JP2019217122A - Robot, method for controlling robot and program - Google Patents

Abstract

To provide a robot that can express various emotions more than before to a predetermined object depending on favorability rating, and to provide a method for controlling a robot and a program.SOLUTION: A robot 100 includes: a user recognition unit 111 for recognizing a predetermined object; and a response control unit 116 for determining an approaching mode during approaching the predetermined object on the basis of favorability rating to the predetermined object recognized by the user recognition unit 111, and controlling the own device to express an emotion to the predetermined object depending on the determined approaching mode during approaching the predetermined object.SELECTED DRAWING: Figure 2

Description

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

  In order for devices such as robots to become widespread, it is important how to express familiarity with a predetermined object. At present, there is known a technique for changing a distance between a robot and a predetermined target in accordance with a favorable sensitivity to the predetermined target. (For example, see Patent Document 1).

JP 2000-326274 A

  However, the technique of Patent Literature 1 changes only the distance after moving to a predetermined target according to the favorable sensitivity, and is insufficient as an emotional expression according to the favorable sensitivity to the predetermined object. There is a problem.

  Therefore, the present invention has been made in view of the above-described circumstances, and provides a robot, a robot control method, and a program capable of expressing richer emotions than ever before in accordance with a favorable sensitivity to a predetermined object. The purpose is to provide.

In order to achieve the above object, one aspect of the robot according to the present invention is:
Recognition means for recognizing a predetermined object;
Determining means for determining an approach mode during approach to the predetermined object, based on the favorable sensitivity to the predetermined object recognized by the recognition means;
During the approach to the predetermined target, control means for controlling the own device to express emotions to the predetermined target according to the approach mode determined by the determination means,
Comprising,
It is characterized by the following.

In order to achieve the above object, one mode of a robot control method according to the present invention is as follows.
A recognition step of recognizing a predetermined object;
A determining step of determining an approach mode during approaching the predetermined object based on the favorable sensibility for the predetermined object recognized in the recognition step;
During the approach to the predetermined object, a control step of controlling a robot to express an emotion to the predetermined object according to the approach mode determined in the determination step,
including,
It is characterized by the following.

Further, in order to achieve the above object, one embodiment of a program according to the present invention is:
On the robot computer,
A recognition process for recognizing a predetermined object;
A determination process of determining an approach mode during approach to the predetermined object based on the favorable sensitivity to the predetermined object recognized by the recognition process;
During the approach to the predetermined object, a control process of controlling the robot so as to express an emotion on the predetermined object in accordance with the approach mode determined by the determination process,
Run,
It is characterized by the following.

  According to the present invention, it is possible to provide a robot, a robot control method, and a program capable of expressing richer emotions than ever before in accordance with a favorable sensibility for a predetermined object.

FIG. 2 is a front view of the robot according to the embodiment of the present invention. FIG. 2 is a block diagram illustrating a configuration of a robot. It is a figure showing an example of a user action table. FIG. 7 is a diagram illustrating an example of a positivity setting table (a positivity acquisition table). It is a figure showing an example of an approach mode decision table. It is a figure for explaining a linear movement course (approach course). It is a figure for explaining a curved movement course. It is a figure for explaining a meandering movement course. FIG. 5 is a diagram for explaining a state in which the robot faces the user in front of the user. FIG. 4 is a diagram for explaining a state in which the robot faces the user diagonally. It is a flowchart which shows the flow of a response control process (approach control process). It is a figure for explaining a movement course where a detour course (passing course) was formed.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the following embodiment, a robot including a dialogue device according to an embodiment of the present invention will be described as an example.

  The robot 100 is a robot that communicates with a predetermined target. Here, the predetermined target is a user (user) who uses the robot 100, and is typically an owner of the robot 100, a family or a friend of the owner, or the like. In addition, the predetermined target includes an animal and another robot in addition to a human.

  As shown in FIG. 1, the robot 100 is a robot having a three-dimensional shape that imitates a human in appearance. The exterior of the robot 100 is formed mainly of synthetic resin. The robot 100 has a torso part 101, a head part 102 connected to the upper part of the torso part 101, a hand part 103 connected to each of the left and right sides of the torso part 101, and two feet connected to the lower part of the torso part 101. 104. Each part of the head 102, the hand part 103, and the foot part 104 is attached to the body part 101 via a joint member (not shown), and is configured to perform a preset operation according to the control of the control unit 110 described later. I have. The head 102 has a pair of left and right eyes 105 and a mouth 106. Note that the upper, lower, left, and right sides of FIG. 1 are the upper, lower, right, and left sides of the robot 100, respectively.

  Next, the configuration of the robot 100 will be described with reference to FIG. The robot 100 includes a control unit 110, a storage unit 120, an imaging unit 130, a sensor unit 140, a moving unit 150, an operation unit 160, a voice input unit 170, and a voice output unit 180, as shown in FIG. These units are electrically connected to each other via a bus line BL.

  The control unit 110 is configured by a computer having a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like, and controls the operation of the robot 100. The control unit 110 controls the operation of each unit of the robot 100 by reading the control program stored in the ROM by the CPU and executing the control program on the RAM.

  By executing the control program, the control unit 110 executes the user recognition unit 111, the face recognition unit 112, the voice recognition unit 113, the user action recognition unit 114, the positivity setting unit (favorability acquisition unit) 115, and the response control unit ( It functions as an approach control unit) 116.

  The user recognition unit 111 identifies and recognizes a user existing around the robot 100 (for example, within a radius of 3 m from the robot 100). The user recognizing unit 111 is, for example, based on one or both of a face recognition result by a face recognizing unit 112 described later and a speaker recognition result by the voice recognizing unit 113, who is present around the robot 100. Recognize. In addition, the user recognition unit 111, for example, based on the position and size of the face image detected by the face recognition unit 112 from the captured image of the imaging unit 130, the sound source direction recognized by the voice recognition unit 113, and the like, The user recognizes the position of the user and the distance to the user. The user recognition unit 111 functions as a recognition unit of the present invention.

