JP3942586B2 - Communication robot - Google Patents

Description

  The present invention relates to a communication robot, and more particularly to a communication robot that communicates with a human by using body movements such as gestures and voices.

This type of conventional communication robot is disclosed in, for example, Patent Documents 1 and 2 by the present applicant. This communication robot is an interaction-oriented robot that attempts to communicate with humans through voice and actions, and is equipped with various sensors for recognizing humans who are conversation partners and obstacles. In this prior art, for example, an ultrasonic distance sensor is provided as a proximity sensor on a panel on the undercarriage, and the approach is detected by measuring the distance to a person or the like. In addition, touch sensors (contact sensors) are provided on parts such as the shoulder, upper arm, and forearm to detect contact with a person or the like.
JP 2002-357883 A [B25J13 / 00] JP 2002-361484 A [B25J13 / 08]

  However, the space on the robot body is limited, and a plurality of types of sensors must be provided in the limited space. Therefore, the proximity sensor and the contact sensor can only be provided at different specific locations. For this reason, in the above-described prior art, it is possible to detect the approach of the person or the object itself to the robot, but for example, it is difficult to detect in advance which part of the robot the person or the like is trying to contact. Further, even if a part such as an arm that is moving during an action is likely to hit a person, it is difficult to avoid it because it is impossible to detect the part that is likely to hit.

  For example, if a proximity sensor is provided in a place where a person may touch, it is possible to detect the approach to this specific place, but a contact sensor cannot be provided in the same place. The contact state cannot be detected.

  On the other hand, in the above-mentioned prior art, a very sensitive contact sensor is used. However, since the contact sensor is provided only at a specific location and the information is binary of on / off, the contact location is determined. For example, we could not acquire tactile information that conveys intentions and feelings by tactile sensation, such as being grasped strongly or stroked gently. I could not. In order to cope with this tactile communication, the present applicant proposed in Japanese Patent Application No. 2003-80106, filed on March 24, 2003, a communication robot in which a pressure sensor sheet is arranged, for example, on the whole body instead of a touch sensor. ing. However, since the proximity sensor and the pressure sensor are provided separately in this robot, there is the same problem as described above.

  Therefore, in the prior art, in the tactile communication, only a posteriori and limited action of always detecting a contact and dealing with it can be executed.

  Moreover, it is preferable from the above-mentioned space aspect and cost aspect to detect a series of movements in the tactile behavior of people approaching and touching using different types of sensors provided at different locations. It is not a thing.

  Therefore, a main object of the present invention is to provide a communication robot that can detect the approach of a person or an object and can detect the subsequent contact state.

  Another object of the present invention is to provide a communication robot that can control the operation in a predictive manner in accordance with the approach and contact state of a person or an object.

  Another object of the present invention is to provide a communication robot that can reduce the types of sensors.

The invention according to claim 1 is a piezoelectric sensor provided with variable sensitivity, a proximity sensor that changes the sensitivity of the piezoelectric sensor to measure capacitance, or a changing means that causes the piezoelectric sensor to function as a pressure sensor that measures pressure, and a proximity sensor Or it is a communication robot provided with the operation control means which controls operation | movement according to the detection information detected as a pressure sensor.

  According to the first aspect of the present invention, the communication robot includes a piezoelectric sensor, and the changing means changes the sensitivity of the piezoelectric sensor to function as a proximity sensor that measures capacitance or a pressure sensor that measures pressure. That is, in this communication robot, the piezoelectric sensor can be used as two types of sensors, ie, a proximity sensor or a pressure sensor, by dynamically changing the sensitivity of the piezoelectric sensor according to, for example, the situation by the changing means. The operation control means controls the operation in accordance with detection information detected as a proximity sensor or a pressure sensor. For example, by making this piezoelectric sensor function as a proximity sensor, it is possible to detect the proximity state of a person or an object. Actions that can be adapted and actions that avoid contact can be performed. Further, after the approach is detected, when the sensitivity of the piezoelectric sensor is changed by the changing means to function as a pressure sensor, it is possible to detect the contact state at the same location or another location where the approach is detected. Therefore, the operation | movement according to the contact state can be performed.

  The invention according to claim 2 is the communication robot according to claim 1, wherein the operation control means detects the approach or the contact when the approach or the contact is detected based on the detection information detected as the proximity sensor. Make it perform a different action than when it is not.

  In the invention of claim 2, when the approach or contact of a person or an object is detected by the piezoelectric sensor functioning as a proximity sensor, the operation control means performs an operation different from that when the approach or contact is not detected. Let Therefore, when an approach is detected, it is possible to adapt to contact or avoid contact, for example, by predicting the subsequent contact and performing an action different from the previous action. Similarly, when it is detected, it is possible to execute an appropriate operation according to the contact.

  According to a third aspect of the present invention, there is provided a communication robot according to the second aspect, wherein the operation control means reduces the operation speed when an approach is detected based on detection information detected as a proximity sensor.

  According to a third aspect of the present invention, the operation control means reduces the operation speed when an approach of a person or an object is detected. Therefore, for example, it is possible to delay contact with the person and avoid contact, or to stay long in the vicinity of the approaching person and give the other party a chance to communicate with the robot Can do.

  According to a fourth aspect of the present invention, there is provided a communication robot according to the second or third aspect, wherein the motion control means changes the motion trajectory when an approach is detected based on detection information detected as a proximity sensor.

  In the invention of claim 4, the motion control means changes the motion trajectory when an approach of a person or an object is detected. Therefore, for example, it is possible to avoid contact by changing the trajectory in a direction away from an approaching person or object.

  The invention according to claim 5 is the communication robot according to any one of claims 1 to 4, wherein the changing means changes the sensitivity when an approach or a contact is detected based on detection information detected as a proximity sensor. Then, the piezoelectric sensor functions as a pressure sensor, and the operation control means controls the operation according to the detection information detected as the pressure sensor.

