JP2006281348A - Communication robot - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the affinity and safety by estimating the next position and touch behavior of the other party in tactile communication to make the other party feel easy. <P>SOLUTION: This communication robot 10 includes a plurality of piezo sensor sheets, that is, tactile sensors 58 dispersed and arranged on the whole body, and stores a tactile aspect graph data showing the probability transition of the output pattern of the tactile sensors 58 related to the position of the other party making touch behavior. Using the tactile aspect graph data, what is the probability transition of the position and touch behavior of the other party is obtained based upon the position of the other party detected from the obtained tactile sensor output vector, for example, photographing image data of an eye camera 50. The movement is controlled according to the transition destination information such as the obtained next position of the other party to avoid the movement to the concerned position and the movement of the arm, and the movement or turning to make the other party exist in the concerned next position is performed to introduce the other party to the estimated touch behavior. Accordingly, the next position and touch behavior of the other party is estimated to improve the safety and achieve the smooth communication. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a communication robot, and more particularly to, for example, a communication robot that includes a tactile sensor distributed throughout the body and predicts a partner's tactile behavior based on the tactile sense.

  The softness and sensitivity of the surface is important for robots working in everyday environments in terms of safety to humans who can communicate with each other. Furthermore, as the term “skinship” means that the surface of the body is the first interface to touch, so knowing how and where the communication partner is touching leads to knowing the partner's condition. For example, tactile behavior such as grasping firmly or stroking gently is an action to convey one's intention and feeling to the other party, and is one of fundamental communication actions. Therefore, the softness and tactile sensation of the whole body of the robot are very important for a communication robot that operates in an everyday environment.

In view of this, the present applicant has proposed a communication robot having a soft skin and a piezo sensor (tactile sensor or skin sensor) that outputs tactile information arranged in the whole body in Patent Document 1, for example. In the communication robot disclosed in Patent Document 1, it is understood that the way of touching a human being who is a communication partner can be distinguished by paying attention to pressure intensity, duration, frequency of change, or the like from output information of a tactile sensor. Yes.
JP 2004-283975 A

  In the prior art, how to touch at a short time interval (for example, about 2 or 3 seconds) is distinguished, for example, how the pressure changes when stroking. If the situation of the robot is limited to communication, the tactile sensation of the transition of human tactile behavior as the communication partner, such as where to locate and how to touch the next location when captured at long time intervals It is possible to limit to some extent based on

  Therefore, a main object of the present invention is to provide a communication robot capable of predicting the next position and tactile behavior of a partner in tactile communication.

  According to the first aspect of the present invention, a plurality of tactile sensors and tactile aspect graph data storing tactile aspect graph data indicating stochastic transitions of output patterns of the plurality of tactile sensors associated with the positions of communication partners performing tactile behavior are stored. Aspect storage means, acquisition means for acquiring output data of a plurality of tactile sensors, detection means for detecting the position of a partner when the output data is acquired by the acquisition means, output data of the plurality of tactile sensors acquired by the acquisition means, and Based on the detection result by the detection means, the transition destination information acquisition means for acquiring the transition destination information including at least the next position of the opponent from the tactile aspect storage means, and the transition destination information acquired by the transition destination information acquisition means It is a communication robot provided with the control means which controls operation | movement based on it.

  In the invention of claim 1, the communication robot is provided with a plurality of tactile sensors, for example, distributed throughout the body. The tactile aspect storage means stores tactile aspect graph data indicating the probabilistic transition of the output patterns of a plurality of tactile sensors associated with the positions of communication partners performing tactile actions. That is, the tactile aspect graph data indicates a probabilistic state transition in which the output patterns of the plurality of tactile sensors are in a state. For example, output patterns of a plurality of tactile sensors are stored as a plurality of clusters, and a cluster to be a transition destination and a transition probability thereof are stored for each of the plurality of clusters. A cluster shows a typical touch in tactile communication. In addition, information on the position of the opponent is stored in association with each of the plurality of clusters. For example, information related to the position of the opponent is stored as a probability distribution relating to at least the position of the fingertip and the waist of the opponent based on the position of the communication robot such as the center of gravity. That is, in the haptic aspect graph data in this case, the opponent's position indicates the position where the opponent is present (waist position) and the place where the opponent is touching (the position of the fingertip). Therefore, the tactile aspect storage means stores information indicating the position where the other party is present, the place where the opponent is touched, and how the touch at that time changes probabilistically. When the opponent performs a tactile action on the communication robot, the acquisition unit acquires output data of a plurality of tactile sensors. The detecting means detects the position of the other party when acquiring the output data of the tactile sensor. In the embodiment, the detection means includes an eye camera provided on, for example, the head of the robot, and detects the position of the other party based on image data captured by the camera. The transition destination information acquisition unit acquires transition destination information including at least the next position of the opponent from the tactile aspect storage unit based on the output data of the plurality of tactile sensors acquired by the acquisition unit and the detection result by the detection unit. . For example, based on output data and image data of a plurality of tactile sensors, the current cluster in the tactile aspect graph data, that is, the position of the current partner is specified. Then, transition destination information including at least the next position of the other party is acquired from the tactile aspect graph data. From the acquired information of the transition destination, the communication robot can know at least next how the opponent's position, touched place, touched state, etc. change stochastically, that is, at least the next Tactile behavior can be predicted. The control means controls the operation based on the acquired transition destination information. Accordingly, it is possible to predict the tactile behavior such as the next position of the opponent based on the tactile sense, and to control the operation in response to the prediction. For example, safety can be improved by avoiding danger, Smooth communication can be realized by guidance.

  The invention of claim 2 is dependent on the invention of claim 1, and the control means avoids the movement of the opponent to the next position or the movement of the part acquired by the transition destination information acquisition means.

  In the invention of claim 2, the control means avoids movement of the opponent to the next position or movement of the part. For example, the next position of the other party is predicted based on the acquired information of the transition destination, and control data that prohibits or restricts the movement to the predicted position or the movement of the part is generated to control the operation. Therefore, since it is possible to prevent a collision with the opponent or a part such as an arm from hitting the opponent, danger can be avoided and safety can be improved.

  The invention of claim 3 is dependent on the invention of claim 1, and the control means corresponds to the position of the opponent corresponding to the output data of the plurality of tactile sensors acquired by the acquisition means and the opponent acquired by the transition destination information acquisition means. The movement or turning is controlled based on the difference from the next position.

  In the invention of claim 3, the control means controls the movement or turning based on the difference between the position of the opponent corresponding to the output data of the plurality of tactile sensors, that is, the current position of the opponent and the next position of the opponent. . Accordingly, the communication robot can be moved or turned so that the opponent exists at the next position acquired by the transition destination information acquisition means. Therefore, since the opponent can be positioned at the predicted next position, it is expected that the opponent touches the touch location corresponding to the next position or performs the touch corresponding to the next position. The That is, the opponent can be guided to the predicted tactile behavior. Further, it is possible to lead to a further transition destination position or tactile behavior. Accordingly, since the other party can be guided to a desired position or tactile behavior in response to the prediction, smooth communication can be realized.