  The face recognition unit 112 recognizes a user's face existing around the robot 100. The face recognition unit 112 detects a user existing around the robot 100 using a known face detection technology, an object detection technology, or the like, based on a captured image acquired from the imaging unit 130 described later, for example. Subsequently, the face recognition unit 112 detects, for example, a face image corresponding to the face of the user detected from the image captured by the imaging unit 130, and extracts face feature information indicating the feature amount of the face. Then, the face recognizing unit 112 collates the face feature information extracted from the face image with the face feature information included in each piece of user information stored in a user information DB (Data Base) 121 of the storage unit 120 to be described later. A registered user whose degree satisfies a predetermined criterion is specified. If the registered user whose similarity satisfies the predetermined criterion cannot be specified, the face recognizing unit 112 recognizes that the detected user is an unregistered user. The facial feature amount may be any information that can identify the user's face.For example, the facial features such as eyes, nose, mouth, etc. are represented by numerical values such as the shape, size, and arrangement of each part included in the face. It is a thing. Thus, the face recognizing unit 112 functions as the face recognizing unit of the present invention.

  The voice recognition unit 113 recognizes the content of the utterance of the user included in the voice data acquired from the voice input unit 170 described later. The speech recognition unit 113 performs a speech recognition process on the speech data using the speech recognition data stored in the speech recognition DB 122 of the storage unit 120, and generates text data indicating the utterance content of the user as an utterance recognition result. The voice recognition unit 113 functions as a voice recognition unit of the present invention.

  Further, the voice recognition unit 113 recognizes a speaker who has generated a voice included in the voice data acquired from the voice input unit 170. The speech recognition unit 113, for example, deletes background noise and the like from the speech data acquired from the speech input unit 170, and extracts speech feature information indicating a feature amount of the user's speech. Then, the voice recognition unit 113 compares the voice feature information extracted from the voice data with the voice feature information included in each user information stored in the user information DB 121 of the storage unit 120, and determines the similarity as a speaker recognition result. Specifies a registered user satisfying a predetermined criterion. The voice feature information may be any information as long as the voice of the user can be identified. For example, the voice feature information indicates acoustic features such as the pitch, strength, speed, and voice quality of the user by numerical values. As described above, when a plurality of users existing around the robot 100 are recognized by the user recognition unit 111, the voice recognition unit 113 can specify which user has spoken.

  Further, the voice recognition unit 113 recognizes the sound source direction based on the voice data acquired from the voice input unit 170. The voice recognition unit 113 recognizes a sound source direction based on, for example, a sound collection time difference and a sound pressure level difference of each microphone of the microphone unit included in the voice input unit 170.

  The user action recognition unit 114 recognizes a user action with respect to the robot 100 (hereinafter, referred to as “user action”). The user action recognition unit 114 recognizes the user action according to, for example, the utterance content of the user recognized by the voice recognition unit 113, the contact strength detected by the touch sensor of the sensor unit 140, and the like. For example, when the contact strength detected by the touch sensor is lower than a preset reference value, the user behavior recognition unit 114 recognizes the user action as “stroke” and sets the contact strength to the preset reference value. If the value is higher than the value, the user action is recognized as “hitting”. Note that the user action determination unit may recognize the user action by detecting a user's facial expression, behavior, or the like from the image captured by the image capturing unit 130. The user action recognition unit 114 adds the recognized user action to the user information of the corresponding user and stores the information as history information. The user action recognition unit 114 functions as the action recognition unit of the present invention.

  The positivity setting section (positivity acquisition section) 115 sets (acquires) the positivity of each user. The favorable impression setting unit 115 determines the favorable impression for the user in accordance with, for example, the user's behavior with respect to the robot 100. The positivity setting unit 115 refers to, for example, the user action table and the positivity setting table (positivity acquisition table) stored in the storage unit 120 when determining the positivity.

  In the user action table, as shown in FIG. 3, "user action" and "given point" are associated with each other. As the “user action”, “talk”, “stroke”, and the like are listed. The “given point” is a numerical value preset for each user action. By setting different giving points for each user action, weighting is given according to the attribute of the user action (the user's preference for the robot 100). For example, positive values are set for favorable actions such as "talk" and "stroke", and negative values are set for unfavorable actions such as "do not respond" and "hit". By performing such weighting, each user behavior can be distinguished into a favorable behavior and a non-favorable behavior, and if the behavior is favorable, the sensitivity is increased, and the unfavorable behavior is increased. If so, the favorable sensitivity can be reduced.

  In the positivity setting table, as shown in FIG. 4, "favorability" and "favorable points" are associated with each other. “Positive” is classified into five levels of “A”, “B”, “C”, “D”, and “E” according to the degree of favorable. The “favorable point” is the total value of the points given to each user action performed by the same user during the predetermined period. Higher liking points are associated in the order of liking “A” to “E”. That is, one or a plurality of thresholds relating to positivity are provided, and the positivity is classified into a plurality of categories according to the provided thresholds. The “favorable point” is, for example, the total value of the awarded points in the latest predetermined period, a weighted average value obtained by increasing the weight as the favorable point in the latest predetermined period is obtained, and a certain long term including the latest predetermined period. A moving average value or the like obtained from the favorable points can be used.

  For example, the positivity setting unit 115 sets the positivity (positivity level) of each user every time a predetermined period elapses, and sets the current state included in the corresponding user information stored in the user information DB 121 of the storage unit 120. Update and memorize the liking of. Further, the positivity setting unit 115 can appropriately acquire a preset positivity with reference to the corresponding user information stored in the user information DB 121. As described above, the positivity setting unit 115 functions as a recording unit and an acquisition unit of the present invention.

  The response control unit 116 determines a response mode according to a favorable impression of the target user and, based on the determined response mode, expresses an emotion so as to express an emotion among the moving unit 150, the operation unit 160, and the voice output unit 180. At least one is controlled to execute various actions.

