JP5419007B2 - Communication robot development support device - Google Patents

Description

  The present invention relates to a communication robot development support device, and more particularly to a communication robot development support device for developing a communication robot that takes communication behavior by executing a behavior module including a series of behavior programs.

  The present applicant has proposed a communication robot development support apparatus in Patent Document 1. For example, Patent Literature 1 discloses a technique for easily visualizing the relationship between behavior modules for controlling communication behavior (behavior) included in a communication robot on a screen. In the technique of Patent Document 1, since the relationship between behavior modules can be intuitively grasped, for example, creation / editing of rules for autonomous behavior can be appropriately performed.

Japanese Patent Laying-Open No. 2006-88328 [G06F 9/06]

  However, in the technique of Patent Document 1, it is necessary for the developer to specify the detailed operations included in the behavior module one by one. That is, it is necessary for the developer to create individual action programs for executing the detailed operations included in the action module for all the action modules, which requires extra effort for development.

  Therefore, a main object of the present invention is to provide a novel communication robot development support device.

  Another object of the present invention is to provide a communication robot development support device that can easily create a behavior module.

  The present invention employs the following configuration in order to solve the above problems. Note that reference numerals in parentheses, supplementary explanations, and the like indicate correspondence relationships with embodiments described later to help understanding of the present invention, and do not limit the present invention in any way.

The first invention is a communication robot development tool for developing communication robot to take communication behavior by executing the behavior module, classifies the behavior module consisting of a series of behavioral program for each type of communication behavior In addition, a template storage means for storing a template other than the execution information as a template, a template reading means for reading the template of the behavior module corresponding to a desired communication action from the template storage means, and a template read by the template reading means are visualized Te, read by the first display means, and the template reading means for displaying a first input screen including an argument input region for inputting the execution information of the behavior module And Plates, based on the execution information input argument input region of the first input screen includes a behavior module creating means for creating a new behavior module, a communication robot development support system.

  In the first invention, the communication robot development support device (10) is for supporting the development of the communication robot (12) and includes a template storage means (30). In the template storage means, for example, behavior modules are stored as templates for each type of communication behavior. The template reading unit (S33) reads a desired template from the template storage unit. The first display means (26, S35) visualizes the template read by the template reading means and displays the input screen (400, 500) on the display device (26). The input screen includes an argument input area (404, 504). For example, a developer of a communication robot or the like inputs execution information related to execution of the behavior module to the argument input area. The behavior module creating means (S37) creates a new behavior module by taking the execution information input in the argument input area of the input screen into the template data read by the template reading means.

  According to the first invention, a new behavior module can be easily created.

  The second invention is dependent on the first invention and the template read by the template reading means and the execution information input to the argument input area of the first input screen are created by the behavior module creating means. The first information corresponding to the behavior module read by the behavior module reading means, the behavior module reading means for reading the desired behavior module from the behavior module storage means, and the behavior module reading means is already input. A second display means for displaying a second input screen including the argument input area, and an action read by the action module reading means based on the second information re-input in the argument input area of the second input screen Action module update means for updating the module is further provided.

  In the second invention, the action module storage means (32) executes the template read by the template reading means (S33) and the execution input inputted to the argument input area (404, 504) of the input screen (400, 500). Information is stored in association with a behavior module created based on the information. The behavior module reading means (S43) reads a desired behavior module from the behavior module storage means. The second display means (26, S45) displays a second input screen corresponding to the behavior module read by the behavior module reading means on the display device (26). In the argument input area (404, 504) of the second input screen, the execution information when the behavior module is created is already input. For example, the developer of the communication robot or the like can execute the execution information in the argument input area. Enter again. The behavior updating means (S49) updates the behavior module based on the execution information re-input to the argument area.

  According to the second invention, a similar behavior module can be easily created simply by partially re-entering execution information, for example.

  According to the present invention, since the behavior module is stored as a template for each type of communication behavior, a desired template is read out, and execution information relating to the execution of the behavior module as an argument is taken into the template data. You can easily create a new behavior module.