  The invention of claim 4 is dependent on any one of the inventions of claims 1 to 3, further comprising skin made of a flexible material placed on the robot body, wherein the plurality of tactile sensors are distributed in the skin. Includes piezo sensor sheet.

  In the invention of claim 4, for example, a skin made of a flexible material is placed on a robot body including a housing or the like, and a plurality of piezoelectric sensor sheets are dispersedly arranged in the skin as a plurality of tactile sensors. Therefore, for example, a communication robot having a whole body distribution type skin sensor can be realized, and tactile behavior is performed through soft skin, giving a sense of security to the other party and further improving affinity and safety. In addition to being able to achieve smoother tactile communication.

  According to the present invention, since the tactile aspect graph data indicating the stochastic transition of the tactile sensor output associated with the position of the opponent is stored, the next position or tactile behavior of the opponent is predicted based on the tactile sense. be able to. Therefore, by controlling the movement in response to the prediction, for example, it is possible to avoid danger and guide the opponent to the predicted position and tactile behavior, so that safety can be improved and smooth communication is realized. it can.

  The above object, 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.

  Referring to FIG. 1, a communication robot (hereinafter simply referred to as “robot”) 10 of this embodiment includes a carriage 12, and wheels 14 for autonomously moving the robot 10 are disposed on the side of the carriage 12. Is provided. The wheel 14 is driven by a wheel motor (indicated by reference numeral “16” in FIG. 4), and the carriage 12, that is, the robot 10 can be moved in any direction, front, back, left, and right. Although not shown, a collision sensor is attached to the front surface of the carriage 12, and the collision sensor detects contact of a person or other obstacle with the carriage 12.

  A polygonal columnar sensor mounting panel 18 is provided on the carriage 12, and an ultrasonic distance sensor 20 is mounted on each surface of the sensor mounting panel 18. In this embodiment, for example, 24 ultrasonic distance sensors 20 are provided so as to cover 360 degrees. The ultrasonic distance sensor 20 measures the distance from the sensor mounting panel 18, that is, the human body around the robot 10. Specifically, the ultrasonic distance sensor 20 emits an ultrasonic wave, measures the timing at which the ultrasonic wave is reflected from the person and is incident on the ultrasonic distance sensor 20, and outputs distance information between the person and the person. To do.

  On the carriage 12, the human body-like part 22 is attached so as to stand upright. The whole body of the human body 22 as the robot body is covered with skin 24 made of a flexible material, as will be described in detail later. The human body portion 22 includes a housing (not shown) such as an iron plate, for example, and accommodates a computer and other necessary components in the housing. Then, the skin 24 is put on the casing. A microphone 26 is provided at approximately the center of the upper part of the housing under the skin 24. The microphone 26 is for collecting ambient sounds, particularly human voices.

  The human body 22 includes a right arm 28R and a left arm 28L, and the right arm 28R and the left arm 28L, that is, the upper arms 30R and 30L are detachably attached to the trunk portion by shoulder joints 32R and 32L, respectively. The shoulder joints 32R and 32L have three axes of freedom. Forearms 36R and 36L are attached to upper arms 30R and 30L by uniaxial elbow joints 34R and 34L, and hands 38R and 38L are attached to these forearms 36R and 36L. Each axis in each joint of the right arm 28R and the left arm 28L is controlled by a motor (not shown). That is, the four motors of the right arm 28R and the left arm 28L are represented as the right arm motor 40 and the left arm motor 42, respectively, in FIG.

  A head 46 is attached to the upper portion of the human body 18 via a neck joint 44 so as to be able to be elevated and rotated like a human head. The three-axis neck joint 44 is controlled by a head motor 48 shown in FIG. Two eye cameras 50 are provided at positions corresponding to the “eyes” on the front surface of the head 46, and the eye camera 50 takes a picture of a human face approaching the robot 10 and other parts and outputs the video signal. take in. A speaker 52 is provided below the eye camera 50 in front of the head 46. The speaker 52 is used for the robot 10 to communicate by voice to the people around it.

  The torso, head 46 and arms of the human body 22 described above are all covered with the skin 24 made of a flexible material as described above. As shown in FIG. 2, the skin 24 includes a lower urethane foam 54 and a relatively thick silicon rubber layer 56a and a relatively thin silicon rubber layer 56b laminated thereon. A piezo sensor sheet (skin sensor) 58 is embedded between the two silicon rubber layers 56a and 56b. As this piezo sensor sheet 58, for example, a piezo film manufactured by MSI Inc. in the USA and sold by Tokyo Sensor Co., Ltd. is used (http://www.t-sensor.co.jp/PIEZO/TOP/index.html). The robot used in the example is a piezo sensor sheet obtained by cutting an A4 size (model number: 200 × 140 × 28) piezo film with scissors into 1/2, 1/3, 1/4, or 1/6 size. It is. This piezo film has a structure in which a metal thin film is formed on both surfaces of a piezoelectric film (for example, PVDF (polyvinylidene fluoride)), that is, a structure in which a piezoelectric body is sandwiched between conductors. When deformed by pressure or the like, piezoelectricity is generated between the metal thin films on both sides, that is, electric charges appear and a potential difference is generated.

  In the example, as described above, the softness of the skin 24 was obtained using foamed urethane and silicon rubber. If silicon rubber alone is used to obtain a certain degree of thickness and softness, it becomes too heavy and not only increases energy consumption but also weakens against laceration. Therefore, as a result of repeated experiments, the inventors decided to adopt a shape in which the rough shape and thickness are made of urethane foam and the surface is covered with about 20 mm of silicon rubber. Then, two silicon rubber layers were formed, and the above-described piezo sensor sheet 58 was embedded between the silicon rubber layers 56a and 56b. At this time, the silicon rubber layer 56a on the inner side was thickened (about 15 mm), and the silicon rubber layer 56b on the front side was thinned (about 5 mm). In this way, vibrations of the robot 10 and high-frequency vibrations generated when a person presses the surface can be cut, and the film can be easily deformed, so that pressure can be easily measured. In other words, the thickness of the silicon rubber layer depends on the structure and power of the robot 10, but it should be as thin as possible, but it is easy for deformation to be transmitted and vibration that causes noise is difficult to be transmitted. In addition, because it is possible to communicate with people through this soft skin through tactile behavior, it is possible to give a sense of security to people and increase their affinity, and when touching or hitting It can prevent human injury and increase safety.

  The material of the skin 24 may be a soft material, and is not limited to the above-described material, and may be another rubber material, for example. However, the surface metal thin film of the piezo film sheet needs to be made of a material that does not corrode. Further, the thickness of the skin 24 (the thickness of each layer) can be appropriately changed depending on the material.