  When the user behavior recognition unit 114 recognizes the user behavior, the response control unit 116 responds to the behavior of the robot 100 that responds to the user behavior (hereinafter, referred to as a “response behavior (response operation)”). Is determined according to the sensitivity. Answering control section 116, for example, a user behavior recognition unit 114 is "Good morning", if you recognize the greeting such as "Hello", for a high favorability target user, "Good morning", greeting such as "Hello" Following this, a response with utterances such as "Today is good weather" or "What do you want to do today?" . On the other hand, for a low positivity target user, "good morning", to determine formal only greeting such as "Hello", a ritual response as answering manner. Then, the response control unit 116 controls the voice output unit 180 to output a voice indicating the determined response mode.

  In addition, the response control unit 116 determines whether the target user satisfies a preset approach condition when the user activity recognition unit 114 recognizes the user activity, and determines that the target user satisfies the approach condition, The approach mode as the response mode is determined according to the favorable impression. The approach condition includes that the target user has called for the robot 100 to approach the robot 100 in the vicinity of the user (for example, “come”, “come here”, etc.). In this case, the response control unit 116 may determine whether or not the approach condition is satisfied, based on the speech recognition result by the voice recognition unit 113. In the following description, a target user satisfying the approach condition is referred to as an “approach target user”. When it is determined that the approach condition is satisfied, the response control unit 116 refers to, for example, the approach mode determination table illustrated in FIG. 5 and determines, as the approach mode, a moving route (approach route) and a moving speed approaching the approach target user. At least one of (approach speed), an opposing distance (approach distance) with the approach target user at the end of the approach, and an opposing posture (approaching posture) facing the user during or at the end of the approach is determined. Thus, the response control unit (approach control unit) 116 functions as the control unit and the determination unit of the present invention.

  For example, as shown in FIG. 5, the approach state determination table corresponds to “moving path”, “moving speed”, “opposed distance”, and “opposed attitude” as response states (approaching states) for each level of liking. It is attached.

  As the “moving route (approaching route)”, a route pattern that is more direct (linear) toward the approach target user is set as the preference is higher, and a path pattern that makes a detour (meandering) is set as the preference is lower. Have been. For example, a favorable path “A” is a straight path “linear”, a favorable path “C” is a path with a small curvature “curved (large)”, and a favorable path “E” has a large meandering width. The route “meandering (large)” is set.

  More specifically, in the case of the moving path “straight”, as shown in FIG. 6, the robot 100 approaches the approach target user UR along the linear moving path MR. In addition, in the case of the moving path “curved (large)” or “curved (small)”, as illustrated in FIG. 7, the robot 100 approaches the approach target user UR along the curved moving path MR. Note that “curved (large)” indicates that the curvature is smaller than “curved (small)”. In the case of the moving route “meandering (large)” or “meandering (small)”, as shown in FIG. 8, the robot 100 approaches the approach target user UR along the meandering moving route MR. Note that “meandering (large)” indicates that the meandering width is larger than “meandering (small)”. As described above, the “moving route” is set such that the moving distance becomes longer as the likability decreases. The response control unit 116 determines a specific moving route based on the position of the target user recognized by the user recognizing unit 111, the distance to the target user, and the route pattern indicated by the “moving route”.

  The “moving speed (approaching speed)” is set such that the higher the liking is, the higher the speed is, and the lower the liking is, the lower the speed is. For example, the favorable sensitivity “A” is “high” (for example, 0.5 m / sec) faster than a preset standard speed (for example, 0.3 m / sec), and the favorable sensitivity “B” is a standard "Slightly higher" (for example, 0.4 m / sec), which is slightly higher than the speed, "Medium" which is the standard speed for the favorable sensitivity "C", and slightly lower than the standard speed for the favorable sensitivity "D" A certain “slightly low” (for example, 0.2 m / sec) and the favorable sensitivity “E” are set to “low” (for example, 0.1 m / sec) lower than the standard speed.

  The “opposite distance (approach distance)” is set so that the distance between the robot 100 and the approach target user becomes shorter at the end of the movement as the liking is higher. The distance from the user is set to be long. For example, the favorable sensitivity “A” is “short” which is a distance (for example, 0.3 m) shorter than a preset standard distance (for example, 0.5 m), and the favorable sensitivity “B” is more than the standard distance. "Slightly short" which is a slightly shorter distance (for example, 0.4 m), "Medium" which is a standard distance for the favorable sensitivity "C", and a slightly longer distance (for example, 0 for the favorable sensitivity "D") than the standard distance .6m), and "Length", which is a distance longer than the standard distance (for example, 0.7m), is set for the favorable sensitivity "E". As described above, the “facing distance” is set so as to end the approaching movement at the separated position and face the approaching user as the favorable impression decreases.

  The “opposing posture (approaching posture)” is a posture in which the robot 100 faces the user during approaching or at the end of approaching the approaching user, and as shown in FIGS. 9 and 10, the front direction of the approaching user UR. It is represented by an opposing angle α which is an angle formed between the UD and the front direction RD of the robot 100. Here, the front direction UD of the approach target user is a direction perpendicularly penetrating the surface along the left-right direction of the approach target user from the rear to the front. Further, the front direction RD of the robot 100 is a direction that vertically penetrates a surface along the left-right direction of the robot 100 from the rear to the front. The setting is such that the higher the liking is, the more directly facing the approaching user, and the lower the liking is, the more inclined the user is facing the approaching user. For example, “0 °” for the favorable impression “A”, “10 °” for the favorable impression “B”, “20 °” for the favorable impression “C”, “30 °” for the favorable impression “D”, “40 °” is set for the sensitivity “E”. When the opposing angle α is 0 °, the robot 100 faces the approaching user UR directly in front, as shown in FIG. 9, and when the opposing angle α is other than 0 °, as shown in FIG. 100 faces obliquely to the approaching user UR.

