JP2004042151A - Communication robot - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a communication robot for allowing a person to communicate without feeling a sense of incompatibility. <P>SOLUTION: This communication robot 10 includes an eyeball part 76, and an eye camera 46 is arranged in the eyeball part 76. An object position such as the person and a specific object in an image acquired from the eye camera 46 is detected, and the eyeball part 76 is displaced so as to turn in the object direction. Thus, the communication robot 10 becomes a shape of adjusting a line of sight to the person or steadily gazing the object direction. The eyeball part 76 is displaced in relation to a movement of a head part 42 such as turning the line of sight in the head part 42-turning direction when a head part 42 moves. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
[Industrial applications]
The present invention relates to a communication robot, and more particularly, to a communication robot having a function of communicating, for example, having a portion corresponding to human eyes or eyes, that is, an eyeball.
[0002]
[Prior art]
2. Description of the Related Art Some conventional robots perform eye movements to convey a certain intention, such as moving or rotating an eyeball part left and right, for example.
[0003]
[Problems to be solved by the invention]
However, in the prior art, only a single goal-directed eye movement is performed mechanically, and a sense of life cannot be created. Therefore, there is a fear that a person trying to communicate may feel uncomfortable and the feeling of communicating may be hindered.
[0004]
Therefore, a main object of the present invention is to provide a communication robot that can communicate without a sense of incongruity.
[0005]
[Means for Solving the Problems]
The present invention relates to a communication robot having a function of performing communication, comprising: an eyeball, an object detection unit for detecting a target object, and an eyeball unit displacement unit for displacing the eyeball so as to face the object. , A communication robot.
[0006]
[Action]
An eyeball portion that is visually recognized as representing the eye is provided, for example, at a position corresponding to the eye. The target object is detected by the object detection means. For example, a camera is provided at an appropriate position of the communication robot, and the first detection unit detects an object in the image based on the image acquired from the camera, and calculates, for example, a position and a direction thereof. As the object, for example, a human face, a specific object, or the like can be set. In addition, when a plurality of touch sensors are provided at arbitrary positions of the communication robot, the second detection unit detects, for example, a touch sensor that is turned on by being touched by a person as an object. Then, the eyeball is displaced by the eyeball displacing means so as to face the direction of the detected object. In other words, the line of sight is directed toward a person's face, a specific object, or a touched location. Therefore, for example, when interacting with a person, the communication robot looks like a line of sight with the person, or gazes at a specific object direction or a touched part of a topic.
[0007]
In the case of including the head, the eyeball is provided on the head. Since the head is a portion corresponding to the head, the head is attached to, for example, a torso via a neck joint, and is displaced by head displacing means. The head can be displaced, for example, such that the front side of the head, ie the face, faces the direction of the object. The eyeball can be displaced by the eyeball displacing means in relation to the movement of the head. Specifically, for example, angle data proportional to the movement speed of the head is calculated, and the eyeball is displaced based on the angle data, whereby the line of sight is directed in the direction in which the head is headed. In this way, the eyeball section can make a natural eyeball movement according to the movement of the head when the head is moving.
[0008]
【The invention's effect】
According to the present invention, it is possible to match a line of sight with a person or to gaze at an object direction. That is, communication can be achieved by natural eye movements. Further, the eyeball can perform eye movements related to the movement of the head. Therefore, since communication is performed with extremely natural eye movements similar to that of a person, the person can communicate with the communication robot without feeling uncomfortable.
[0009]
The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description of embodiments with reference to the drawings.
[0010]
【Example】
FIG. 1 shows an overall configuration of the present invention. Referring to FIG. 1, a communication robot (hereinafter, may be simply referred to as a “robot”) 10 of this embodiment includes a bogie 12. Wheels 14 for autonomously moving the robot 10 are provided on the lower surface of the robot. The wheels 14 are driven by wheel motors (indicated by reference numeral “70” in FIG. 4), and can move the carriage 12, that is, the robot 10, in any direction in the front, rear, left, and right directions. Although not shown, a collision sensor (indicated by reference numeral “74” in FIG. 4) is mounted on the front surface of the cart 12, and the collision sensor is used to prevent a person or other obstacles from entering the cart 12. Detect contact. Then, when contact with an obstacle is detected while the robot 10 is moving, the driving of the wheels 14 is immediately stopped, and the movement of the robot 10 is suddenly stopped to prevent collision.