  The above-described piezo sensor sheet, that is, a skin sensor (tactile sensor) 58 is embedded over the whole body of the human body-like portion 22, thereby detecting pressure applied to the skin 24 by contact with a human or the like as pressure (tactile) information. . In this embodiment, as shown in FIG. 3, 48 piezo sensor sheets 501-548 are embedded over the entire body of the robot 10. That is, it can be said that the robot 10 has a whole body distributed skin sensor. As for the embedding situation (place), a piezo sensor sheet is embedded in a part that is easily touched by humans, for example, the top of the head, shoulders, and arms (including hands) without gaps so that pressure can be detected accurately and reliably. A piezo sensor sheet was embedded with an acceptable gap in a part that is not supposed to be used, such as a foot or a flank. This solved the trade-off between detection accuracy and manufacturing cost. It should be noted that these 48 piezo sensor sheets 501-548 are sometimes shown without distinction by reference numeral 58 in some cases.

  The electrical configuration 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 CPU 60 for overall control. The CPU 60 is connected to a memory 64, a motor control board 66, a sensor input / output board 68, and a sound through a bus 62. An input / output board 70 is connected.

  Although not shown, the memory 64 includes a ROM, an HDD, and a RAM. A control program for the robot 10 is written in advance in the ROM and HDD. The control program includes, for example, a program for executing communication behavior and a program for communicating with an external computer. The memory 64 also stores data for executing a communication action. The data is, for example, voice data or voice data (voice synthesis data) to be generated from the speaker 52 when executing each action. And angle control data of each joint axis for presenting a predetermined gesture. The RAM is used as a temporary storage memory and a working memory.

  The motor control board 66 is constituted by a DSP (Digital Signal Processor), for example, and controls each axis motor such as each arm and head. That is, the motor control board 66 receives control data from the CPU 60, and controls three motors for controlling the angles of the three axes of the right shoulder joint 32R and one motor for controlling the angles of the one axis of the right elbow joint 34R. The rotation angles of the four motors (collectively shown as “right arm motor” in FIG. 4) 40 are adjusted. Further, the motor control board 66 adjusts the rotation angle of a total of four motors (collectively shown as “left arm motor” in FIG. 4) 42, three axes of the left shoulder joint 32L and one axis of the left elbow joint 34L. The motor control board 66 also adjusts the rotation angle of the triaxial motor 48 of the head 46 (collectively shown as “head motor” in FIG. 4) 48. The motor control board 66 controls two motors 16 that collectively drive the wheels 14 (collectively shown as “wheel motors” in FIG. 4).

  The above-described motors of this embodiment are stepping motors or pulse motors for simplifying the control except for the wheel motors 16, but may be direct-current motors similarly to the wheel motors 16.

  Similarly, the sensor input / output board 68 is configured by a DSP, and takes in signals from each sensor and camera and gives them to the CPU 60. That is, data related to contact from each of the collision sensors (not shown) is input to the CPU 60 through the sensor input / output board 68. Further, a video signal from the eye camera 50 is input to the CPU 60 after being subjected to predetermined processing by the sensor input / output board 68 as necessary.

  As shown in FIG. 5, the sensor input / output board 68 further includes a plurality (12 in the embodiment) of substrates 72, 72..., And each substrate 72 is provided with one PIC microcomputer 74. . The PIC microcomputer 74 is composed of, for example, an ASIC, and outputs voltage data (for example, 10 bits) from the A / D converter 76 similarly provided on the substrate 72 as a bit serial signal.

  As shown in FIG. 5, the skin sensor 58 has a piezo film 78 sandwiched between electrodes or conductors 80a and 80b. When pressure is applied to the piezo film 78, the piezo film 78 generates a voltage, and the voltage is applied to the two conductors 80a. And appear between 80b. However, since the voltage generated at this time has a high potential but the current is weak, it is difficult to capture the generated voltage as it is in the computer 60 (FIG. 4) because of a lot of noise. Therefore, in this embodiment, the substrate 72 shown in FIG. 5 is disposed at a position close to the skin sensor 58, and a high-impedance reading device, that is, an A / D converter 76 is disposed therein, and this A / D converter. The voltage value converted at 76 is read by the PIC microcomputer 74 and output as a serial signal, which is sent to the CPU 60. In addition, as an example of arrangement | positioning of the electrode of a piezo film sheet, the conductor 80a is arrange | positioned at the surface side of the skin 24, and the conductor 80b is arrange | positioned at the housing | casing side.

  As the A / D converter 76, a 4-channel 10-bit converter is used in the embodiment. Therefore, one substrate 72 can handle the four skin sensors 58. The substrate 72 is provided with four pairs of terminals 82a and 82b for four piezo sensor sheets, to which electrodes 80a and 80b are connected, respectively. A noise removing capacitor 84 is connected between the terminals 82a and 82b. Accordingly, the voltage from the skin sensor 58 applied between the terminals 82a and 82b is subjected to noise removal, and then subjected to current amplification by the operational amplifier 86 and is input to one channel of the A / D converter 76 described above. Although only one skin sensor 58 is shown in FIG. 5, the other skin sensors 58 and associated circuitry are similarly configured.

  As described above, 48 piezo sensor sheets 501 to 548 are embedded in the skin 24 of the human body 22 throughout the whole body. However, when all of them are read by the CPU or computer 60 for controlling the robot, noise is generated. Not only is it easy to pick up, but it also requires a large number of A / D ports of the computer, which is not realistic. Therefore, as described above, the readers (substrate 72, A / D converter 76) are arranged in the vicinity of the skin sensor 58, and each output is connected by one serial cable, for example, RS232C (trade name). A so-called daisy chain was formed. Therefore, the bit serial signal output from the PIC microcomputer 74 of one board 72 shown in FIG. 5 is given to the serial input port of the PIC microcomputer 74 of the board 72 at the next stage. The next-stage PIC microcomputer 74 adds the data read from the A / D converter 76 that it is in charge of to the data sent from the previous-stage PIC microcomputer 74 and outputs the result as a bit serial signal. Therefore, the computer 60 can take in the detection information from the skin sensor 58 of the whole body with one serial port.

  Since the detection data output from each PIC microcomputer 74 includes an identifier indicating which of the 48 piezo sensor sheets 501 to 548 shown in FIG. 3 and information on the pressure value, the computer 60 It is possible to easily specify which (part) of the piezo sensor sheet is subjected to how much pressure.

  However, in this embodiment, for each of the five piezo sensor sheets 535-539 and 544-548 provided at the left and right hands, there are four inputs to the A / D converter 76. The output of one piece (right hand: 539, left hand: 548) and one outside (right hand: 536, left hand: 545) are arranged in parallel. Therefore, in this embodiment, there are substantially four skin sensors 58 on the left and right hands, so that the output data of the skin sensor 58 is 46-dimensional data.