  Although the description has been made assuming that the front direction of the body 101 and the head 102 of the robot 100 match, the body 102 of the robot 100 (face The angle formed between the front direction of the part ()) and the front direction of the approach target user may be defined as the opposing angle α. That is, as the favorable impression decreases, the head 102 may be rotated left or right to turn the head 102 (face) of the robot 100 away from the approaching user.

  As described above, the approach state determination table illustrated in FIG. 5 indicates that the robot 100 performs a favorable operation such as approaching a user with high liking quickly and quickly, while the robot 100 performs a favorable operation with respect to a user with low liking. The approach mode is set so as to perform an unfavorable operation such as moving over time and stopping the approach at a remote position. In the approach state determination table, a priority operation or a non-priority operation may be set according to the degree (level) of liking.

  Furthermore, the response control unit 116 determines whether or not to execute a voluntary action of the robot 100 (hereinafter, referred to as “self-issue action”) even when the user action recognition unit 114 does not recognize the user action. However, if it is determined to be executed, the response mode for executing the self-issue operation is determined according to the favorable impression. When determining whether or not to execute the self-issuing action, for example, a time during which the user action recognition unit 114 does not recognize the user action, that is, a time during which the target user does not act on the robot 100 for a predetermined time (for example, 30 seconds). Minutes) may be used as a determination condition. The self-issue action includes an utterance (for example, "What are you doing?", "Let's talk", etc.) or an action (for example, a hand that draws the target user's attention to encourage the target user to communicate with the robot. Shaking operation, etc.), and the utterance content and the degree of the operation are appropriately adjusted according to the favorable sensitivity. It should be noted that regardless of whether or not there is a user action and whether or not the approach condition is satisfied, the target user with high liking may be spontaneously approached.

  In addition, the user recognition unit 111, the face recognition unit 112, the voice recognition unit 113, the user action recognition unit 114, the sensibility setting unit 115, and the response control unit 116 may realize each function by a single computer, Each function may be realized by a separate computer.

  The storage unit 120 includes a rewritable nonvolatile semiconductor memory, a hard disk drive, and the like, and stores various data necessary for the control unit 110 to control each unit of the robot 100. The storage unit 120 stores, for example, the above-described user action table, positivity setting table, approach state determination table, and the like in a predetermined storage area. The storage unit 120 functions as a recording unit of the present invention.

  Further, the storage unit 120 has a plurality of databases each storing various data. The storage unit 120 has, for example, a user information DB 121, a voice recognition DB 122, and a dialog information DB 123.

  The user information DB 121 accumulates and stores various information on registered users as user information. The user information includes, for example, user identification information assigned in advance to identify a user, face feature information indicating a feature amount of a user's face, voice feature information indicating a feature amount of a user's voice, and a user for each predetermined period. It includes history information such as actions and liking points, and current liking.

  The voice recognition DB 122 stores voice recognition data used for voice recognition processing. The speech recognition data includes, for example, an acoustic model representing each feature (frequency characteristic) of a phoneme, which is the minimum unit of a sound that distinguishes the meaning from another word, a word dictionary for associating phoneme features with words, a sequence of words And a language model representing its connection probability.

  The conversation information DB 123 stores the conversation information necessary for the response control unit 116 to generate voice data related to the response and utterance of the robot 100 in the conversation with the user. The conversation information is, for example, response information used to generate a response of the robot 100 to the user's utterance, utterance information used to generate an utterance of the robot 100 when spontaneously speaking to the user, a response of the robot 100 And speech synthesis information used to generate speech data indicating a synthesized speech waveform of the utterance. In the response information, for example, text data indicating the utterance content of the user and text data indicating the response content of the robot 100 are associated with each other. The utterance information includes, for example, text data indicating the utterance content of the robot 100. The speech synthesis information includes, for example, speech data indicating a speech waveform of each phoneme used to convert text data into speech.

  The imaging unit 130 is configured by a camera including an image sensor such as a lens, a charge coupled device (CCD) image sensor, and a complementary metal oxide semiconductor (CMOS) image sensor, and captures an image around the robot 100. The imaging unit 130 is installed, for example, at the front upper part of the head 102, captures an image of the front of the head 102, and generates and outputs digital image data. The camera is attached to a motor-driven gantry (gimbal or the like) operable to change the direction in which the lens faces, and is configured to be able to track a user's face or the like. The imaging unit 130 functions as an imaging unit of the present invention.

  The sensor unit 140 acquires information (physical quantity) indicating the external state and the internal state of the robot 100, converts the information into a predetermined signal as appropriate, and supplies the signal to the control unit 110. The sensor unit 140 includes various sensors such as a touch sensor that detects pressure received by a user's contact, an acceleration sensor that detects acceleration of the robot 100, and a temperature and humidity sensor that detects external temperature and humidity. The touch sensor is disposed, for example, above the head 102 in order to recognize a physical action of the user such as “stroke” or “hit” according to the magnitude of the contact pressure. The contact strength may be detected from accelerations in three orthogonal directions using an acceleration sensor.

  The moving unit 150 is a part that enables the robot 100 to move. The moving unit 150 includes, for example, wheels provided on the bottoms of the left and right feet 104 of the robot 100, a motor that drives the left and right wheels to rotate, and a drive circuit that drives and controls the motors. The drive circuit supplies a pulse signal for driving to the motor in accordance with the control signal received from the control unit 110. The motor rotationally drives the left and right wheels according to the driving pulse signal, and moves the robot 100. Thus, the moving unit 150 functions as the moving unit of the present invention. Note that the number of motors is arbitrary as long as the left and right wheels are configured to rotate independently of each other and the robot 100 can travel forward, backward, turn, accelerate, and decelerate. For example, the left and right wheels may be driven by one motor by providing a coupling mechanism or a steering mechanism. Further, the number of drive circuits can be appropriately changed according to the number of motors.