[0011]
In this embodiment, the height of the robot 10 is set to about 100 cm so as not to give a feeling of intimidation to a person, especially a child. However, this height can be arbitrarily changed.
[0012]
A polygonal column sensor mounting panel 16 is provided on the carriage 12, and an ultrasonic distance sensor 18 is mounted on each surface of the sensor mounting panel 16. The ultrasonic distance sensor 18 measures a distance between the mounting panel 16, that is, a person mainly around the robot 10.
[0013]
The lower part of the robot 12 is mounted on the carriage 12 so that the lower part is surrounded by the mounting panel 16 and the body of the robot 10 stands upright. The body includes a lower body 20 and an upper body 22, and the lower body 20 and the upper body 22 are connected by a connecting portion 24. Although not shown, the connecting portion 24 has a built-in elevating mechanism, and by using the elevating mechanism, the height of the upper body 22, that is, the height of the robot 10 can be changed. The lifting mechanism is driven by a waist motor (indicated by reference numeral “68” in FIG. 4) as described later. The height 100 cm of the robot 10 described above is a value when the upper body 22 is at the lowest position thereof. Therefore, the height of the robot 10 can be set to 100 cm or more.
[0014]
One omnidirectional camera 26 and one microphone 28 are provided substantially at the center of the upper body 22. The omnidirectional camera 26 captures an image around the robot 10 and is distinguished from an eye camera 46 described later. The microphone 28 captures ambient sounds, especially human voices.
[0015]
Upper arms 32R and 32L are attached to both shoulders of upper torso 22 by shoulder joints 30R and 30L, respectively. The shoulder joints 30R and 30L each have three degrees of freedom. That is, the shoulder joint 30R can control the angle of the upper arm 32R around each of the X axis, the Y axis, and the Z axis. The Y axis is an axis parallel to the longitudinal direction (or axis) of the upper arm 32R, and the X axis and the Z axis are axes orthogonal to the Y axis from different directions. The shoulder joint 30L can control the angle of the upper arm 32L around each of the A axis, the B axis, and the C axis. The B axis is an axis parallel to the longitudinal direction (or axis) of the upper arm 32L, and the A axis and the C axis are axes orthogonal to the B axis from different directions.
[0016]
Forearms 36R and 36L are attached to respective distal ends of the upper arms 32R and 32L via elbow joints 34R and 34L. The elbow joints 34R and 34L can control the angles of the forearms 36R and 36L around the W axis and the D axis, respectively.
[0017]
In the X, Y, X, W axes and the A, B, C, D axes for controlling the displacement of the upper arms 32R and 32L and the forearms 36R and 36L (FIG. 1), "0 degree" is the home position. In this home position, the upper arms 32R and 32L and the forearms 36R and 36L are directed downward.
[0018]
Further, although not shown, touch sensors are provided on the shoulder portion of the upper body 22 including the shoulder joints 30R and 30L, the upper arms 32R and 32L, and the forearms 36R and 36L, respectively, and these touch sensors are provided. , To detect whether a person has contacted these parts of the robot 10. These touch sensors are also indicated generally by reference numeral 72 in FIG.
[0019]
Balls 38R and 38L corresponding to the hands are fixedly attached to the tips of the forearms 36R and 36L, respectively. In place of the spheres 38R and 38L, when a function of a finger is required unlike the robot 10 of this embodiment, a "hand" shaped like a human hand can be used.