  When reading the output, specifically, the computer 60 polls the PIC microcomputer 74 at the final stage that outputs the bit serial data at a cycle of 50 msec, for example, and detects all the piezo sensor sheets 501 to 548 at a cycle of 50 msec. Data can be read. The detection data is output from the A / D converter 76 (FIG. 5) in, for example, 32 stages of positive and negative, for a total of 64 stages. That is, the lower 4 bits of the 10 bits are discarded as noise components, and only the upper 6 bits of data are output from each PIC microcomputer 74 (FIG. 5).

  Then, the computer 60 uses the 64 levels of data detected by the skin sensor 58, for example, touch such as strength of touch, duration of pressed state or frequency of voltage change waveform (frequency of pressure change). The state can be measured. Depending on the strength of the touch, it can be judged, for example, whether it was struck badly, whether it was lightly struck, whether it was gently touched, or whether it was touched lightly. It is possible to determine the continuation state such as “whether it has been struck”, “whether it has been pressed”, etc., and depending on the frequency of the pressure change, for example, “whether it is struck”, “whether it is being stroked”, “whether it is tickled” The type of touching can be determined. And the robot 10 can control operation | movement according to how to touch (contact state). Such operation control is also disclosed in detail in Patent Document 1 (Japanese Patent Application Laid-Open No. 2004-283975) by the present applicant, so please refer to it.

  Returning to FIG. 4, the synthesized voice data is given from the CPU 60 to the speaker 52 via the sound input / output board 70, and the voice or voice according to the data is outputted from the speaker 52 accordingly. . Also, the voice input from the microphone 26 is taken into the CPU 60 via the sound input / output board 70.

  Further, a communication LAN board 88 and a wireless communication device 90 are connected to the CPU 60 via the bus 62. The communication LAN board 88 is configured by a DSP and gives transmission data sent from the CPU 60 to the wireless communication apparatus 90. The transmission data from the wireless communication apparatus 90 is omitted from illustration, but a network such as a wireless LAN or the Internet is not shown. Via an external computer. The communication LAN board 88 receives data via the wireless communication device 90 and gives the received data to the CPU 60. That is, the communication LAN board 88 and the wireless communication device 90 allow the robot 10 to perform wireless communication with an external computer or the like.

  Further, the CPU 60 is connected to a haptic aspect graph database (DB) 92 via a bus 62. However, this database may be provided so as to be accessible on an external computer or an external network.

  The tactile aspect graph DB 110 stores data indicating the probabilistic state transition (tactile aspect graph data) with the tactile sensor output associated with the position and orientation of the partner during tactile communication as a state. The aspect graph is used as a description of object (pattern) recognition, and shows how the appearance of the object (generally based on camera image information) changes (transitions). The output of the tactile sensor is stored in association with the position and posture of the opponent at the time of tactile communication. Therefore, the robot 10 views the position of the opponent through the tactile sensor 58. Therefore, the probabilistic transition of the tactile sensor output indicating how the other party looks through the tactile sense is expressed as a tactile aspect graph using the term aspect graph. Data that associates the output of the tactile sensor 58 with the position / posture of the communication partner at the time of tactile communication is referred to as a map.

  For example, the robot 10 measures the output data of the skin sensor 58 during communication, and uses the map data included in the tactile aspect graph data, so that the position where the partner who is performing the tactile action is currently located. Can grasp what kind of posture it is. Humans obtain various information through skin sensations in daily interaction and skinship communication. For example, even if they are hugged from behind, they can grasp how they are hugged. With this map data, the robot 10 can recognize what kind of tactile action the other party is taking by tactile sense, even if it is not seen with the eyes, like a human being.

  Further, the robot 10 can grasp what kind of tactile sensor output is to be probabilistically changed next based on the measured output data of the tactile sensor 58 and tactile aspect graph data. That is, the robot 10 can predict how to be touched next based on the tactile sensor output (cluster) of the transition destination. Further, since the tactile sensor output is associated with the position and posture of the opponent, the robot 10 can predict the next position and posture of the communication partner. Specifically, in this embodiment, the position and posture of the opponent are grasped from the probability distribution regarding the three-dimensional positions of the opponent's fingertips and waist. Therefore, the robot 10 can predict where the opponent's waist is located and where the fingertip touches in the next tactile behavior.

  Thus, the robot 10 can predict the next action (touch action) of the opponent. In addition, since the tactile aspect graph data indicates the probabilistic transition of the tactile sensor output as described above, not only the transition to the previous one but also the transition to the further destination is grasped based on the tactile aspect graph data. It is possible. Therefore, the robot 10 can also predict the near future behavior (the opponent's position, touched location, touched manner, etc.) such as the next behavior of the opponent.

  When creating the tactile aspect graph data, first, a map in which the output of the tactile sensor 58 is associated with the position / posture of the communication partner during tactile communication is created.

  Touching actions for communication purposes include handshake and hug, but typical actions are thought to have many typical actions because the meaning of the action is transmitted to anyone. Therefore, by combining the skin sensor 58 of the robot 10 and the three-dimensional motion measurement system, the output of the skin sensor 58 during communication between the robot 10 and the human and the position / posture of the human measured by the three-dimensional motion measurement system. Mapping is performed.

  For example, a creation system 94 as shown in FIG. 6 is used. The creation system 94 includes a creation computer 96, and a plurality of cameras 98 are connected to the creation computer 96. The creation computer 96 includes a three-dimensional motion measurement data DB 102, a skin sensor data DB 104, a map DB 92, and a haptic aspect graph DB 110 inside or outside the computer 96.

  The creation system 94 has a function as a motion capture system. In this embodiment, for example, an optical motion capture system of VICON (http://www.vicon.com/) is applied. The motion capture system is not limited to an optical system, and various known systems can be applied.

  For example, a plurality of infrared reflection markers are attached to the robot 10 and the human 100, and a camera with an infrared strobe is applied as the camera 98. The plurality of cameras 98 are installed in different directions with respect to the robot 10 and the human 100, and in principle, at least three cameras may be provided. In this embodiment, for example, twelve cameras 98 are installed in the environment.

  As shown in FIG. 7, the infrared reflective marker includes both the robot 10 and the human 100 at four positions on the head, six positions on the arm × 2 (left and right), one position × 2 (left and right) on the fingertip of the index finger, and the front of the torso It is attached to a total of 23 locations, two at the part and three at the rear of the fuselage. Since it is desired to obtain the position of the other party during communication in the coordinate system with reference to the robot 10, a marker is also attached to the robot 10 in order to determine the origin and orientation of the robot 10. Although the origin of the robot 10 may be set arbitrarily, it is set at the center (center of gravity) position, for example. Although the coordinate axis direction is also arbitrary, for example, both shoulder directions are set to the Y axis, the depth direction is set to the X axis, and the vertical direction is set to the Z axis.