  The operation section 160 is a section that performs a predetermined operation under the control of the control section 110. The operation unit 160 includes driven members such as the head 102, the hand 103, and the foot 104, a driving member such as an actuator that drives the driven member, and a driving circuit that drives and controls the driving member. . Under the control of the control unit 110 (the response control unit 116), the operation unit 160 drives the driven member in accordance with the drive signal supplied from the drive circuit, for example, an operation of moving a limb or moving the head up and down. Perform the swinging operation to the left and right. As described above, the operation unit 160 functions as an operation unit of the present invention.

  The sound input unit 170 includes a microphone unit including a plurality of microphones, an A / D (Analog to Digital) converter, and the like. For example, sound collected by a plurality of microphones installed at different positions on the head 102 is used. , And outputs digital audio data (audio information) on which signal processing such as A / D conversion and encoding has been performed to the control unit 110. The voice input unit 170 functions as a voice input unit of the present invention.

  The audio output unit 180 includes a speaker, a D / A (Digital to Analog) converter, and the like, and performs decoding, D / A conversion, amplification, and the like on audio data supplied from the control unit 110 (the response control unit 116). The signal processing is performed, and an analog audio signal is output from, for example, a speaker provided in the mouth 106. Thus, the audio output unit 180 functions as the audio output unit of the present invention.

  Next, the response control processing executed by the robot 100 will be described with reference to the flowchart shown in FIG. The response control process is a process of controlling the behavior of the robot 100 so that the robot 100 responds to the target user in a response manner according to the favorable feeling. The control unit 110 starts the response control process when the user recognition unit 111 recognizes a user existing around the robot 100.

  When starting the response control process, the control unit 110 first sets a favorable impression for the target user (step S101). The control unit 110 (the positivity setting unit 115) refers to the user action table and the positivity setting table stored in the storage unit 120, and specifies the level of positivity associated with the positivity point, thereby setting the target. Set the user's favorability. Note that the positivity setting unit 115 may acquire the preset current positivity included in the user information of the target user with reference to the user information DB 121.

  Next, the control unit 110 determines whether or not the target user exists around the robot 100 (Step S102). The control unit 110 confirms whether or not the target user is still around based on the captured image captured by the imaging unit 130. When it is determined that the target user does not exist around the robot 100 (step S102: NO), the control unit 110 ends the response control process.

  On the other hand, when it is determined that the target user exists around the robot 100 (step S102: YES), the control unit 110 determines whether the user action has been recognized (step S103). The control unit 110 (user action recognition unit 114) newly recognizes the user action based on, for example, the content of the utterance of the user recognized by the voice recognition unit 113, the contact strength detected by the touch sensor of the sensor unit 140, and the like. It is determined according to whether or not it has been performed.

  When determining that the user action is not recognized (step S103: NO), the control unit 110 determines whether or not to execute the self-issue operation (step S104). The control unit 110 (the response control unit 116) determines, for example, according to whether or not the time during which the user action recognition unit 114 does not recognize the user action has continued for a predetermined time. If it is determined that the self-issuing action is not to be executed (step S104: NO), the control unit 110 returns the process to step S102, and checks whether the target user is still around.

  When it is determined in step S103 that the user action has been recognized (step S103: YES), the control unit 110 (the response control unit 116) determines whether or not the approach condition is satisfied (step S105). The response control unit 116 determines, for example, according to whether or not there is a call for the target user to call the robot 100 near the target user.

  After executing step S105, or when it is determined in step S104 that the self-issuing operation is to be executed (step S104: YES), the control unit 110 (the response control unit 116) determines a response mode according to the favorable impression (step S104). S106). The response control unit 116 responds to the favorable response based on the determination result in step S104 (the determination result of executing the self-issue operation) and the determination result in step S105 (the determination result as to whether the approach condition is satisfied). To determine.

  If it is determined in step S104 that the self-issue operation is to be performed, the response control unit 116 appropriately adjusts the content of the utterance and the degree of the operation according to the favorable sensitivity set in step S101, and responds to the execution of the self-issue operation. Determine the mode.

  If it is determined in step S105 that the approach condition is not satisfied, the response control unit 116 determines a response mode for executing a response action to the user action recognized in step S103 according to the favorable feeling.

  If it is determined in step S105 that the approach condition is satisfied, the response control unit 116 determines, as the response mode, the moving route and the moving speed approaching the approaching user, the facing distance between the approaching user at the end of the approach, and the user. At least one of the opposing postures during approaching or at the end of approaching is determined.

  Subsequently, the control unit 110 controls each unit to execute various actions based on the response manner determined in step S106 (step S107). If the response control unit 116 determines in step S106 the response mode for executing the self-issuing operation according to the favorable impression, the response control unit 116 controls each unit based on the response mode and executes the self-issuing operation. If the response mode related to the self-issue is, for example, an utterance that prompts the target user to communicate with the robot 100, voice data indicating the content of the utterance is generated, and the voice output unit 180 is controlled to generate the generated voice data. Output the sound based on

  If the response control unit 116 determines in step S106 that the approach condition is not satisfied and the response mode for executing the response action to the user action recognized in step S103 according to the preference, based on the response mode, Each part is controlled to execute a response action. If the response mode related to the response action is, for example, an utterance that responds to the greeting of the target user, audio data indicating the utterance content is generated, and the audio output unit 180 is controlled to generate an audio based on the generated audio data. Output.

  In step S106, the response control unit 116 satisfies the approach condition, the moving route and the moving speed approaching the approaching user, the facing distance to the approaching user at the end of the approach, and the approaching or facing approaching user. When at least one of the facing postures at the time of the approach end is determined, the approaching behavior is executed by controlling the moving unit 150 and the operating unit 160 based on the response manner.

  After executing Step S107, the control unit 110 returns the process to Step S102, and checks whether the target user is still around.