  In the fifth aspect of the invention, when the approach or contact is detected by the piezoelectric sensor functioning as a proximity sensor, the changing means changes the sensitivity and causes the piezoelectric sensor to function as a pressure sensor. Then, the operation control means controls the operation according to the detection information detected as the pressure sensor. Therefore, for example, the function of the piezoelectric sensor is changed from a proximity sensor to a pressure sensor in accordance with the approach or contact of a person or the like, so that the subsequent contact state can be detected and an appropriate operation corresponding to the contact state is performed. Can be executed.

  A sixth aspect of the present invention is a communication robot according to any one of the first to fifth aspects, wherein the piezoelectric sensor is arranged in a portion that moves when the gesture is expressed.

  According to the sixth aspect of the present invention, the piezoelectric sensor is arranged in a portion that moves when expressing by gesture. Therefore, for example, if the piezoelectric sensor functions as a proximity sensor by the changing means, it is possible to detect that the moving part has approached a person or the like while performing the gesture, and thus contact with the opponent. Can be avoided. Further, if the sensitivity is changed so that the piezoelectric sensor functions as a pressure sensor, the contact state with respect to that portion can also be detected.

  A seventh aspect of the present invention is a communication robot according to any one of the first to sixth aspects, wherein the piezoelectric sensor is disposed at least on any of the head, torso, shoulder, arm or hand.

  In the invention of claim 7, since the sensor is disposed at least on any of the head, torso, shoulder, arm or hand, it is possible to detect the approach and contact state of a person or the like with respect to the disposed portion. Moreover, the approached part and the contacted part can be grasped | ascertained, and the operation | movement according to it can be performed.

  The invention according to claim 8 is the communication robot according to any one of claims 1 to 7, wherein the piezoelectric sensors are distributedly arranged.

  In the invention of claim 8, since the piezoelectric sensors are arranged in a distributed manner, it is possible to detect the approach and contact state of a person or the like with respect to each arranged part. For example, when distributed throughout the body, proximity states from various directions to the whole body can be detected, and contact states to various parts of the whole body can be detected. Then, it is possible to grasp the approached part or the contacted part and execute an operation corresponding thereto. For example, it is possible to estimate the size of the other party from the difference in height of the portion where the piezoelectric sensor that has detected the approach or the like is arranged, and to perform an operation corresponding to the size.

  A ninth aspect of the present invention is a communication robot according to any one of the first to eighth aspects, further comprising skin made of a flexible material placed on the robot body, and the piezoelectric sensor is disposed in the skin.

  According to the ninth aspect of the present invention, for example, a soft skin is put on a robot main body including a casing and the piezoelectric sensor is disposed in the skin. Therefore, for example, it is possible to detect the state of approach and contact in the tactile behavior via the soft skin, and communication can be performed accordingly. In addition, it is possible to give the partner a sense of security and increase the affinity, and also improve the safety at the time of contact.

  According to the present invention, the piezoelectric sensor can be caused to function as two types of sensors, ie, a proximity sensor and a pressure sensor, by dynamically changing the sensitivity thereof, so that the approach of a person or the like is detected by this one type of sensor. It is possible to detect the state of contact after contact. Therefore, it is possible to predict a contact and perform an operation according to the prediction, and after the contact, an operation according to the state of the contact can be performed. New communication behavior that has been In addition, the types of sensors can be reduced, which can contribute to space and cost.

  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.

  With reference to FIG. 1, a communication robot (hereinafter, simply referred to as “robot”) 10 of this embodiment includes a carriage 12, and wheels on which the robot 10 autonomously moves are disposed on a side surface of the carriage 12. 14 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.

  On the carriage 12, the human body-like portion 18 is attached so as to stand upright. The whole body of the human body 18 as the robot body is covered with skin 20 made of a flexible material, as will be described in detail later. The human body 18 includes a housing (not shown) such as an iron plate, for example, and houses a computer and other necessary components in the housing. Then, the skin 20 is put on the casing. A microphone 22 is provided at approximately the center of the upper part of the housing under the skin 20. The microphone 22 is for collecting ambient sounds, particularly human voices.

  The human body portion 18 includes a right arm 24R and a left arm 24L, and the right arm 24R and the left arm 24L, that is, the upper arms 26R and 26L are detachably attached to the trunk portion by shoulder joints 28R and 28L, respectively. The shoulder joints 28R and 28L have three degrees of freedom. Forearms 32R and 32L are attached to upper arms 26R and 26L by uniaxial elbow joints 30R and 30L, and hands 34R and 34L are attached to these forearms 32R and 32L. Each axis in each joint of the right arm 24R and the left arm 24L is controlled by a motor (not shown). That is, the four motors of the right arm 24R and the left arm 24L are represented as the right arm motor 36 and the left arm motor 38, respectively, in FIG.

  A head 42 is attached to the upper portion of the human body 18 via a neck joint 40 so as to be able to be elevated and rotated in the same manner as a human head. The three-axis neck joint 40 is controlled by a head motor 44 shown in FIG. Two eye cameras 46 are provided at positions corresponding to the “eyes” on the front surface of the head 42, and these eye cameras 46 photograph a human face and other parts approaching the robot 10 and output the video signals thereof. take in. A speaker 48 is provided below the eye camera 46 in front of the head 42. The speaker 48 is used for the robot 10 to communicate by voice to the people around it.