[0020]
Although the shape and dimensions of the robot 10 are appropriately set, in other embodiments, for example, the upper body 22 includes a front surface, a back surface, a right side surface, a left side surface, a top surface and a bottom surface, and a right side surface and a left side. The surface may be formed so that the surface faces diagonally forward. In other words, the upper body 22 may be formed to have a trapezoidal shape in which the width of the front surface is shorter than the width of the rear surface, and the shape of the upper body 22 viewed from above is trapezoidal. In such a case, the shoulder joints 30R and 30L are attached to the right side surface and the left side surface via left and right support parts whose surfaces are parallel to the left and right side surfaces, respectively. The rotation range of the upper arm 32R and the upper arm 32L is regulated by the left and right side surfaces or the surface (mounting surface) of the support portion, and the upper arms 32R and 32L do not rotate beyond the mounting surface. However, if the inclination angles of the left and right side surfaces, the interval between the B axis and the Y axis, the lengths of the upper arms 32R and 32L, and the lengths of the forearms 36R and 36L are appropriately set, the upper arms 32R and 32L can move beyond the front. Since the robot can be pivoted further inward, the arms of the robot 10 can intersect forward even if there is no freedom of the arms by the W axis and the D axis. Therefore, even when the degree of freedom of the arm is small, close communication such as hugging with a person located in front can be achieved.
[0021]
Above the center of the upper body 22, a head 42 is attached via a neck joint 40. The neck joint 40 has three degrees of freedom, and can be angle-controlled around each of the S axis, T axis, and U axis. The S-axis is an axis directly above the neck, and the T-axis and the U-axis are axes orthogonal to the S-axis in different directions. A speaker 44 is provided on the head 42 at a position corresponding to the mouth of a person. The speaker 44 is used by the robot 10 to communicate by voice or voice with people around it. However, the speaker 44 may be provided on another part of the robot 10, for example, on the body.
[0022]
Further, the head 42 is provided with eyeball portions 76R and 76L at positions corresponding to the eyes. The eyeball portions 76R and 76L include eye cameras 46R and 46L, respectively. Note that the right eyeball 76R and the left eyeball 76L are collectively referred to as the eyeball 76, and the right eye camera 46R and the left eye camera 46L are sometimes referred to as the eye camera 46. The eye camera 46 captures an image of a person's face and other parts or objects approaching the robot 10 and captures the video signal.
[0023]
Each of the omnidirectional camera 26 and the eye camera 46 may be a camera using a solid-state imaging device such as a CCD or a CMOS.
[0024]
For example, the eye camera 46 is fixed in the eyeball 76, and the eyeball 76 is attached to a predetermined position in the head 42 via an eyeball support (not shown). The eyeball support has two degrees of freedom, and can be angle-controlled around each of the α axis and the β axis (FIG. 2). The α axis and the β axis are axes set with respect to the head 42, the α axis is an axis in a direction going upward of the head 42, and the β axis is orthogonal to the α axis and is on the front side of the head 42. This is an axis in a direction orthogonal to the direction in which (face) faces. In this embodiment, when the head 42 is at the home position, the α axis is set to be parallel to the S axis, and the β axis is set to be parallel to the U axis. In such a head 42, the eyeball supporting portion is rotated around each of the α axis and the β axis, whereby the eyeball portion 76 or the tip (front) side of the eye camera 46 is displaced, and the camera axis, that is, the line of sight is changed. Moved (FIG. 3).
[0025]
In addition, in the α axis and the β axis for controlling the displacement of the eye camera 46, “0 degree” is the home position. In this home position, the camera axis of the eye camera 46 is, as shown in FIG. The front side (face) is turned in a direction to face, and the line of sight is in a normal vision state.
[0026]
The configuration of the control system of the robot 10 shown in FIG. 1 is shown in the block diagram of FIG. As shown in FIG. 4, the robot 10 includes a microcomputer or a CPU 50 for overall control. The CPU 50 is connected to a memory 54, a motor control board 56, a sensor input / output board 58, and an audio The input / output board 60 is connected.