  As described above, according to the present embodiment, the robot 100 determines the response mode according to the user's favorable feeling and controls each unit based on the determined response mode, thereby performing various actions. Execute. For example, when a predetermined approach condition is satisfied, the response control unit 116 determines, as a response mode, a moving route and a moving speed when approaching the approaching user, and an opposition between the approaching user at the end of the approach. At least one of the distance and the facing posture facing the user during approaching or at the end of approaching is determined according to the preference, and the approaching action is performed based on the determined response mode. Thereby, the response according to the favorable sensitivity can be improved.

  In addition to the case where the robot 100 satisfies the approach condition, the robot 100 determines a response mode related to the response action to the user action according to the favorable impression, and executes the response action based on the determined response mode. Further, the response mode relating to the self-issuing action is determined according to the favorable impression, and the self-issuing action is executed based on the determined response mode. Thereby, the response according to the favorable sensitivity can be improved.

  Note that the present invention is not limited to the above embodiment, and various modifications and applications are possible. The above embodiment may be modified as follows.

  In the above-described embodiment, the positivity setting unit (positivity acquisition unit) 115 sets or acquires the positivity for the predetermined target recognized by the user recognition unit 111, and the positivity setting unit (positivity acquisition unit). The response control unit (approach control unit) 116 has determined an approach state during approach to the predetermined object based on the favorable sensibility for the predetermined object set or acquired by the 115. However, the present invention is not limited to this, and the response control unit (approach control unit) 116 determines the approach state while approaching the predetermined object based on the favorable sensibility for the predetermined object recognized by the user recognition unit 111. May be determined.

  For example, when the user recognition unit 111 recognizes another user different from the approach target user on the moving route, the positivity setting unit 115 sets the positivity for the other user, and the response control unit 116 The response mode of the approach target user and the response mode of the other user may be determined according to the favorable impression of the other user. In this case, the response control unit 116 determines a detour route (passing route) for passing through the moving route determined as the response mode of the approach target user while avoiding other users. Here, the approaching user is an example of the predetermined object of the present invention, and the other user is an example of another object of the present invention.

  Explaining with reference to FIG. 12, the response control unit 116 is configured to, when there is another user UR ′ on the movement route MR approaching the approach target user UR, pass through the other user UR ′. Then, a new moving route MR that forms the detour route (passing route) DR is determined. At this time, the response control unit 116 forms the detour route DR in accordance with the favorable feeling for the other user UR '. The response control unit 116 forms the detour route DR so as to pass right next to the other user UR ', for example, when the favorableness of the other user UR' is high, and Is low, a bypass route DR that makes a large turn around another user UR 'is formed. In addition, the response control unit 116 determines, as a response mode of the other user UR ', a moving posture when the robot 100 passes through the detour route DR according to a favorable feeling for the other user UR'. The response control unit 116 determines that the front direction of the head 102 (face) of the robot 100 is different from the moving posture, for example, when the favorable attitude of the other user UR 'is high, when passing through the detour route DR. (It is assumed that the robot 100 is looking at the other user UR '), and when the other user UR' has a low preference, when the robot 100 passes through the detour route DR, the robot The front direction of the head 102 (face) of 100 is not directed to another user UR ′ (it is directed to the other side). Thereby, various expressions can be produced according to the favorable impression, and the response according to the favorable impression can be further improved.

  In the above embodiment, the robot 100 includes, for example, a communication unit having a wireless communication module or the like, and stores various data stored in the storage unit 120 using a cloud system built on the Internet. These data may be made accessible as needed.

  In addition, the robot 100 includes a display unit including a display device such as an LCD (Liquid Crystal Display), a PDP (Plasma Display Panel), and an organic EL (Electro-Luminescence) display. It may be further diversified. For example, a display unit may be provided in the eye unit 105 of the robot 100 to display an image expressing different eyes according to the liking, so that the expression of the robot 100 may be produced.

  In the above-described embodiment, the control program executed by the CPU of the control unit 110 is stored in the ROM or the like in advance. However, the present invention is not limited to this. By implementing a control program for executing the various processes described above on an existing general-purpose computer, a framework, and an electronic device such as a workstation, the present invention can be implemented. The device may function as a device corresponding to the robot 100 according to the embodiment.

  The method of providing such a program is arbitrary. For example, the program is stored in a computer-readable recording medium (flexible disk, CD (Compact Disc) -ROM, DVD (Digital Versatile Disc) -ROM) and distributed. Alternatively, the program may be stored in a storage on a network such as the Internet, and provided by downloading the program.

  In the case where the above processing is performed by sharing an OS (Operating System) and an application program, or when the OS and the application program cooperate with each other, only the application program may be stored in a recording medium or a storage. It is also possible to superimpose a program on a carrier wave and distribute it via a network. For example, the program may be posted on a bulletin board (Bulletin Board System: BBS) on a network, and the program may be distributed via the network. Then, the above-described processing may be executed by activating the distributed program and executing it in the same manner as other application programs under the control of the OS.

  The present invention is capable of various embodiments and modifications without departing from the broad spirit and scope of the invention. The above-described embodiment is for describing the present invention, and does not limit the scope of the present invention. That is, the scope of the present invention is shown not by the embodiments but by the claims. Various modifications made within the scope of the claims and within the meaning of the invention equivalent to the scope of the claims are considered to be within the scope of the invention.

  Hereinafter, the invention described in the claims of the present application is additionally described.