  The body, head 42 and arms of the human body 18 described above are all covered with the skin 20 made of a flexible material as described above. As shown in FIG. 2, the skin 20 includes a lower urethane foam 50 and a relatively thick silicon rubber layer 52a and a relatively thin silicon rubber layer 52b laminated thereon. A piezo sensor sheet 54 is embedded between the two silicon rubber layers 52a and 52b. As this piezo sensor sheet 54, for example, a piezo film manufactured by MSI Inc. in the US 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 addition, when a charged object approaches one side of the metal thin film, charges are collected on the one thin film side, so that the capacitance can be measured if the potential difference at this time can be read. That is, the piezo sensor sheet 54 can function as two types of sensors, that is, a pressure sensor and a proximity sensor, by controlling sensitivity as described later.

  In the example, as described above, the softness of the skin 20 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 54 was embedded between the silicon rubber layers 52a and 52b. At this time, the silicon rubber layer 52a on the inner side was thickened (about 15 mm), and the silicon rubber layer 52b 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, it is possible to give a sense of security to the other party and enhance the affinity, and when people touch or collide It can prevent injury and increase safety.

  The material of the skin 20 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 20 (the thickness of each layer) can be appropriately changed depending on the material.

  The above-described piezo sensor sheet, that is, the skin sensor 54 is embedded over the whole body of the human body-like portion 18, thereby detecting pressure applied to the skin 20 by contact with a human or the like as pressure sense information. Also, with this piezo sensor sheet 54, the capacitance can be measured when the detection sensitivity is increased, so that the change in capacitance due to the approach of a human or the like is measured in a non-contact manner to detect the proximity state. You can also. Therefore, in this embodiment, as shown in FIG. 3, 48 piezo sensor sheets 501-548 are embedded over the entire body of the robot 10. 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 in an unacceptable gap, such as on the back, legs, or 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 54 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 56 for overall control. The CPU 56 is connected to a memory 60, a motor control board 62, a sensor input / output board 64, and a sound through a bus 58. An input / output board 66 is connected.

  Although not shown, the memory 60 includes a ROM, an HDD, and a RAM. The ROM and HDD are pre-stored with a control program for the robot 10, and voice or voice data to be generated from the speaker 48, predetermined data The angle control data of each joint axis for performing the gesture is stored. The RAM is used as a temporary storage memory and a working memory.

  The motor control board 62 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 62 receives control data from the CPU 56, and controls three motors for controlling the angles of the three axes of the right shoulder joint 28R and one motor for controlling the angles of the one axis of the right elbow joint 30R. The rotation angles of the four motors (collectively shown as “right arm motor” in FIG. 4) 36 are adjusted. Further, the motor control board 62 adjusts the rotation angle of a total of four motors (collectively shown as “left arm motor” in FIG. 4) 38, three axes of the left shoulder joint 28L and one axis of the left elbow joint 30L. The motor control board 62 also adjusts the rotation angle of the three-axis motor 44 of the head 42 (collectively shown as “head motor” in FIG. 4). The motor control board 62 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 64 is configured by a DSP, and takes in signals from each sensor and camera and gives them to the CPU 56. That is, data relating to contact from each of the collision sensors (not shown) is input to the CPU 56 through the sensor input / output board 64. The video signal from the eye camera 46 is input to the CPU 56 after being subjected to predetermined processing by the sensor input / output board 64 as necessary.

  As shown in FIG. 5, the sensor input / output board 64 further includes a plurality of (in the embodiment, twelve) boards 68, 68..., And each board 68 is provided with one PIC microcomputer 70. . The PIC microcomputer 70 is composed of, for example, an ASIC, and outputs voltage data (for example, 10 bits) from the A / D converter 72 similarly provided on the substrate 68 as a bit serial signal.

  As shown in FIG. 5, the piezo sensor sheet 54 is obtained by sandwiching a piezo film 74 between electrodes or conductors 76a and 76b. When a pressure is applied, the piezo film 74 generates a voltage, and the voltage is divided into two conductors. Appears between 76a and 76b. 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 into the computer 56 (FIG. 4) because of a lot of noise. Therefore, in this embodiment, the substrate 68 shown in FIG. 5 is disposed at a position close to the piezo sensor sheet 54, and a high impedance reader, that is, an A / D converter 72 is disposed therein, and this A / D conversion is performed. The voltage value converted by the device 72 is read by the PIC microcomputer 70 and output as a serial signal, which is sent to the CPU 56. In addition, as an example of arrangement | positioning of the electrode of a piezo film sheet, the conductor 76a is arrange | positioned at the surface side of the skin 20, and the conductor 76b is arrange | positioned at the housing | casing side.

  As the A / D converter 72, a 4-channel 10-bit converter is used in the embodiment. Accordingly, one substrate 68 can handle the four piezoelectric sensor sheets 54. The substrate 68 is provided with four pairs of terminals 78a and 78b for the four piezo sensor sheets 54, and electrodes 76a and 76b are connected to the respective pairs. A noise removing capacitor 80 and a resistor R are connected between the terminals 78a and 78b. One end of the capacitor 80 and the resistor R on the terminal 78a side is connected to one (+) side input of the operational amplifier 82, and the other end of the capacitor 80 and the resistor R on the terminal 78b side is grounded. The (−) side input of the operational amplifier 82 is grounded via the resistor R1. The feedback circuit connecting the (−) side input of the operational amplifier 82 and the output side of the operational amplifier 82 includes a circuit including a plurality of resistors R21, R22,... R2n and a multiplexer 84 in order to make the resistance variable. Provided.