[0027]
Although not shown, the memory 54 includes a ROM and a RAM. The ROM stores a control program of the robot 10 in advance and stores voice or voice data to be generated from the speaker 44. The RAM can be used as a temporary storage memory and also as a working memory.
[0028]
The motor control board 56 is composed of, for example, a DSP (Digital Signal Processor), and controls each axis motor of each arm, head, eyeball, and the like. That is, the motor control board 56 receives the control data from the CPU 50, and controls the three motors for controlling the respective angles of the X, Y and Z axes of the right shoulder joint 30R and the angle of the axis W of the right elbow joint 34R. The rotation angles of a total of four motors (one motor and a right motor) 62 in FIG. 4 are adjusted. The motor control board 56 has four motors, three motors for controlling the angles of the A, B and C axes of the left shoulder joint 30L and one motor for controlling the angle of the D axis of the left elbow joint 34L. (They are collectively shown as “left arm motor” in FIG. 4.) The rotation angle of 64 is adjusted. The motor control board 56 also adjusts the rotation angles of three motors (collectively shown in FIG. 4 as “head motors”) 66 that control the respective angles of the S, T and U axes of the head 42. I do. The motor control board 56 also controls a waist motor 68 and two motors 70 for driving the wheels 14 (collectively shown as “wheel motors” in FIG. 4).
[0029]
Further, the motor control board 56 adjusts the rotation angles of two motors (collectively shown as “right eyeball motor” in FIG. 4) 78 for controlling the respective angles of the α axis and the β axis of the right eyeball 76R. Further, the rotation angle of two motors (collectively shown as “left eyeball motor” in FIG. 4) 80 for controlling the respective angles of the α axis and the β axis of the left eyeball section 76L is adjusted.
[0030]
The above-described motors in this embodiment are stepping motors or pulse motors for simplifying the control except for the wheel motor 70, but may be DC motors like the wheel motor 70.
[0031]
Similarly, the sensor input / output board 58 is also configured by a DSP, and takes in signals from each sensor and camera and supplies them to the CPU 50. That is, data on the reflection time from each of the ultrasonic distance sensors 18 is input to the CPU 50 through the sensor input / output board 58. A video signal from the omnidirectional camera 26 is input to the CPU 50 after being subjected to predetermined processing by the sensor input / output board 58 as necessary. The video signal from the eye camera 46 is similarly supplied to the CPU 50. In FIG. 4, the touch sensors described in FIG. 1 are collectively represented as “touch sensors 72”, and signals from those touch sensors 72 are given to the CPU 50 via the sensor input / output board 58. .
[0032]
Synthesized voice data is supplied from the CPU 50 to the speaker 44 via the voice input / output board 60, and a voice or voice according to the data is output from the speaker 44 in response thereto. Then, the voice input from the microphone 28 is taken into the CPU 50 via the voice input / output board 60.
[0033]
The robot 10 of this embodiment includes an eyeball 76, detects the position of a specific target object such as a person or an object, and displaces the eyeball 76 so as to face the object. Further, in the robot 10, eye movement related to the movement of the head 42 is performed.
[0034]
For example, in the robot 10, when a human face is present in the image acquired from the eye camera 46, the position of the human face as the target object is detected, and the eyeball unit 76 and the eye camera 46 are turned in that direction. Control angle data of α-axis and β-axis are calculated. When the angle data is given to the right eyeball motor 78 and the left eyeball motor 80, the eyeball 76 and the eye camera 46 are displaced, and the line of sight is directed toward the face of the person. Therefore, the robot 10 can match the line of sight with the person, and the person can communicate without feeling uncomfortable.
[0035]
When the head 42 is operating, control angle data of the α-axis and β-axis are calculated based on, for example, the movement speed around the axes (S-axis and U-axis) for controlling the neck joint. In some cases, the eyeball support may be controlled. In this case, the movement of the eyeball portion 76 that matches the movement of the head 42 can be represented, and it is possible to prevent a person from feeling uncomfortable.