  In this embodiment, in order to grasp where the human 100 is touching and in what kind of posture, a total of three markers, the fingertip and the waist (rear torso), are used for map creation. From the position of the fingertip, it is grasped where the human 100 is touching. The posture is determined from the relative positional relationship between any position of the upper body and the fingertip. Among them, the position of the waist has the advantage that the posture is easy to understand by being difficult to see from the camera 98 and knowing whether the opponent is standing or squatting because the legs are out of the waist, etc. Was adopted for correspondence. If other positions such as heads and shoulders are also used in the association map, the position and posture of the opponent 100 can be grasped in more detail.

  Then, in order to collect data, an experiment is conducted in which communication is performed between the robot 10 and the opponent 100 in the environment of the camera 98. Specifically, communication with tactile behavior such as handshake and hug is performed. The robot 10 acts autonomously based on the control program and data registered in the memory 64 and the acquired various sensor data.

  FIG. 8 shows an example of map creation processing operation in the creation system 94. First, in step S1, the creation computer 96 accumulates data during communication through experiments. The acquired data is the output data of the skin sensor 58 and the fingertip and waist position data of the partner 100 by three-dimensional motion measurement. Specifically, the three-dimensional motion data during communication is measured at, for example, 60 Hz (60 frames per second), and the skin sensor data is measured at 20 Hz (20 frames per second). In order to synchronize the skin sensor data and the position data, at the start of the experiment, a human 100 hits the top of the robot 10 and the like, and measures the data at that time so that the time axes of both data are aligned. .

  The operation at the time of communication is imaged by a plurality of cameras 98, and the creation computer 96 acquires time-series image data from each camera 98. By the image processing of the acquired image data for each frame, two-dimensional coordinate data of each marker (including the fingertip and the waist) in each image data is extracted. Then, time-series three-dimensional coordinate data of each marker is calculated from the two-dimensional coordinate data by the principle of triangulation. The coordinates of the fingertip marker are expressed in two types: a body coordinate system that is an orthogonal coordinate system in which the origin is fixed to the body of the robot 10 and a head coordinate system that is an orthogonal coordinate system in which the origin is fixed to the head of the robot 10. Is done. The coordinates of the waist marker are expressed in the body coordinate system. The time-series three-dimensional coordinate data of each marker calculated in this way is stored in the three-dimensional motion meter side data DB 102. When the motion measurement data is viewed for each frame, the three-dimensional positions of the fingertips and waist for each frame are known, so that the posture of the opponent 100 can be grasped, and where the opponent 100 is located. You can see where you are touching. Further, when the motion measurement data is viewed in time series, the change in the positions of the fingertips and the waist can be understood, so that it is possible to grasp what kind of tactile behavior is being performed.

  Further, time-series output data of the skin sensor 58 is taken in from the robot 10 to the creation computer 96 via the wireless LAN, for example, or after completion of the experiment, and stored in the skin sensor data DB 104. Skin sensor data is 46-dimensional data including outputs from 48 skin sensors 501-548 as described above, and each element has a value of 0 to 32 (absolute value of −31 to 32).

  Subsequently, in step S3, a frame whose skin sensor output is greater than or equal to a threshold is selected. The threshold value is experimentally obtained for each of the 48 skin sensors 501 to 548 and set in advance. It is judged whether any one of the 48 skin sensors 501 to 548 exceeds the threshold value, and if there is one exceeding the threshold value, the frame is adopted for map creation. To do.

  In step S5, clustering is performed by the ISODATA method based on the skin sensor data using the selected frame data. That is, in the 46-dimensional skin sensor data space, a portion (cluster) in which the distribution of patterns is dense is found, or similar patterns in each pattern are grouped together. For example, first, for each selected frame, a 46-dimensional vector is created in which the sensor exceeding the threshold is “1” and the other sensors are “0”. Among these vectors, those with less than 5 frames having the same pattern are discarded as noise. For each element of the remaining pattern, when the value is “1”, the value of 1.5 times the threshold value of the sensor is used, and when the value is “0”, the value of 0.5 times the threshold value of the sensor is used. The vector used was generated and used as the initial cluster nucleus of ISODATA. The ISODATA method is a method of clustering, and the cluster generated in the previous stage is divided or merged according to a certain standard to derive a final cluster (Reference: Junichiro Toriwaki, Television Society Textbook Series 9 “Cognitive Engineering— Pattern recognition and its application- ", Corona, 1993). The derived cluster represents a typical touch in communication.

  Subsequently, in step S7, fingertip data processing is executed. The processing of this fingertip data is shown in detail in FIG. In the first step S21 in FIG. 9, the spatial density distribution in each coordinate system is obtained for the coordinates of the fingertip (head / torso coordinate system) paired with the skin sensor data in each cluster. That is, the measurement time corresponding to the skin sensor data in each cluster or the fingertip coordinate data of the frame is processed. For example, the coordinate system is divided into 50 × 50 × 50 [mm] voxels, and the number of fingertip coordinates in each voxel is counted. Finally, the density is obtained by dividing the number of elements in the voxel by the number of elements in the cluster.

  Next, in step S23, the maximum value of the density distribution is compared between the coordinate systems, and the larger coordinate system is adopted. That is, it is determined whether the head or the torso is touched. For example, at the time of communication, the robot 10 moves the neck joint and performs body expression such as crawling and tilting. However, when the fingertip of the human 100 touches the head of the robot 10 at that time, the position of the fingertip depends on the movement of the head. It changes together. The position of the fingertip does not change if it is expressed in the head coordinate system, but if it is expressed in the torso coordinate system, it changes even though the same place of the robot 10 is touched. Therefore, if the coordinates of the fingertip are spread over a wide range in the coordinate system, the coordinate system is inappropriate. If the coordinates of the fingertip are in a certain range in the coordinate system, the coordinate system is appropriate. It is. As for the maximum value of the density distribution, it can be seen that the head is touched when the head coordinate system is larger, and the body is touched when the body coordinate system is larger.

  In step S25, a cluster whose density peak value in the adopted coordinate system exceeds the threshold is taken into the map as an effective cluster. The density peak value is a density when the vicinity of 26 of the voxel is collected, and is a sum of the densities near 26. The threshold value of the peak value is obtained experimentally and is set to 0.6, for example. As a result of selection based on this threshold value, a stable position or posture that can be estimated is adopted as a cluster. For example, if the other hand is hanging while communicating with one hand, the posture cannot be estimated and is discarded. By this step S25, the probability distribution of the coordinates of the fingertip paired with the skin sensor output data in the effective cluster is stored as a map in association with the effective cluster. When step S25 ends, the process returns to step S9 in FIG.