  In this substrate 68, the voltage from the piezo sensor sheet 54 applied between the terminals 78a and 78b is subjected to noise removal, and then current amplified by the operational amplifier 82 and input to one channel of the A / D converter 72 described above. Is done. Further, the amplification factor of the operational amplifier 82 is dynamically adjusted by the resistors R21 to R2n, the multiplexer 84, and the like, and thus the sensitivity of the piezo sensor sheet 54 is dynamically varied. Specifically, the multiplexer 84 is controlled by the computer 56 directly or by the PIC microcomputer 70 that receives an instruction from the computer 56 to control which of the resistors R21 to R2n is connected to the circuit, and thereby the piezo sensor sheet 54. The sensitivity of is changed. The piezo sensor sheet 54 functions as a pressure sensor that measures pressure when the sensitivity is normal, and functions as a non-contact proximity sensor that measures capacitance when the sensitivity is increased.

  In FIG. 5, only one piezo sensor sheet 54 is shown for the sake of clarity, but the other piezo sensor sheets 54 and circuits related thereto are configured in the same manner. Further, the plurality of resistors R21 to R2n and the multiplexer 84 are examples, and the variable resistor circuit for adjusting the amplification factor of the operational amplifier 82 can be changed as appropriate.

  The inventors have confirmed that the resistance value of the feedback circuit sufficient for measuring the capacitance is, for example, about 330 kΩ (amplification factor is about 330 times). That is, when the resistance R1 is 1 kΩ and the resistance of the feedback circuit is 330 kΩ, the piezo sensor sheet 54 can be used as a proximity sensor. Therefore, as an example, if the range of the variable resistor provided in the feedback circuit changes from about 0Ω to about 500 kΩ, a sufficient function can be exhibited. When this variable resistor is discretely configured as a plurality of R2n as shown in FIG. 5, for example, these R2n may be set to 0Ω, 10 kΩ, 50 kΩ, 100 kΩ, and 500 kΩ as an example. In this case, it functions as a proximity sensor when 500 kΩ is selected, and functions as a pressure sensor when another resistance is selected.

  As described above, 48 piezo sensor sheets 54 are embedded throughout the body 20 in the skin 20 of the human body-like portion 18. However, if all of them are read by the CPU or computer 56 for robot control, it is easy to pick up noise. In addition, it requires a large number of computer A / D ports, which is not realistic. Therefore, as described above, the reading device (substrate 68, A / D converter 72) is distributed in the vicinity of the piezo sensor sheet 54, and each output is connected by one serial cable, for example, RS232C (trade name). Formed a so-called daisy chain. Therefore, the bit serial signal output from the PIC microcomputer 70 of one board 68 shown in FIG. 5 is given to the serial input port of the PIC microcomputer 70 of the board 68 at the next stage. The next-stage PIC microcomputer 70 adds the data read from the A / D converter 72 in charge to the data sent from the previous-stage PIC microcomputer 70 and outputs the result as a bit serial signal. Therefore, the computer 56 can capture the detection information from the piezo sensor sheet 54 of the whole body with one serial port.

  Since the detection data output from each PIC microcomputer 70 includes an identifier indicating which of the 48 piezo sensor sheets 501 to 548 shown in FIG. 3 and information on the pressure value or capacitance, The computer 56 can easily specify how much pressure (which part) of the piezo sensor sheet 54 is subjected to, or can easily specify which (part) piezo sensor sheet 54 has been approached. .

  Specifically, the computer 56 polls the last stage PIC microcomputer 70 that outputs bit serial data at a cycle of 50 msec, for example, and reads the detection data of all the piezo sensor sheets 501 to 548 at a cycle of 50 msec. it can. The detection data is output from the A / D converter 72 (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 70 (FIG. 5).

  Then, when the piezo sensor sheet 54 functions as a pressure sensor, the computer 56 uses, for example, 64 steps of data detected as the pressure sensor, for example, how to touch and how long it is pressed. Alternatively, the contact state such as the frequency of the voltage change waveform (pressure change frequency) can be measured. Depending on the strength of the touch, it can be judged whether it was struck by a heavy hit, lightly struck, a gentle hand placed, a light touch, etc. 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 has been struck” The type of touching can be determined. Such control of the operation according to the contact state is disclosed in detail in Japanese Patent Application No. 2003-80106 filed on March 24, 2003 by the applicant of the present application. Also in this pressure sensor mode, the contact state can be detected in more detail by appropriately changing the sensitivity according to the behavior pattern of the robot 10, for example.

  Further, when the sensitivity of the piezo sensor sheet 54 is changed to a high sensitivity so as to function as a proximity sensor, the computer 56 uses the data detected as the proximity sensor, that is, changes in the detected data (electrostatic capacitance). By capturing the change in capacity, it is possible to measure the approaching state of a person or an object, and to detect that there has been contact.

  Returning to FIG. 4, the synthesized voice data is given to the speaker 48 from the CPU 56 via the sound input / output board 66, and accordingly, the voice or voice according to the data is outputted from the speaker 48. . Further, the voice input from the microphone 22 is taken into the CPU 56 via the sound input / output board 66.

  The robot 10 of this embodiment dynamically uses the skin sensor 54 as a proximity sensor or a pressure sensor by dynamically changing the sensitivity of the skin sensor (piezoelectric sensor) 54 in accordance with the behavior pattern and situation of the robot 10. Based on the detected data, it is possible to execute a communication action corresponding to the situation.

  For example, by first increasing the sensitivity of the piezo sensor sheet 54, the piezo sensor sheet 54 is caused to function as a non-contact type proximity sensor that measures capacitance. When the skin sensor 54 reacts and detects the approach or contact of a person or an object, the robot 10 is different from the operation when no approach or contact is detected according to the approach or contact. A predetermined operation is executed.

  The fact that there is a response to the skin sensor 54 in the high sensitivity state means that there is a person or an object nearby. For example, if it is moving, a subsequent contact is predicted. In this embodiment, as described above, a plurality of piezo sensor sheets 54 are distributed throughout the body, such as the head, torso, shoulders, arms or hands, so that proximity states from various directions to the whole body can be detected. . So, for example, by slowing down, you can delay getting closer, ensuring safety, and staying closer to the other person makes it easier to talk to the other person Can give an opportunity to communicate. Furthermore, when there is a reaction in the traveling direction, contact can be avoided and safety can be improved by changing the moving direction.