[0036]
In addition, the robot 10 can detect the position of a specific object in the image acquired from the eye camera 46 as well as the face of the person, and turn his or her gaze in that direction. That is, for example, when introducing a poster or the like affixed to a wall or the like, a natural operation such as detecting the position of a referral such as a poster and performing an introduction while interposing an operation of turning a face toward this is introduced. You can do it too.
[0037]
Specifically, this robot 10 operates according to the flowcharts shown in FIGS. 5 and 6, for example. This flowchart shows an example of the operation when introducing a poster installed on a wall.
[0038]
First, in steps S1 to S5, initialization is performed. In step S1, the CPU 50 extracts angle data from the memory 54 and sends it to the motor control board 56 so that the head 42 is directed forward. That is, an angle “0” is given to each of the motors (the head motor 66) that control the rotation angles of the S axis, the T axis, and the U axis. Therefore, in this step S1, the head 42 faces the front. In step S <b> 3, the CPU 50 extracts the angle data from the memory 54 and sends it to the motor control board 56 so that the eyeball unit 76 faces the front. That is, the angle “0” is given to the motors (the right eyeball motor 78 and the left eyeball motor 80) that control the rotation angles of the α axis and the β axis. Therefore, in this step S3, the eyeball unit 76 and the eye camera 46 face the front, and a normal vision state in which the line of sight faces the front. In step S5, the value of the current time is set in the variables t1, t2, and t3. Although not shown, the variable c is set to a value smaller than 1 as an initial setting.
[0039]
In the following step S7, the CPU 50 captures the video signal from the eye camera 46 into the sensor input / output board 58, and in step S9, processes the video signal to detect the position of a human face in the camera image, if any. . Various methods can be used as a method for determining a human face. For example, a skin color portion may be determined to be a human face. Further, for example, data such as a specific person's face or characteristics may be stored in the memory 54 in advance, and the line of sight may be adjusted only to the specific person by referring to the data.
[0040]
In step S11, it is determined whether a predetermined time has elapsed from time t1 and whether a human face is present in the image. If “YES” in the step S11, in a step S15, the CPU 50 calculates the angle data of the α axis and the β axis corresponding to the detected direction of the face of the person, and transmits the angle data to the motor control board 56. Send to The calculated α-axis and β-axis control angles are given to the right eyeball motor 78 and the left eyeball motor 80, respectively. Therefore, in step S15, if there is a person in the field of view, the eyeball 76 is turned toward the face of the person, and the person and the robot 10 can match their eyes.
[0041]
In the following step S17, the current time is set in the variable t1, and after this step S17, the process proceeds to step S25. As described above, when the processing in step S15 is performed and the line of sight is turned toward the face of the person, the variable t1 is reset to the current time. Therefore, the processing in step S15 is performed until a certain time has elapsed thereafter. Is not executed.
[0042]
On the other hand, if “NO” in the step S11, that is, if the fixed time has not yet elapsed from the time t1, or if there is no human face in the image, the CPU 50 determines in the step S13 that the time t2 It is determined whether or not a predetermined time has passed since and the motor for adjusting the rotation angle of the S-axis or the U-axis of the head motor 66 is operating.
[0043]
If “YES” in the step S13, the moving speeds Vs and Vu of the operating S axis and the U axis are detected in a step S19, and the control angles of the α axis and the β axis are determined based on the moving speeds Vs and Vu. Calculate the data. Specifically, the rotation angle of the α-axis is set to a value (aVs) proportional to the movement speed Vs of the S-axis, and the rotation angle of the β-axis is set to a value (bVu) proportional to the movement speed Vu of the U-axis. You. Here, a and b are constants and are set as appropriate. Then, the angle data is sent to the motor control board 56, and the angle “aVs” is given to the motor controlling the α-axis rotation angle, and the angle “bVu” is given to the motor controlling the β-axis rotation angle. Then, since the α-axis and the β-axis correspond to the S-axis and the U-axis, respectively, the eyeball 76 is directed in the moving direction of the head 42, that is, the direction in which the face of the head 42 faces. In this way, when the head 42 is moved, the line of sight is directed in the direction in which the head 42 is moved, and an extremely natural eye movement according to the head movement can be realized. In this embodiment, specifically, such eye movements are realized when the head 42 operates by the processing of steps S35 and S39 described later. In the following step S21, the current time is set in the variable t2, and after this step S21, the process proceeds to step S25. As described above, since the variable t2 is reset to the current time, the process of step S19 is not performed until a certain time has elapsed thereafter.