(Note)
(Appendix 1)
Recognition means for recognizing a predetermined object;
Determining means for determining an approach mode during approach to the predetermined object, based on the favorable sensitivity to the predetermined object recognized by the recognition means;
During the approach to the predetermined target, control means for controlling the own device to express emotions to the predetermined target according to the approach mode determined by the determination means,
Comprising,
A robot characterized in that:

(Appendix 2)
Further comprising a moving means for moving the own device,
The approach mode during approach to the predetermined object includes an approach path,
The control means, while approaching the predetermined object, controls the moving means to express the emotion on the predetermined object according to the determined approach route,
The robot according to supplementary note 1, characterized in that:

(Appendix 3)
Further comprising operating means for operating the own device,
The approach mode during approach to the predetermined object includes an approach posture,
The control means controls the operating means to express the emotion on the predetermined object according to the determined approaching posture during the approach to the predetermined object,
3. The robot according to claim 1 or 2, wherein

(Appendix 4)
Imaging means for capturing an image;
Face recognition means for performing face recognition by detecting a face from the image captured by the imaging means,
Voice input means for inputting voice,
Voice recognition means for performing voice recognition from the voice input by the voice input means,
Further comprising
The recognition unit recognizes the predetermined target by the face recognition unit and / or the voice recognition unit.
4. The robot according to any one of supplementary notes 1 to 3, wherein:

(Appendix 5)
Acquisition means for acquiring the positivity for the predetermined object recognized by the recognition means,
Behavior recognition means for recognizing the behavior of the predetermined target with respect to the own device,
Further comprising
The acquisition unit acquires the favorable impression according to the behavior of the predetermined target recognized by the behavior recognition unit.
The robot according to any one of supplementary notes 1 to 4, characterized in that:

(Appendix 6)
The acquisition means,
If the behavior of the predetermined object recognized by the behavior recognition means is a favorable behavior, the degree of the favorable impression is increased from the current favorable impression, and if the behavior is not the favorable behavior, Lowering the degree of the favorability than the current favorability,
The robot according to supplementary note 5, wherein

(Appendix 7)
Further comprising recording means for recording information;
The recording means,
When the behavior of the predetermined object recognized by the behavior recognition means is the favorable behavior, the favorable impression is recorded higher than a threshold value, and when the behavior is not the favorable behavior, the favorable impression is recorded. Is recorded below the threshold,
The obtaining means obtains the favorable sensitivity of the predetermined object via the recording means,
7. The robot according to supplementary note 6, wherein:

(Appendix 8)
The action recognition means,
When the own device is stroked by the predetermined object, or when the own device is spoken by the predetermined object, the action of the predetermined object is recognized as the favorable action,
When the own device is hit by the predetermined object, or when the own device is ignored by the predetermined object, the action of the predetermined object is recognized as not the favorable action.
8. The robot according to appendix 6 or 7, wherein:

(Appendix 9)
The control means may determine, as the approach mode during approach to the predetermined object, an approach route, an approach speed, an opposing angle, and an approach attitude when the own device approaches the predetermined object based on the favorable sensitivity. decide,
The robot according to supplementary note 1, characterized in that:

(Appendix 10)
The control means, while approaching the predetermined object,
A first opposing angle corresponding to a first approaching posture such as facing the predetermined object from the front at a first approach speed on a linear first approach path when the favorable sensitivity is higher than a threshold value; In, the own device controls the own device so as to approach the predetermined target,
When the favorable sensitivity is lower than the threshold, a second meandering second route that does not face the predetermined object from the front at a second approach speed lower than the first approach speed in a meandering second approach route. At a second facing angle corresponding to the approaching posture, the own device is controlled so that the own device approaches the predetermined target,
The robot according to supplementary note 9, wherein:

(Appendix 11)
The control means, while approaching the predetermined object, as the emotional expression,
When the favorable impression is higher than a threshold, the own device is controlled to perform a favorable operation or a preferential operation to the predetermined target,
When the favorable impression is lower than the threshold, the own device is controlled to perform an unfavorable operation or a non-priority operation on the predetermined target,
The robot according to supplementary note 1, characterized in that:

(Appendix 12)
The control means, while approaching the predetermined object, changes the degree of the favorable operation or the degree of the preferential operation for the predetermined object according to the favorable sensitivity,
12. The robot according to supplementary note 11, wherein:

(Appendix 13)
The control means, while approaching the predetermined object, changes the degree of the unfavorable operation or the degree of the non-priority operation with respect to the predetermined object according to the favorable sensitivity,
13. The robot according to appendix 11 or 12, wherein

(Appendix 14)
The predetermined object includes a human or an animal or another robot,
14. The robot according to any one of supplementary notes 1 to 13, wherein:

(Appendix 15)
While approaching the predetermined object, when the recognition unit recognizes another object different from the predetermined object,
The determining means, based on another favorable feeling for the other object recognized by the recognition means, determines a passing mode during the passing for the other object,
The control means, during the passage to the other object, controls the own device to express another emotion to the other object according to the passage manner determined by the determination means,
The robot according to supplementary note 1, characterized in that:

(Appendix 16)
The control means includes:
As the passing mode to the other object, determine a passage route to pass avoiding the other object,
16. The robot according to supplementary note 15, wherein:

(Appendix 17)
The control means includes:
As the passing mode to the other object, determine a passing posture when passing through the passing route,
19. The robot according to supplementary note 16, wherein:

(Appendix 18)
The other object includes a human or animal or other robot,
18. The robot according to any one of supplementary notes 15 to 17, wherein:

(Appendix 19)
A recognition step of recognizing a predetermined object;
A determining step of determining an approach mode during approaching the predetermined object based on the favorable sensibility for the predetermined object recognized in the recognition step;
During the approach to the predetermined object, a control step of controlling a robot to express an emotion to the predetermined object according to the approach mode determined in the determination step,
including,
A method for controlling a robot, comprising:

(Appendix 20)
On the robot computer,
A recognition process for recognizing a predetermined object;
A determination process of determining an approach mode during approach to the predetermined object based on the favorable sensitivity to the predetermined object recognized by the recognition process;
During the approach to the predetermined object, a control process of controlling the robot so as to express an emotion on the predetermined object in accordance with the approach mode determined by the determination process,
Run,
A program characterized by the following.