  Returning to FIG. 8, waist data processing is executed in step S9. The waist data processing is shown in detail in FIG. In the first step S31 of FIG. 10, the spatial density distribution is obtained for the coordinates of the waist (body coordinate system) paired with the skin sensor data in each cluster. Similar to the above fingertip data processing, for the waist coordinate data of the frame corresponding to the skin sensor data in each cluster, count how many waist coordinates are in each voxel, and then voxel by the number of elements in the cluster Divide the number of inner elements. In step S33, a cluster whose density peak value exceeds the threshold value is taken into the map as a valid cluster. By this step S33, the probability distribution of the coordinates of the hips paired with the skin sensor output data in the effective cluster is stored in association with the effective cluster as a map. When step S33 ends, the waist data processing ends, and the map creation processing ends.

  In addition, when the marker positions of parts other than the fingertips and the waist, such as shoulders and heads, are used for map creation, data processing at each position may be executed in the same manner.

  In this way, map data in which the pattern (cluster) of the skin sensor output is associated with the position / posture of the other party is created and stored in the map DB 92 of FIG. An overview of the map is shown in FIG. For example, a cluster whose cluster representative value is a 46-dimensional skin sensor vector A is associated with a probability distribution A related to the three-dimensional position of the fingertip and / or waist of the communication partner 100. That is, when the robot 10 is touching the skin sensor vector A, the probability distribution A represents the representative fingertip position and posture (waist position) of the opponent 100 corresponding to the touch method. The cluster representative value is the center value, and the variance indicates the spread of the cluster.

  FIG. 12 shows an example of cluster representative values. That is, it is a 46-dimensional skin sensor vector and represents one way of touching. An example of the three-dimensional position (average of distribution) of the fingertip position 106 and the waist position 108 of the opponent 100 corresponding to the cluster of FIG. 12 is shown in FIG. In FIG. 13, the same position distribution is displayed from two different viewpoints. It can be grasped that the fingertip of the opponent 100 touches both shoulders of the robot 10 and the waist is above the front of the robot 10. That is, it can be seen that the opponent 100 in this case stands in front of the robot 10 and both hands touch the shoulders of the robot 10.

  The creation computer 96 creates tactile aspect graph data based on the map created in this manner that associates the output pattern of the skin sensor with the position or posture of the partner 100 performing the tactile action.

  FIG. 14 shows an example of the operation of the tactile aspect graph creation process of the creation computer 96. First, in step S41, the time-series skin sensor data stored in the skin sensor data DB 104 is converted into the transition data of the cluster of the skin sensor output stored in the map DB 92. Specifically, it is determined to which cluster the output data (skin sensor output vector) at each detection time (frame) belongs. For example, the distance between the skin sensor output vector and each cluster is calculated, and the nearest cluster is detected. Then, the transition data of the cluster is created by associating the skin sensor data at each detection time with the identification information of the belonging cluster.

  Next, in step S43, the transition destination cluster is detected for each cluster based on the cluster transition data, and the number of transitions is counted for each transition destination cluster. Subsequently, in step S45, for each cluster, for example, the transition probability for each transition destination cluster is calculated by dividing the number of transitions of each transition destination cluster by the total number of transitions.

  In step S47, the transition destination cluster (cluster identification information) and its transition probability value are stored in association with the cluster (cluster identification information).

  The data including the transition destination cluster and the transition probability associated with the cluster and the map data of the map DB 92 created in this way are stored in the haptic aspect graph DB 110 as haptic aspect graph data. FIG. 15 shows an example of the contents of the haptic aspect graph data. As shown in FIG. 15A, for each cluster number, a transition destination cluster number and its transition probability value are stored. When there are a plurality of transition destinations, data including a plurality of transition destination cluster numbers and transition probabilities is stored. Further, as shown in FIG. 15B, data indicating the association between the cluster number and the partner position data number is stored. The cluster number is identification information of the cluster, and the partner position data number is identification information of the partner position data, that is, data indicating a probability distribution regarding the three-dimensional positions of the fingertip and waist of the communication partner. Further, as shown in FIG. 15C, data related to the tactile sensor output (cluster) is stored for each cluster number. Further, as shown in FIG. 15D, data on the opponent position is stored for each opponent position data number. At the time of communication, the cluster number corresponding to the measured tactile sensor output vector is specified, and the transition destination cluster number and its transition probability can be acquired by referring to the data in FIG. Then, by referring to the data of FIG. 15B, the partner position data number corresponding to the cluster number of the transition destination can be specified, so that the next partner position data can be acquired. Therefore, the next partner position, that is, the next action can be predicted based on the next partner position data and the transition probability.

  The data shown in FIGS. 15B, 15C, and 15D corresponds to map data in which the tactile sensor output (cluster) is associated with the position and orientation of the communication partner.

  FIG. 16 shows a conceptual diagram of a haptic aspect graph. As described above, the tactile aspect graph shows the stochastic state transition with the tactile sensor output as the state. The tactile sensor output (cluster) is associated with the position of the communication partner. In this embodiment, the position of the opponent is a probability distribution of the three-dimensional positions of the opponent's fingertips and waist. The position of the waist corresponds to the position of the opponent viewed from the robot 10, and the position of the fingertip corresponds to the place touched. That is, the tactile aspect graph shows not only how to be touched but also how the position of the other party and the place to touch change probabilistically. Therefore, the tactile aspect graph can be used to predict where the opponent will be located next (or in the near future), where to touch, and how to touch.

  Each node in FIG. 16 indicates the position of the opponent and the touched location associated with the tactile sensor output (cluster). In the upper right node, a graph of the cluster representative value is shown as the tactile sensor output (cluster), and the position of the other party corresponding to the cluster and the touched location are shown. In the other nodes, only the position of the opponent and the place being touched are shown, and the corresponding tactile sensor output is omitted. The place where the black sphere is touched (the position of the fingertip) indicates the position of the opponent (the position of the waist). A line segment extending downward from the ellipsoid indicates the height of the opponent's waist.

  In the example of FIG. 16, the upper right node indicates that the opponent stands on the right side of the robot and touches the right shoulder of the robot. The line between nodes shows the connection of each node, and each node makes a transition to a connected node stochastically. The upper left node shows a state where the opponent stands on the right side of the robot and touches the top of the robot. The middle right node shows that the opponent is standing in front of the robot and touching the head of the robot. The middle left node shows a state where the opponent stands in front of the robot and touches the top of the robot. The lower right node shows a state where the opponent is squatting in front of the robot and touching the chest of the robot. For example, as the next action after standing on the right side of the robot and touching the robot's right shoulder (the upper right node), the opponent touches the top of the robot in the same position (the upper left node), or It can be predicted that the robot moves forward and touches the head of the robot (the middle node on the right).