[0044]
On the other hand, if “NO” in the step S13, that is, if the fixed time has not elapsed from the time t2, or if the motor for adjusting the rotation angles of the S axis and the U axis is not operating, In a succeeding step S23, the CPU 50 takes out the angle data from the memory 54 and sends it to the motor control board 56 so that the eyeball portion 76 faces the front. That is, the angle “0” is given to the motors (the right eyeball motor 78 and the left eyeball motor 80) that control the rotation angles of the α axis and the β axis. Therefore, in this step S23, the eyeball section 76 and the eye camera 46 are turned to the front. That is, for example, the line of sight moved in step S15 or step S19 is returned to the front.
[0045]
After finishing the processing in step S17, S21 or S23, the CPU 50 determines in a succeeding step S25 whether a predetermined time has elapsed from the time t3. If “NO” in the step S25, the process returns to the step S7. Therefore, the processing from step S7 to step S25 is repeatedly performed until a certain time has elapsed from time t3. On the other hand, if “YES” in the step S25, the process proceeds to a step S27 in FIG.
[0046]
In step S27, the CPU 50 determines whether or not the variable c is 1 or more. If “NO” in the step S27, that is, if the variable c is smaller than 1, the process proceeds to a step S31. In step S31, the CPU 50 captures the video signal from the eye camera 46 into the sensor input / output board 58, and in step S33 performs the video signal processing to detect the position of the poster on the wall from the camera image. Note that data such as the characteristics of the poster is stored in the memory 54 in advance, and the poster position is detected by referring to this data.
[0047]
In this embodiment, it is assumed that a poster is set within the field of view of the robot 10, but if the poster is not in the image acquired in step S31, for example, the processing of steps S35 and S37 is skipped. The user may move on, or the head 42 may be moved to scan the surroundings to find the poster.
[0048]
In step S35, the CPU 50 calculates S-axis, T-axis, and U-axis angle data corresponding to the direction of the detected poster position, and sends the data to the motor control board 56. Therefore, in step S35, the front side (face) of the head 42 is turned in the direction in which the poster is located, and the robot 10 looks like watching the poster.
[0049]
Then, in step S37, the variable t3 is set to the current time, and the variable c is set to 1. After step S37, the process returns to step S7 in FIG. 5, and the process is repeated. As described above, since the variable t3 is reset to the current time, the processing after step S27 is not executed until a certain time has elapsed thereafter. Further, since the variable c is set to 1, when the next step S27 is processed, “YES” is determined and the process proceeds to step S29.
[0050]
On the other hand, if “YES” in the step S27, it is determined whether or not the variable c is 2 or more in a step S29. If “NO” in the step S29, that is, if the variable c is 1 or more and less than 2, the process proceeds to a step S39. In step S39, the CPU 50 takes out the angle data from the memory 54 and sends it to the motor control board 56 so that the head 42 faces upward. That is, an angle “0” is given to each of the motors controlling the rotation angles of the S-axis and the T-axis, and an angle “45” is given to the motor controlling the rotation angles of the U-axis. Therefore, in this step S39, the head 42 that has been turned in the poster direction is turned forward and obliquely upward, and looks like looking up. Note that this appearance is an operation assuming, for example, a case where a person is guided in front of the robot 10 in advance.