  Further, for example, when waiting, there is a person or moving object nearby and the person is trying to come into contact with the person, so the operation of looking at the place or direction where the user feels the sign is natural for the person. Further, when the opponent suddenly touches the skin sensor 54, the response of the skin sensor 54 swings out and shows a maximum value or exceeds a predetermined threshold value, so that the robot 10 shows a natural surprise as a human reaction, and then the skin sensor 54 You can start communication with lower sensitivity. In addition, for a partner who is in contact with the scary robot 10 and does not readily touch it and cannot grasp the opportunity of communication, the user can talk to the person that it is safe to touch, for example, and can promote communication.

  Further, during the communication action, the robot 10 and the person basically communicate with each other at a short distance, and the opportunity for the robot 10 and the person to collide increases. In particular, when expressing with gestures (gestures) using the body, in this embodiment, the arms 24R and 24L are frequently moved. However, when only the touch sensor is provided as in the prior art, it is difficult to understand until it hits the opponent. There wasn't. That is, in the prior art, for example, the robot pushes the hand toward the opponent during a handshake action, but at this time, if the distance from the opponent cannot be measured well, the robot bumps into the middle. Similarly, in the case of a by-by action (waving a hand while speaking “bye-bye”), the user will also collide when waving. In addition, in the case of hugging behavior (if you spread your arms and hug your loved one when the other person approaches you), you will collide with someone who is beside you when you spread your arms.

  However, if there is a proximity sensor at the same location as the pressure sensor in this moving part as in this embodiment, that is, if the pressure sensor functions as a proximity sensor, the risk of collision before collision is detected. Can do. Therefore, if the sensitivity of the part (arm in this embodiment) that is frequently moved during the communication action is increased, when the skin sensor 54 of the part reacts, the operation speed of the part is reduced or the action is performed. The danger of contact can be avoided by changing the trajectory. Then, for example, it is possible to return the operation after confirming that the reaction is lost and the safety is ensured when operating in the avoidance path.

  Further, when the robot 10 functions as a proximity sensor, the robot 10 functions as a pressure sensor by reducing the sensitivity of the skin sensor 54 when detecting the approach or contact of a person or an object. Then, it is possible to measure a state in which a person or an object has actually contacted thereafter, and perform a predetermined operation according to the contact state.

  For example, in the case of a handshake action, when the skin sensor 54 reacts, it is possible to grasp the place where the opponent is about to grip before actually shaking, so the handshake action is adapted to the action of the opponent. be able to. In addition, when there is a reaction in the skin sensor 54 arranged on a portion other than the arm that has been pushed out, it can be grasped in advance that the opponent is not responding to the handshake, and accordingly, an operation corresponding to that can be executed. In this way, it is possible to grasp the part where the approach or contact is detected, and execute an appropriate operation according to the part. Further, by lowering the sensitivity of the skin sensor 54 at the place where the reaction occurred and functioning as a pressure sensor to measure the strength of the pressure, it is possible to know how much the opponent is gripping during the handshake. It is possible to detect whether or not he is trying to release his hand, and the handshake operation can be terminated in accordance with the operation of the opponent.

  In addition, for example, in the case of a cuddle action, by making the skin sensors 54 distributed throughout the body highly sensitive, based on the difference in the placement part (height) of the skin sensor 54 without using vision, The size of the opponent can be estimated before touching. That is, the robot 10 of this embodiment has a body height of about 100 cm, for example, but when the opponent is an adult or the like, the opponent tries to hold the arm so that the arm contacts the shoulder portion of the robot 10. Is expected. On the other hand, when the other party is a child or the like, the other party is expected to turn his arm around the robot 10 and hold it. Therefore, for example, when there is a response to the skin sensor 54 around the shoulder, the opponent is large. On the other hand, when there is no response to the skin sensor 54 around the shoulder, for example, there is a reaction to the skin sensor 54 around the lower abdomen. You can see that the opponent is small. And the operation | movement adapted to the difference of the part with this reaction can be performed. For example, if the opponent is large, you will hug it so that it will be sweet, and if the opponent is small, you will gently hold it without applying force. Can be executed. And during this hugging operation, you can know how much you are hugging by measuring the contact state such as pressure intensity and duration with a pressure sensor, so that you can hold or touch gently You can hug.

  According to this embodiment, by increasing the sensitivity of the piezo sensor sheet 54, it can be used as a proximity sensor to adaptively operate or avoid danger according to the approach of a person or an object during communication. The operation can be controlled predictively. Furthermore, in the situation where it is actually touched, by using the pressure sensor with reduced sensitivity, pressure sensing can be performed at the same place where the approach has occurred, and an operation according to the contact state can be executed. In this manner, new tactile communication that is impossible with the conventional technology in which the proximity sensor and the contact sensor are separately provided can be realized. In addition, since the piezo sensor sheet 54 can function as two types of sensors, a proximity sensor and a pressure sensor, depending on the sensitivity, the types of sensors can be reduced, contributing to space and cost.

  Some specific operations are described below with reference to corresponding flow diagrams. However, it should be pointed out in advance that each operation is merely an example.

  The flowchart shown in FIG. 6 shows an example of the operation when the robot 10 executes “patrol behavior” (pretend to move around and guard).

  In the first step S1 of FIG. 6, the CPU 56 instructs the sensor input / output board 64 to change the sensitivity of all the skin sensors 54 to high sensitivity and enter the proximity sensor mode.