Reference Signs List 100 robot, 101 body part, 102 head part, 103 hand part, 104 foot part, 105 eye part, 106 mouth part, 110 control part, 111 user recognition part, 112 face recognition part Reference numeral 113, a voice recognition unit, 114, a user behavior recognition unit, 115, a positivity setting unit (a positivity acquisition unit), 116, a response control unit (approach control unit), 120, a storage unit, 121, a user information DB, 122 ... voice recognition DB, 123 ... dialog information DB, 130 ... imaging unit, 140 ... sensor unit, 150 ... moving unit, 160 ... operation unit, 170 ... voice input unit, 180 ... voice output unit, BL ... bus line, DR ... Detour route (passing route), MR: moving route (approaching route), RD: front direction, UD: front direction, UR: approaching user (predetermined target), UR ': other user (other target)

Claims (20)

Recognition means for recognizing a predetermined object;
Determining means for determining an approach mode during approach to the predetermined object, based on the favorable sensitivity to the predetermined object recognized by the recognition means;
During the approach to the predetermined target, control means for controlling the own device to express emotions to the predetermined target according to the approach mode determined by the determination means,
Comprising,
A robot characterized in that: Further comprising a moving means for moving the own device,
The approach mode during approach to the predetermined object includes an approach path,
The control means, while approaching the predetermined object, controls the moving means to express the emotion on the predetermined object according to the determined approach route,
The robot according to claim 1, wherein: Further comprising operating means for operating the own device,
The approach mode during approach to the predetermined object includes an approach posture,
The control means controls the operating means to express the emotion on the predetermined object according to the determined approaching posture during the approach to the predetermined object,
The robot according to claim 1, wherein: Imaging means for capturing an image;
Face recognition means for performing face recognition by detecting a face from the image captured by the imaging means,
Voice input means for inputting voice,
Voice recognition means for performing voice recognition from the voice input by the voice input means,
Further comprising
The recognition unit recognizes the predetermined target by the face recognition unit and / or the voice recognition unit.
The robot according to any one of claims 1 to 3, wherein: Acquisition means for acquiring the positivity for the predetermined object recognized by the recognition means,
Behavior recognition means for recognizing the behavior of the predetermined target with respect to the own device,
Further comprising
The acquisition unit acquires the favorable impression according to the behavior of the predetermined target recognized by the behavior recognition unit.
The robot according to any one of claims 1 to 4, wherein: The acquisition means,
If the behavior of the predetermined object recognized by the behavior recognition means is a favorable behavior, the degree of the favorable impression is increased from the current favorable impression, and if the behavior is not the favorable behavior, Lowering the degree of the favorability than the current favorability,
The robot according to claim 5, wherein: Further comprising recording means for recording information;
The recording means,
When the behavior of the predetermined object recognized by the behavior recognition means is the favorable behavior, the favorable impression is recorded higher than a threshold value, and when the behavior is not the favorable behavior, the favorable impression is recorded. Is recorded below the threshold,
The obtaining means obtains the favorable sensitivity of the predetermined object via the recording means,
The robot according to claim 6, wherein: The action recognition means,
When the own device is stroked by the predetermined object, or when the own device is spoken by the predetermined object, the action of the predetermined object is recognized as the favorable action,
When the own device is hit by the predetermined object, or when the own device is ignored by the predetermined object, the action of the predetermined object is recognized as not the favorable action.
The robot according to claim 6, wherein: The control means sets the approach route, approach speed, opposing angle, approach attitude when the own device approaches the predetermined object as the approach mode during approach to the predetermined object based on the favorable sensitivity. decide,
The robot according to claim 1, wherein: The control means, while approaching the predetermined object,
A first opposing angle corresponding to a first approaching posture such as facing the predetermined object from the front at a first approach speed on a linear first approach path when the favorable sensitivity is higher than a threshold value; In, the own device controls the own device so as to approach the predetermined target,
When the favorable sensitivity is lower than the threshold, a second meandering second route that does not face the predetermined object from the front at a second approach speed lower than the first approach speed in a meandering second approach route. At a second facing angle corresponding to the approaching posture, the own device is controlled so that the own device approaches the predetermined target,
The robot according to claim 9, wherein: The control means, while approaching the predetermined object, as the emotional expression,
When the favorable impression is higher than a threshold, the own device is controlled to perform a favorable operation or a preferential operation to the predetermined target,
When the favorable impression is lower than the threshold, the own device is controlled to perform an unfavorable operation or a non-priority operation on the predetermined target,
The robot according to claim 1, wherein: The control means, while approaching the predetermined object, changes the degree of the favorable operation or the degree of the preferential operation for the predetermined object according to the favorable sensitivity,
The robot according to claim 11, wherein: The control means, while approaching the predetermined object, changes the degree of the unfavorable operation or the degree of the non-priority operation with respect to the predetermined object according to the favorable sensitivity,
The robot according to claim 11, wherein: The predetermined object includes a human or an animal or another robot,
The robot according to any one of claims 1 to 13, wherein: While approaching the predetermined object, when the recognition unit recognizes another object different from the predetermined object,
The determining means, based on another favorable feeling for the other object recognized by the recognition means, determines a passing mode during the passing for the other object,
The control means, during the passage to the other object, controls the own device to express another emotion to the other object according to the passage manner determined by the determination means,
The robot according to claim 1, wherein: The control means includes:
As the passing mode to the other object, determine a passage route to pass avoiding the other object,
The robot according to claim 15, wherein: The control means includes:
As the passing mode to the other object, determine a passing posture when passing through the passing route,
The robot according to claim 16, wherein: The other object includes a human or animal or other robot,
The robot according to any one of claims 15 to 17, wherein: A recognition step of recognizing a predetermined object;
A determining step of determining an approach mode during approaching the predetermined object based on the favorable sensibility for the predetermined object recognized in the recognition step;
During the approach to the predetermined object, a control step of controlling a robot to express an emotion to the predetermined object according to the approach mode determined in the determination step,
including,
A method for controlling a robot, comprising: On the robot computer,
A recognition process for recognizing a predetermined object;
A determination process of determining an approach mode during approach to the predetermined object based on the favorable sensitivity to the predetermined object recognized by the recognition process;
During the approach to the predetermined object, a control process of controlling the robot so as to express an emotion on the predetermined object in accordance with the approach mode determined by the determination process,
Run,
A program characterized by the following.

Images