[0051]
Then, in a step S41, the CPU 50 extracts predetermined audio data from the memory 54 and sends it to the audio input / output board 60. Therefore, a synthesized voice such as “Look at the poster” is output from the speaker 44. In this way, by outputting a synthesized voice related to a specific object, it is possible to interact with a person and to convey the meaning of the movement of the head 12 and the eyeball 76.
[0052]
In this way, it is possible to talk with a person while sandwiching an extremely natural operation such as an operation of turning the face to the poster to be introduced (step S35). It should be noted that even when there is no nearby person, it can be expected that the person comes closer by such an operation to invite a person or an operation to emit a voice (steps S39 and S41).
[0053]
In a succeeding step S43, the variable t3 is set to the current time again, and the variable c is set to 2. After step S43, the process returns to step S7 of FIG. 2, and the process is repeated. As described above, since the variable t3 is set to the current time again, the processes after step S27 are not executed until a certain time has elapsed thereafter. In addition, since the variable c is set to 2, the CPU 50 determines “YES” when processing the next step S29, and ends the operation processing of the head 42 and the eyeball section 76. It is needless to say that such a series of operation processes in FIGS. 5 and 6 may be repeatedly executed, for example, every predetermined time.
[0054]
According to this embodiment, when interacting with a person or the like, it is possible to match the line of sight with the person or gaze at a specific object direction to be talked about. That is, communication can be achieved by natural eye movements. Further, the eyeball portion 76 can perform eyeball movement related to the movement of the head portion 42. Therefore, since communication is performed with extremely natural eye movements similar to those of humans and animals, humans can communicate with the robot 10 without feeling uncomfortable.
[0055]
The embodiments of the present invention may be applied with various changes. For example, in the above-described embodiment, the head 12 is displaced in step S35 so that the face is turned to the poster. However, only the eyeball 76 may be turned to the poster.
[0056]
Further, in each of the above-described embodiments, the face of the head 42 is directed in a predetermined direction such as obliquely upward and forward in step S39. However, for example, processing similar to steps S7, S9, and S35 may be performed. Then, the position of the human face may be detected, and the face of the head 42 may be turned in the direction of the face.
[0057]
In each of the above-described embodiments, it is assumed that there is a person in the field of view of the robot 10. For example, before processing step S7, the user moves the head 42 to scan the surroundings to detect the person. If it is possible to find the person, or if there is no person even looking around, for example, as in step S41, a synthetic voice or the like may be issued in advance to bring the person closer or into the field of view. Good. In addition, as a behavior for calling a person, other methods such as emitting light may be applied.
[0058]
Further, in another embodiment, for example, when touched by a person, the touch sensor 72 in the ON state is detected to calculate the position of the touched point, and the necessary angle data is given to the necessary motors. The head 42 and the eyeball section 76 may be appropriately moved so as to direct the line of sight in that direction. That is, the on state touch sensor 72 may be detected as an object, and the displacement of the eyeball 76 and the head 12 may be controlled so as to face the direction of the object. Also in this case, similarly to the above-described embodiments, the eyeball portion 76 can be moved in association with the movement of the head 42, and an extremely natural eyeball movement can be performed. I don't remember.
[0059]
Further, when turning to the touched portion, for example, the distance (angle) that the head 42 and the eyeball portion 76 move is shared based on the relationship between the current eye position and the touched position. The head 42 and the eyeball 76 may be displaced so as to direct the line of sight to the direction of the touched position. Specifically, the distance to the touched portion is calculated, and the head 42 is moved to half the distance, for example, and the eyeball 76 is moved to the other half. Such eye movements are also natural related to the movement of the head 42 and do not cause any discomfort to humans.
[0060]
The technique for turning the head 42 to the touched location is described in detail in Japanese Patent Application No. 2001-166018 filed by the present applicant.
[0061]
Further, in each of the above-described embodiments, in controlling the left and right eyeball units 76, the same angle data is given to the left and right eyeball motors so that both eyes are aligned and moved, but separate angle data is calculated respectively. The left and right eyeball portions 76 may be moved separately by giving the information. In some cases, the left and right eyeball portions 76 may be simultaneously moved by one eyeball motor.