  In this embodiment, when the action is started in this way, the skin sensor 54 is changed to high sensitivity so as to function as a proximity sensor. However, the skin sensor 54 is set to high sensitivity as an initial state. It may be set to function as a proximity sensor.

  Next, in step S3, the program and data for executing this “patrol action” are read from the memory 60 to start the “patrol action”, and in step S5, the wheel motor 16 is controlled via the motor control board 62. The robot 10 is controlled and moved at random.

  In step S7, it is determined whether any skin sensor 54 has a reaction. That is, the detection data is acquired from the sensor input / output board 64, and it is determined whether or not the data indicating that the approach is detected is input. The process of step S7 is repeatedly executed, for example, at a constant period in parallel with the operation of the patrol action.

  If “YES” in this step S7, the wheel motor 16 is controlled in a succeeding step S9 to lower the moving speed. As a result, it is possible to give an opportunity for communication to an approaching partner.

  Further, in step S11, it is determined whether or not there is a reaction in the traveling direction, that is, whether or not data indicating that an approach has been detected is input from the skin sensor 54 disposed on the traveling direction side. If “NO” in this step S11, it is understood that the person or thing detected in the step S7 does not exist in the traveling direction, and the process directly returns to the step S7. On the other hand, if “YES” in the step S11, since a person or an object exists in the traveling direction, the moving direction is changed to another direction by controlling the wheel motor 16 to avoid a collision in the subsequent step S13. Then, the process returns to step S7.

  The flowchart shown in FIG. 7 is an example of an operation when the robot 10 executes “standby action (single play)” (playing alone by putting a head or arms together while waiting for communication). Indicates.

  In the first step S21 in FIG. 7, the CPU 56 instructs the sensor input / output board 64 to increase the sensitivity of all the skin sensors 54 and set the proximity sensor mode. Next, in step S23, the program and data for executing the “standby action” are read from the memory 60, and the “standby action” is started.

  In step S25, it is determined whether any skin sensor 54 has a reaction. That is, for example, detection data is acquired from the sensor input / output board 64 at regular intervals, and it is determined whether data indicating that an approach or contact has been detected is input. If “YES” in the step S25, it is determined whether or not the reaction has been shaken in a succeeding step S27. That is, it is determined whether the detection data shows a maximum value or exceeds a predetermined threshold value. Here, “YES” means that there is a contact.

  If “NO” in the step S27, the part of the piezo sensor sheet 54 that is likely to be touched is specified, and the head motor 44 is controlled to perform the operation of viewing the part. As a result, it is possible to present a natural action of grasping a place where a person is going to touch and looking at the place or direction before being touched.

  In a succeeding step S31, it is determined whether or not a state of being touched even if there is a reaction is repeated a predetermined number of times or more. If “NO”, the process returns to the step S25. On the other hand, if “YES” in the step S31, it seems that the other party does not readily touch the robot 10 and is terribly responding. Therefore, in the subsequent step S33, the audio data is transferred from the memory 60 to the sound input / output board 66. Then, a synthesized voice “It is okay to touch” is output from the speaker 48. As a result, communication can be promoted to the other party who is hesitant. When step S33 ends, the process returns to step S25.

  If “YES” in the step S27, that is, if the opponent is suddenly touched, a surprising operation is executed by controlling the right arm motor 36, the left arm motor 38, or the head motor 44 in a step S35. , Present a surprising gesture. In step S37, the sensor input / output board 64 is instructed to lower the sensitivity of the skin sensor 54 in the touched portion. The part touched in this way functions as a pressure sensor so that the contact state can be detected, and then communication with the partner is started.

  The flowchart shown in FIG. 8 shows an example of an operation when risk avoidance at the time of communication action is executed. In the first step S51 in FIG. 8, the CPU 56 instructs the sensor input / output board 64 to increase the sensitivity of the skin sensors 531 to 539 and 540 to 548 in the portions of the arms 24R and 24L to be moved and move the arms. Only the portions 24R and 24L are set to the proximity sensor mode. Next, in step S53, a program and data for executing a predetermined communication action are read from the memory 60, and the communication action is started.

  In step S55, whether or not there is a reaction in the skin sensors 54 of the arms 24R and 24L in the proximity sensor mode is determined at regular intervals, for example.

  If “YES” in this step S55, that is, if the detection data from the skin sensor 54 of the moving arms 24R and 24L indicates that an approach has been detected, in the subsequent step S57, the right arm The motor 36 and the left arm motor 38 are controlled to reduce the operating speed of the moving arms 24R and 24L. In subsequent step S59, the right arm motor 36 and the left arm motor 38 are controlled to change the trajectories of the arms 24R and 24L to obstacle avoiding trajectories. The obstacle avoidance trajectory is, for example, close to the piezo sensor sheets 531 to 539 and 540 to 548 arranged in the moving arms 24R and 24L in such a direction that the reaction becomes small, that is, away from the obstacle. It may be registered in advance in the memory 60 as control data for operating the arm 24R or 24L in such a direction that the degree of the movement becomes smaller. In this case, control data corresponding to the piezo sensor sheet 54 that has reacted is given from the memory 60 to the motor control board 62 to control the right arm motor 36, the left arm motor 38, etc. The trajectory can be easily taken.

  Subsequently, in step S61, whether or not there is a reaction in the skin sensors 54 of the arms 24R and 24L during the action accompanied by the avoidance is determined at regular intervals, for example. If “YES” in this step S61, that is, if the detection data acquired from the sensor input / output board 64 indicates that an approach or contact has been detected, the arm is detected in a succeeding step S63. It is determined whether or not the reaction of the skin sensors 54 of 24R and 24L has been shaken, that is, whether or not there has been contact.