  FIG. 17 shows an example of the prediction processing operation of the robot 10. The robot 10 starts the prediction process when touched by the opponent during communication. In the first step S61 in FIG. 17, when the output data from the skin sensor 58 is detected, the CPU 60 of the robot 10 acquires a skin sensor output vector. That is, pressures detected by all skin sensors 58 are detected as pressure value data via the sensor input / output board 68, and a 46-dimensional skin sensor output vector is generated and stored in the memory 64.

  Next, in step S63, the cluster closest to the acquired skin sensor vector is selected from the clusters stored in the haptic aspect graph DB 110. For example, the distance between the current skin sensor output vector and each cluster is calculated from each cluster representative value and the variance of the elements in each cluster.

  Subsequently, in step S65, image data is acquired in the memory 64 by directing the eye camera 50 to a position corresponding to the selected cluster. For example, the other party position data corresponding to the selected cluster number is acquired from the tactile aspect graph data in the memory 64, and the angle control of the head motor 48 that enables the eye camera 50 to photograph the position based on the other party position data. Data is generated and the rotation angle of the head motor 48 is controlled via the motor control board 66. Then, the image data from the eye camera 50 is acquired in the memory 64 via the sensor input / output board 68.

  In the subsequent step S67, an attempt is made to detect a partner at a position corresponding to the acquired partner position data based on the image data. In step S69, it is determined whether or not a partner is detected at the position. If “NO” in the step S69, that is, if the opponent is not photographed at the position, the opponent does not exist at the position corresponding to the tactile sensor output. In this case, since the cluster could not be specified, the process returns to step S61.

  On the other hand, if “YES” in the step S69, that is, if the opponent is photographed at the position, the opponent position data number acquired in the step S65 is stored in the memory 64 as information indicating the current position in the step S71. To remember. In subsequent step S73, the transition destination information of the selected cluster is acquired from the tactile aspect graph data in the memory 64. For example, the cluster number and probability value of the transition destination corresponding to the selected cluster number are acquired, the data of the cluster corresponding to the cluster number of the transition destination is acquired, and the partner position data corresponding to the transition destination cluster number The number is specified, and the partner position data corresponding to the partner position data number is acquired.

  In step S75, it is determined whether further information is necessary. For example, when it is set to predict and guide a future action further than the next action of the opponent, it is determined that the information ahead is necessary. If “YES” in the step S75, the transition destination information of the transition destination cluster acquired in the step S73 is acquired in the memory 64 from the tactile aspect graph data in a step S77. As a result, necessary information (transition destination cluster number, transition probability, transition destination cluster data, corresponding partner position data, etc.) ahead of the next action, that is, in the near future is acquired.

  If “NO” in the step S75, or when the step S77 is ended, a prediction handling process is executed in a step S79, and the operation is controlled based on the transition destination information acquired in the step S73 or S77. When step S79 ends, the prediction process of FIG. 17 ends. An example of the operation of the prediction handling process in step S79 is shown in FIG.

  In first step S91 in FIG. 18, it is determined whether or not to avoid danger. For example, it is determined whether or not risk avoidance based on prediction is set in the program or data of the tactile communication behavior currently being executed.

  If “YES” in the step S91, the next position or action of the opponent is predicted in a step S93 based on the acquired transition destination information. For example, the partner position data having the highest probability value is predicted as the next partner position and stored in the memory 64 as the predicted position. Alternatively, all possible partner position data acquired as a transition destination may be predicted as the next partner position and stored as a predicted position in the memory 64 so as to consider all possibilities. For example, when the opponent position data of the transition destination includes the right side, the front side, and the left side of the robot, all of the right side, front side, and left side of the robot may be predicted as the positions of the next opponent. Note that the weight may be considered according to the probability value.

  In step S95, prohibition of movement toward the predicted position is set. For example, a flag prohibiting movement is set to ON, and the position of the opponent's waist among the predicted positions is stored as a movement prohibited position. In step S97, prohibition of movement of a part such as an arm to the predicted position is set. For example, the flag prohibiting the movement of the arm is set to ON, and from the predicted position, the position of the opponent's fingertip and waist, the area between the fingertip and the waist, and the like are stored as the operation prohibited position. When step S97 ends, the prediction handling process ends.

  On the other hand, if “NO” in the step S91, it is determined whether or not to guide in a step S99. For example, it is determined whether or not guidance based on prediction is set in the program or data of the tactile communication behavior currently being executed.

  If “YES” in the step S99, the position or action of the other party to be guided is selected from the acquired transition destination information in a step S101. For example, in the example of FIG. 16, when the current partner position is the upper right node and when predicting up to two nodes ahead, the node to be guided from the node other than the upper right is selected by the cluster number or the partner position data number Is done. For example, when it is desired to guide the partner to touch the chest, the partner position data number corresponding to the lower right node is selected.

  Subsequently, it is determined whether or not the selection in step S103 is guidance to the destination of the transition destination. If “YES” in the step S103, an opponent position on the near side of the opponent position selected in the step S101 is selected in a step S105. When it is desired to guide further to the transition destination, it is necessary to change the path on the near side in order to lead to the position to be guided, and therefore, the partner position corresponding to the one ahead from the current node is selected. For example, when the partner position data number corresponding to the lower right node is selected in step S101, the partner position data number corresponding to the middle right node that is the node closest to the current node on the near side is selected. Etc. are selected.

  If step S105 ends or if “NO” in the step S103, in a step S107, the opponent moves or turns so that the opponent exists at the selected position. Specifically, based on the difference between the partner position data corresponding to the selected partner position data number and the partner position data corresponding to the current position, the wheel motor 16 is set so that the partner exists at the selected position. Change the control data. Based on the changed control data, the robot 10 turns or moves, and as a result, the opponent is located at a position selected as viewed from the robot 10.

  For example, as a continuation of the above example, when the current opponent position is the right side of the robot 10 (the upper right node in FIG. 16) and the selected position is the front side of the robot 10 (the middle right node in FIG. 16). For example, by calculating the difference (angle difference) between the direction from the robot 10 to the current position and the direction to the selected position and turning the robot 10 by the angle difference, the opponent is positioned in front of the robot 10. Can do. In this case, as the next action, it is expected that the opponent touches the head of the robot, which is the touch location corresponding to the position, with the corresponding touch method, as in the middle right node. Therefore, the other party's tactile behavior can be induced in response to the prediction. If it is detected by checking the skin sensor output vector or the image data that the opponent touches the lower neck of the robot 10, the lower left node of the transition destination, that is, the opponent squats in front and touches the chest. It becomes possible to lead to the state. If the step S107 is ended or if “NO” in the step S99, the prediction corresponding process is ended.