[0062]
Further, in each of the above-described embodiments, two eyeballs 76 are provided on the left and right as in the case of a person. However, the eyeballs 76 only need to be visually recognizable as parts corresponding to the eyes. The number of the eyeball portions 76 is arbitrary and can be changed as appropriate. Similarly, the number of eye cameras 46 may be changed as appropriate.
[0063]
In each of the above-described embodiments, the position of an object such as a person or a poster is detected based on the image from the eye camera 46. However, if possible, the position of the object is detected using the image from the omnidirectional camera 26. May be detected.
[0064]
In each of the above-described embodiments, the position of the object is detected to control the displacement of the eyeball 76 or the head 12. However, for example, the direction of the object is detected, or a time difference of the image is generated. By detecting an object (e.g., an object that is likely to move) that is likely to move, an object is detected, and the displacement of the eyeball 76 or the head 12 is controlled so as to face the object. It may be.
[0065]
Further, in each of the above-described embodiments, the eyeball portion 76 is configured to include the eye camera 46. However, in some cases, the eyeball camera 46 is provided in a portion other than the eyeball portion 76 such as another portion of the head 42. You may. In this case, the eyeball 76 is not a visual organ because there is no camera, and becomes a part or dummy representing an apparent eye. However, the eyeball 76 is provided at a position corresponding to the eye and is the eye of the robot 10. Since it is configured to be visually recognized as such, there will be no particular problem for a person communicating with the robot 10.
[0066]
In each of the above-described embodiments, the eyes and face are merely turned to the object. However, in some cases, an operation such as pointing in the direction of the object using an arm portion or the like may be combined with an eye movement. . That is, by giving necessary angle data to motors of appropriate axes such as the right arm motor 62 and the left arm motor 64, for example, an introduction operation such as pointing the direction of an object such as a poster with the tip (hand) of the forearm. May be further combined.
[Brief description of the drawings]
FIG. 1 is a front view showing an entire configuration of an embodiment of the present invention.
FIG. 2 is an enlarged front view showing the head of the embodiment of FIG. 1;
FIG. 3 is a front view showing the displacement of the eyeball unit of the embodiment in FIG. 1;
FIG. 4 is a block diagram showing the embodiment of FIG. 1;
FIG. 5 is a flowchart showing a part of the operation in the embodiment in FIG. 1;
FIG. 6 is a flowchart showing another portion of the operation in the embodiment in FIG. 1;
[Explanation of symbols]
10. Communication robot
20 ... lower body
22 ... upper torso
40 ... neck joint
42 ... head
46, 46R, 46L ... eye camera
50 ... CPU
54… Memory
56… Motor control board
58… Sensor input / output board
60: Voice input / output board
66… Head motor
76, 76R, 76L ... Eyeball
78… Right eyeball motor
80… Left eyeball motor

Claims (7)

A communication robot with a function to communicate,
Eyeball,
A communication robot comprising: object detection means for detecting a target object; and eyeball displacement means for displacing the eyeball so as to face the object. A torso, a head attached to the torso via a neck joint, and a head provided with the eyeball portion, and further comprising head displacement means for controlling the neck joint to displace the head;
The communication robot according to claim 1, wherein the eyeball displacing unit displaces the eyeball in association with the movement of the head. The communication robot according to claim 2, wherein the eyeball displacing unit displaces the eyeball based on angle data proportional to the movement speed of the head. The communication robot according to claim 2, wherein the head displacing unit displaces the head such that a front side of the head is directed toward the object. The communication robot according to claim 1, wherein the object includes a human face. The communication robot according to claim 1, wherein the object detection unit includes a first detection unit that detects the object based on an image obtained from a camera. 7. The object detection device according to claim 1, wherein the object detection unit includes a second detection unit that detects the on-state touch sensor among the plurality of touch sensors provided at an arbitrary position of the communication robot as the object. 8. Communication robot described.

Images