  If “NO” in the step S63, that is, if an obstacle (a person or an object) is present in a place where the object is likely to collide but is not yet in contact, the process returns to the step S61 and the skin sensor 54 is monitored. Continue communication behavior by avoidance trajectory.

  On the other hand, if “YES” in the step S63, that is, if an obstacle is touched, the right arm motor 36 and the left arm motor 38 are controlled in a subsequent step S65 to interrupt the communication action. In this embodiment, since the robot 10 is covered with the soft skin 20, it is safe even if it collides.

  In step S 67, the audio data is given to the sound input / output board 66 and “I ’m sorry” is uttered from the speaker 48. When the process of step S67 is completed, the process returns to step S61, and the reaction of the skin sensor 54 is monitored again at regular intervals.

  On the other hand, if “NO” in the step S61, that is, if there is no obstacle at the place where the arms 24R and 24L are likely to collide, the right arm motor 36 and the left arm motor 38 are controlled in a step S69. Return the speed and trajectory of the arms 24R, 24L to normal ones. Then, the process returns to step S55, and the reaction of the skin sensor 54 is watched at regular intervals while continuing normal communication behavior.

  The flowchart shown in FIG. 9 shows an example of an operation when “shake a hand! In the first step S81 in FIG. 9, the CPU 56 instructs the sensor input / output board 64 to increase the sensitivity of the skin sensors 531-539 and 540-548 arranged on the moving arms 24R, 24L and move the arms 24R. , 24L is set to the proximity sensor mode.

  Next, in step S83, a program and data for executing this “shake a hand! Action” are read from the memory 60, and a “shake a hand! Action” is started. When this action is started, first, in step S85, the voice data is given to the sound input / output board 66 to output the synthesized voice of “shake handshake” from the speaker 48. Subsequently, in step S87, the control data is sent. The operation is given to the motor control board 62 to control the right arm motor 36 and the left arm motor 38, and the operation of inserting a hand is executed.

  In step S89, for example, detection data is acquired from the sensor input / output board 66 at regular intervals, and it is determined whether or not the skin sensors 531 to 539 and 540 to 548 arranged on the left arms 24R and 24L have reacted. .

  If “YES” in the step S89, that is, if the opponent brings his hand close to any of the arms 24R and 24L of the robot 10 to shake hands, in the subsequent step S91, the sensor input / output board 64, the sensitivity of the skin sensor 54 at the place of reaction is lowered, and the skin sensor 54 at that place is changed to the pressure sensor mode. In step S93, the right arm motor 36 and the left arm motor 38 are controlled to perform an operation of shaking the arms 24R and 24L around the gripped position, that is, a handshake operation.

  In step S95, for example, the detection data from the sensor input / output board 64 is acquired at regular intervals, and the pressure detected by the skin sensor 54 at the place where it is gripped by measuring the strength of the pressure as the contact state. Determine if has become smaller.

  If “YES” in this step S95, the opponent is trying to release his hand to finish the handshake, and in the subsequent step S97, the right arm motor 36 and the left arm motor 38 are controlled to end the handshake operation. Then, in step S99, the right arm motor 36 and the left arm motor 38 are controlled to perform the operation of pulling out the hand that has been drawn out, and this process is terminated.

  On the other hand, if “NO” in the step S89, it is determined whether or not there is a reaction in the skin sensor 54 disposed in another part in a step S101. If “NO” in the step S101, the process returns to the step S89 to repeat the monitoring of the skin sensor 54 for every fixed period. On the other hand, if “YES” in this step S101, that is, if the other party is trying to grasp a part other than the arm that the opponent has presented, the speaker 48 speaks “It is not a handshake” in the following step S103. . Then, the process proceeds to step S99, and the operation of pulling out the hand that has been handed out is executed, and this process ends.

  The flowchart shown in FIG. 10 shows an example of an operation in the case of “Hold me! In the first step S111 of FIG. 10, the CPU 56 instructs the sensor input / output board 64 to increase the sensitivity of all the skin sensors 54 and set the whole body skin sensors 54 to the proximity sensor mode. Next, in step S113, the program and data relating to “Hold me! Action” are read from the memory 60, and “Hold me! Action” is started. When this action is started, voice data is given from the memory 60 to the sound input / output board 66 in step S115, and a synthesized voice of “Hold me” is outputted from the speaker 48. Subsequently, in step S117, control data is given to the motor control board 62 to control the right arm motor 36 and the left arm motor 38, and an operation of spreading a hand for holding is executed.

  In step S119, detection data is acquired from the sensor input / output board 64 at regular intervals, for example, and it is determined whether or not there is a response to the piezo sensor sheet 509-513 arranged around the shoulder. If “YES” in this step S119, that is, if a large opponent is trying to hug the robot 10, the sensitivity of the skin sensor 54 of the whole body is lowered and changed to the pressure sensor mode in step S121. . Subsequently, in step S123, the right arm motor 36 and the left arm motor 38 are controlled to start the operation of making a ring with the arms 24R and 24L and hugging the opponent.

  In step S125, for example, detection data is acquired from the sensor input / output board 64 at regular intervals, and the skin sensor 54 is pressed a little by measuring the strength of pressure as a contact state. Determine if. Here, the determination is made based on data from the skin sensor 54 disposed in a contact portion such as an arm, chest, or stomach. At this time, the contact state may be reliably detected by changing the sensitivity within a range where the skin sensor 54 functions as a pressure sensor.

  If “YES” in the step S125, that is, if the pressure value data indicates a value in a predetermined range such as a state where the pressure data is strongly held by the opponent, for example, the right arm motor 36 and the left arm motor 38 are controlled in a step S127. Then, the hugging operation is stopped, and “more tight” is uttered from the speaker 48 in step S129. As described above, in this embodiment, when the skin sensor 54 is used to detect that the partner is large, the partner can be hugged so that the partner can be pampered by strongly hugging and uttering sweet words.