  FIG. 19 shows an example of the operation of the change process when the communication action is executed. When the prohibition of movement or arm movement is set to avoid danger in the prediction handling process of FIG. 18, the control data is changed by this change process. In the first step S121 of FIG. 19, it is determined whether or not movement prohibition is set based on a movement prohibition flag or the like. If “YES” in the step S121, the prohibited position set in the step S95 is acquired in a step S123, and whether or not the control data of the communication action to be executed includes a movement to the prohibited position side in a step S125. Judging. If “YES” in the step S125, control data changed so as not to move to the prohibited position side is generated in a step S127. For example, when the forbidden position is the front and the control data includes a forward movement, the portion related to the forward movement of the control data of the wheel motor 16 is cleared. In this way, the control is executed based on the changed control data, so that the movement to the prohibited position is not performed, so that the collision with the opponent can be prevented and the danger can be avoided.

  If step S127 ends or if “NO” in step S125, or if “NO” in step S121, prohibition of arm movement is set based on the operation prohibition flag or the like in step S129. Judge whether or not. If “YES” in the step S129, the prohibition position set in the step S97 is acquired in a step S131, and whether or not the control data of the communication action to be executed includes the movement of the arm to the prohibition position in a step S133. Determine whether. If “YES” in the step S133, control data changed so as not to move the arm to the prohibited position is generated in a step S135. For example, when the forbidden position is on the right side and the control data includes a right-hand motion, the control data for the right arm motor 40 is cleared and the current angle is maintained. Further, for example, when the prohibition position is on the right side and the control data includes the movement of the right arm, and the movement of the right arm can be replaced with the left arm, for example, the action of shaking the hand with the right arm, You may make it produce | generate the control data changed so that the said operation | movement may be performed with a left arm. If step S135 ends, or if “NO” in step S133, or if “NO” in step S129, this change processing ends. In this way, since the control is executed based on the changed control data, the movement of the part such as the arm to the prohibited position is not performed, so it is possible to prevent the part such as the arm from hitting the opponent beforehand. And avoid danger.

  According to this embodiment, the tactile aspect graph data indicating the stochastic transition of the tactile sensor output associated with the position of the opponent (for example, the waist and the fingertip) is stored. Since it is possible to acquire information on a partner who is predicted to change the tactile behavior, such as a position, a touched place, and a touched manner, the next position and tactile behavior of the partner can be predicted. And by controlling operation | movement according to prediction, since danger, such as a collision with the other party, can be avoided, safety | security can be improved. In addition, since the opponent can be guided to the predicted position or tactile behavior based on the difference between the acquired information of the transition destination and the current position of the other party, smooth communication can be realized.

  If attention is paid to the personal characteristics of the transition of tactile behavior, the robot 10 can also perform personal recognition using tactile aspect graph data. In this case, the haptic aspect graph data for each of a plurality of communication partners is stored in the haptic aspect graph DB 110 in association with the partner ID, for example. During communication, by measuring the output vector of the tactile sensor and detecting the position of the other party from the image data, the position of the other party, the place where the person is touching, and the way of touching (cluster) are identified and the history is stored. To record the transition of tactile behavior. Then, it is possible to specify who the communication partner is by determining which partner ID corresponds to the tactile aspect graph data of the partner ID.

It is an illustration figure which shows the communication robot of one Example of this invention. It is an illustration figure which shows the skin used for the communication robot of FIG. 1 Example, and the piezo sensor sheet | seat embedded in it. It is an illustration figure which shows the arrangement position of a piezo sensor sheet | seat. It is a block diagram which shows the electrical structure of the communication robot of FIG. 1 Example. It is an illustration figure which shows partially the sensor input / output board which inputs a detection signal from the piezo sensor sheet | seat in the communication robot of FIG. 1 Example. It is an illustration figure which shows the outline | summary of the map creation system for creating the tactile aspect graph with which the communication robot of FIG. 1 embodiment is equipped. It is an illustration figure which shows the attachment position of the infrared reflective marker at the time of map creation. It is a flowchart which shows an example of operation | movement of the map creation process of the computer for creation of FIG. It is a flowchart which shows the operation | movement of the processing of the fingertip data of FIG. It is a flowchart which shows the operation | movement of the process of waist data of FIG. It is an illustration figure which shows the outline | summary of the map data memorize | stored in map DB. It is an illustration figure which shows an example of a cluster representative value. It is an illustration figure which shows the outline | summary of distribution of the other party's fingertip and waist position corresponding to the cluster representative value of FIG. It is a flowchart which shows an example of operation | movement of the tactile aspect graph preparation process of the computer for preparation of FIG. It is an illustration figure which shows an example of the content of the data memorize | stored in tactile aspect graph DB, FIG.15 (A) shows the data including the transition destination cluster number for every cluster number, and its transition probability, FIG.15 (B) is FIG. FIG. 15C shows data indicating the tactile sensor output (cluster) for each cluster number, and FIG. 15D shows data for each partner position number data. Indicates partner position data. It is an illustration figure which shows the concept of a tactile aspect graph. It is a flowchart which shows an example of the operation | movement in the prediction process of the communication robot of FIG. 1 Example. It is a flowchart which shows an example of the operation | movement of the prediction corresponding | compatible process of FIG. It is a flowchart which shows an example of the operation | movement in the change process at the time of communication action execution of the communication robot of FIG. 1 Example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Communication robot 22 ... Human body-like part 24 ... Skin 50 ... Eye camera 58,501-548 ... Tactile sensor (piezo sensor sheet)
60 ... CPU
64 ... Memory 66 ... Motor control board 68 ... Sensor input / output board 70 ... Sound input / output board 92 ... Map DB
94 ... Creation system 96 ... Creation computer 110 ... Tactile aspect graph DB

Claims (4)

Multiple tactile sensors,
Tactile aspect storage means for storing tactile aspect graph data indicating probabilistic transitions of patterns of outputs of a plurality of tactile sensors associated with positions of communication partners performing tactile behavior;
Obtaining means for obtaining output data of the plurality of tactile sensors;
Detecting means for detecting the position of the opponent when the output data is acquired by the acquiring means;
Based on the output data of the plurality of tactile sensors acquired by the acquiring unit and the detection result by the detecting unit, the transition destination that acquires at least the next position of the opponent from the tactile aspect storage unit A communication robot comprising: an information acquisition unit; and a control unit that controls an operation based on the transition destination information acquired by the transition destination information acquisition unit.   The communication robot according to claim 1, wherein the control unit avoids movement of the opponent to a next position or movement of a part acquired by the transition destination information acquisition unit.   The control means is based on a difference between the position of the opponent corresponding to the output data of the plurality of tactile sensors acquired by the acquisition means and the next position of the opponent acquired by the transition destination information acquisition means. The communication robot according to claim 1, wherein the communication robot controls movement or turning. It is further equipped with a skin made of a flexible material to be put on the robot body,
The communication robot according to any one of claims 1 to 3, wherein the plurality of tactile sensors include a plurality of piezo sensor sheets distributed in the skin.

Images