasped in the step of grasping the number of viewers, an operation pattern storage unit that stores a plurality of operation patterns that the robot performs for the viewers when the robot performs explanation Selecting any one of the operation patterns;
Determining the motion of the robot based on the motion pattern selected in the step of selecting the motion pattern;
Only including,
In the step of grasping the number of viewers, a change in the number of viewers is detected,
In the step of selecting the operation pattern, when the viewer status grasping unit detects a change in the number of viewers, the operation pattern corresponding to a new number is selected,
A robot control method for determining an operation of the robot based on the selected operation pattern in the step of determining the operation . A robot control method for controlling a robot that performs explanation,
Determining the number of viewers in the vicinity of the robot;
Based on the number of viewers grasped in the step of grasping the number of viewers, an operation pattern storage unit that stores a plurality of operation patterns that the robot performs for the viewers when the robot performs explanation Selecting any one of the operation patterns;
Determining the motion of the robot based on the motion pattern selected in the step of selecting the motion pattern;
Including
Storing a setting as to whether or not a viewer who satisfies a predetermined condition is to be a special operation target,
In the step of storing the motion pattern, a special motion pattern intended for a viewer who satisfies the predetermined condition is stored in association with the predetermined condition,
In the step of grasping the number of viewers, a viewer who satisfies the predetermined condition is detected from the viewers,
In the step of selecting the operation pattern, when a viewer who satisfies the predetermined condition is detected, the viewer who satisfies the predetermined condition in the step of storing the setting may be set as the special operation target. If set, select the special operation pattern corresponding to the predetermined condition,
The robot control method in which the predetermined condition is a case where a predetermined number or more of new viewers are grasped or a long time viewer who is present in the vicinity of the robot for a predetermined time or more is grasped . A robot control method for controlling a robot that performs explanation,
Determining the number of viewers in the vicinity of the robot;
Based on the number of viewers grasped in the step of grasping the number of viewers, an operation pattern storage unit that stores a plurality of operation patterns that the robot performs for the viewers when the robot performs explanation Selecting any one of the operation patterns;
Determining the motion of the robot based on the motion pattern selected in the step of selecting the motion pattern;
Only including,
The motion pattern storage unit is configured to group the viewers into a number less than the predetermined number when the robot has a first motion pattern in which each of the viewers is a motion target and the viewer has more than a predetermined number. Storing a second operation pattern for each divided group as an operation target;
In the step of selecting the operation pattern, the first operation pattern is selected when the number of viewers grasped in the step of grasping the number of viewers is equal to or less than a predetermined number, and the number is larger than the predetermined number And a robot control method for selecting the second operation pattern. The robot control method according to any one of claims 11 to 13 ,
The movement pattern is a line-of-sight movement pattern for controlling the line of sight of the robot to the viewer,
Further comprising grasping a position of the viewer with respect to the robot;
A robot control method for determining a gaze motion of the robot to the viewer based on the motion pattern selected in the step of selecting the viewer position and the motion pattern in the step of determining the motion of the robot. The robot control method according to claim 14 , wherein
The step of determining the motion of the robot includes the step of determining an eye contact point that directs the robot's line of sight based on the motion pattern selected in the step of selecting the position of the viewer and the motion pattern. Method. The robot control method according to claim 15 , wherein
The operation pattern includes a setting of a frequency of directing the line of sight of the robot to the eye contact point and a time for the robot to direct the line of sight to the eye contact point,
In the step of determining the movement of the robot, a line-of-sight movement that directs the line of sight of the robot to the eye contact point according to the set frequency and the time based on the movement pattern selected in the step of selecting the movement pattern To determine the robot control method. The robot control method according to any one of claims 13 to 16 ,
The robot is configured such that at least one of a neck, a torso, or an eye is movable,
A robot control method for determining one or more movements of the neck, the trunk, or the eyes of the robot so that the robot performs the line-of-sight movement in the step of determining the movement of the robot. Computer to control the robot to explain
Viewer status grasping means for grasping the number of viewers in the vicinity of the robot;
An operation pattern storage means for storing a plurality of operation patterns that the robot performs for the viewer when the robot performs explanation;
An action pattern selection means for selecting any one of the action patterns from the action pattern storage means based on the number of viewers grasped by the viewer status grasping means;
Motion determining means for determining the motion of the robot based on the motion pattern selected by the motion pattern selecting means;
To function as,
The viewer status grasping means detects a change in the number of viewers,
When the viewer status grasping unit detects a change in the number of viewers, the motion pattern selection unit selects the motion pattern corresponding to a new number,
The operation determining means determines an operation of the robot based on the operation pattern selected by the operation pattern selection unit .   Computer to control the robot to explain
  Viewer status grasping means for grasping the number of viewers in the vicinity of the robot;
  An operation pattern storage means for storing a plurality of operation patterns that the robot performs for the viewer when the robot performs explanation;
  An action pattern selection means for selecting any one of the action patterns from the action pattern storage means based on the number of viewers grasped by the viewer status grasping means;
  Motion determining means for determining the motion of the robot based on the motion pattern selected by the motion pattern selecting means;
  A setting storage means for storing a setting as to whether or not a viewer who satisfies a predetermined condition is a special operation target;
Function as
  The operation pattern storage means stores a special operation pattern for a viewer who satisfies the predetermined condition in association with the predetermined condition,
  The viewer status grasping means detects a viewer that satisfies the predetermined condition from the viewers,
  The operation pattern selection means is a case where the viewer condition grasping unit detects a viewer who satisfies the predetermined condition, and the viewer who satisfies the predetermined condition is set as the special operation target in the setting storage unit. If it is set to do, select the special operation pattern corresponding to the predetermined condition,
  The predetermined condition is a case where a predetermined number or more of new viewers are recognized, or a case where a long-time viewer who is in the vicinity of the robot for a predetermined time or longer is recognized.   Computer to control the robot to explain
  Viewer status grasping means for grasping the number of viewers in the vicinity of the robot;
  An operation pattern storage means for storing a plurality of operation patterns that the robot performs for the viewer when the robot performs explanation;
  An action pattern selection means for selecting any one of the action patterns from the action pattern storage means based on the number of viewers grasped by the viewer status grasping means;
  Motion determining means for determining the motion of the robot based on the motion pattern selected by the motion pattern selecting means;
Function as
  The motion pattern storage means includes a first motion pattern in which the robot targets each of the viewers and a group of the viewers less than the predetermined number when the viewers are larger than the predetermined number. Storing a second operation pattern for each divided group as an operation target;
  The operation pattern selection unit selects the first operation pattern when the number of viewers grasped by the viewer state grasping unit is equal to or less than a predetermined number, and when the number is larger than the predetermined number, the first operation pattern is selected. A program for selecting two operation patterns.

Images