  If “NO” in the step S119, that is, if there is no response to the skin sensor 54 around the shoulder, it is determined whether or not there is a response to the piezo sensor sheets 526, 527, and 529 arranged in the lower abdomen in the step S131. to decide. If “NO” in the step S131, the process returns to the step S119, and the monitoring of the skin sensor 54 is repeated at regular intervals. On the other hand, if “YES” in this step S131, that is, if a small opponent is trying to hug the robot 10, the sensitivity of the whole body skin sensor 54 is lowered in step S133, and the pressure sensor mode is changed. Subsequently, in step S135, an operation of making a ring with the arms 24R and 24L and hugging the opponent is started.

  In step S137, for example, detection data is acquired from the sensor input / output board 64 at regular intervals, and the pressure sensor is measured as a contact state to determine whether the skin sensor 54 is touched. To do. Also at this time, the sensitivity of the pressure sensor 54 may be adjusted appropriately as described above.

  If “YES” in this step S137, that is, if the pressure value data indicates a value within a predetermined range such as a state where the force is hung on the other party, the right arm motor 36 and the left arm motor 38 in step S139. To stop the hugging movement. In step S141, an operation of moving one arm up and down is executed, and in step S143, the speaker 48 utters “good”. As described above, in this embodiment, when the skin sensor 54 is used to detect that the partner is small, the partner can be gently held by gently hugging and issuing a friendly word.

  When the process of step S129 or step S143 is completed, the right arm motor 36, the left arm motor 38, and the wheel motor 16 are controlled in step S145 to perform the operation of spreading and separating the arms 24R, 24L, and the holding operation is performed in step S147. The process is terminated.

  According to this embodiment, by dynamically changing the sensitivity of the piezo sensor sheet 54, the sensor can function as a proximity sensor and a pressure sensor, and by disposing the piezo sensor sheet 54 throughout the body, Both a distributed proximity sensor and a whole body distributed pressure sensor can be realized simultaneously. Therefore, new tactile communication can be realized by dynamically changing the sensitivity according to the situation and using the functions of the sensor properly.

  In the above-described embodiment, the piezo sensor sheets 54 are distributed throughout the body, such as the head, torso, shoulders, arms, and hands, of the human body-like portion 18, but the piezo sensor sheets 54 are executed by the robot 10. You may arrange | position as many as required to a required part according to predetermined | prescribed operation | movement.

  In each of the above-described embodiments, since the piezo sensor sheet 54 is functioned as a proximity sensor, the ultrasonic distance sensor used in the prior art robot is not provided, but it is necessary to measure the distance to a person or the like. For example, an ultrasonic distance sensor may be provided as necessary.

FIG. 1 is an illustrative view showing a communication robot according to an embodiment of the present invention. FIG. 2 is an illustrative view showing the skin used in FIG. 1 embodiment and a piezo sensor sheet embedded therein. FIG. 3 is an illustrative view showing an arrangement position of the piezo sensor sheet. FIG. 4 is a block diagram showing the electrical configuration of the embodiment of FIG. FIG. 5 is an illustrative view partially showing a sensor input / output board for inputting a detection signal from the piezo sensor sheet in FIG. 1 embodiment. FIG. 6 is a flowchart showing an operation when performing the “patrol action” in the embodiment of FIG. 1. FIG. 7 is a flowchart showing an operation when performing “waiting action” in the FIG. 1 embodiment. FIG. 8 is a flowchart showing an operation when risk avoidance is performed during the communication action in the FIG. 1 embodiment. FIG. 9 is a flowchart showing an operation when “shake a hand! Act” in the embodiment of FIG. FIG. 10 is a flowchart showing an operation when “hold me! Act” in the embodiment of FIG. 1.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Communication robot 18 ... Human body part 20 ... Skin 54,501-548 ... Piezo sensor sheet | seat 56 ... CPU
62 ... Motor control board 64 ... Sensor input / output board 66 ... Sound input / output board 68 ... Substrate 70 ... PIC microcomputer 72 ... A / D converter 74 ... Piezo film 76a, 76b ... Conductor 82 ... Operational amplifier 84 ... Multiplexer

Claims (9)

A piezoelectric sensor provided with variable sensitivity ,
Depending on the proximity sensor that changes the sensitivity of the piezoelectric sensor to measure capacitance or the changing means that causes the piezoelectric sensor to function as a pressure sensor that measures pressure, and detection information detected as the proximity sensor or the pressure sensor A communication robot comprising motion control means for controlling motion.   The communication robot according to claim 1, wherein when the approach or contact is detected based on the detection information detected as the proximity sensor, the operation control unit performs an operation different from that when no approach or contact is detected.   The communication robot according to claim 2, wherein the operation control unit reduces the operation speed when an approach is detected based on detection information detected as the proximity sensor.   The communication robot according to claim 2 or 3, wherein the motion control means changes the motion trajectory when an approach is detected based on detection information detected as the proximity sensor. When the approach or contact is detected based on detection information detected as the proximity sensor, the changing means changes sensitivity to cause the piezoelectric sensor to function as a pressure sensor,
The communication robot according to claim 1, wherein the operation control unit controls the operation according to detection information detected as the pressure sensor.   The communication robot according to claim 1, wherein the piezoelectric sensor is disposed in a portion that moves when expressing by gesture.   The communication robot according to claim 1, wherein the piezoelectric sensor is disposed at least on any of a head, a torso, a shoulder, an arm, and a hand.   The communication robot according to claim 1, wherein the piezoelectric sensors are arranged in a distributed manner. It is further equipped with a skin made of a flexible material to be put on the robot body
The communication robot according to claim 1, wherein the piezoelectric sensor is disposed in the skin.

Images