  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.

It is an illustration figure which shows an example of the development assistance apparatus and communication robot of one Example of this invention. It is a block diagram which shows the internal structure of the development assistance apparatus shown in FIG. It is an illustration figure which shows an example of the table memorize | stored in action module DB. It is an external view which shows an example of the communication robot shown in FIG. It is a block diagram which shows the electrical structure of a communication robot. It is an illustration figure which shows an example of the composer screen displayed with a development assistance apparatus. It is an illustration figure which shows an example of the input screen displayed with a development assistance apparatus. It is an illustration figure which shows another example of the input screen displayed with a development assistance apparatus. It is a flowchart which shows an example of operation | movement of a development assistance apparatus. It is a flowchart which shows an example of operation | movement of the action module creation / update process of FIG. It is a flowchart which shows an example of operation | movement of a communication robot. FIG. 12 is a flowchart showing an example of an operation in the case of a behavior execution process belonging to the “Talk (greeting)” template among the behavior execution processes belonging to the template of FIG. 11; FIG. 12 is a flowchart showing an example of behavior execution processing belonging to the “Guide” template among the behavior execution processing belonging to the template of FIG. 11. FIG. 12 is a flowchart showing an example of behavior execution processing belonging to a “Navi (guidance)” template among behavior execution processing belonging to the template of FIG. 11;

  Referring to FIG. 1, a communication robot development support apparatus (hereinafter simply referred to as “development support apparatus”) 10 of this embodiment is a communication robot that takes a communication action by executing an action module consisting of a series of action programs. (Hereinafter simply referred to as “robot”).

  The development support apparatus 10 is configured by a computer such as a personal computer (PC) or a workstation, and includes, for example, a CPU 20 as shown in FIG. The CPU 20 is also called a microcomputer or a processor, and is connected via a bus 22 to a memory 24, a display device 26 such as a liquid crystal display or CRT, and an input device 28 such as a mouse and a keyboard.

  Although not shown, the memory 24 includes a ROM, an HDD, and a RAM. The ROM and HDD store a program for controlling the operation of the development support apparatus 10, and various data such as screen data displayed on the display device 26 is also stored. Furthermore, the ROM and the HDD store programs and data for controlling the behavior of the robot 12. Here, the behavior indicates the communication behavior of the robot 12 realized by the behavior module, and a plurality of behavior modules are stored in the ROM and HDD in association with each behavior. The RAM is used as a work memory or a buffer memory.

  Further, the CPU 20 is connected to a plurality of databases (hereinafter referred to as DB) via the bus 22. Specifically, the CPU 20 is connected to the template DB 30 and the behavior module DB 32.

  In the template DB 30, behavior modules are templated for each type of behavior such as “Talk (greeting)”, “Guide (direction guidance)”, and “Navi (guidance)”, and stored in advance. As will be described in detail later, the developer of the robot 12 reads the template data (template data) stored in the template DB 30 and stores information (execution information) related to the execution of the behavior module in the template data. A new behavior module can be created.

  The behavior module DB 32 stores behavior modules created based on templates and execution information. As an example, as shown in FIG. 3, in association with the behavior (Bi) of the robot 12, the type of the template (Tj) and the execution information (arg1, arg2, arg3,...) Given as an argument to the template. , ArgN) is stored as a table.

  However, the configuration of the table shown in FIG. 3 is merely an example, and the configuration of the table may be any configuration as long as it can realize the technical idea of the present invention.

  Further, the CPU 20 is connected to the communication LAN board 34 via the bus 22. The communication LAN board 34 is constituted by a DSP, for example, and provides transmission data given from the CPU 20 to the wireless communication device 36, and the wireless communication device 36 transmits the transmission data to the robot 12 via the network 200. The communication LAN board 34 receives data via the wireless communication device 36 and provides the received data to the CPU 20.

  Here, the robot 12 that supports development by the development support apparatus 10 will be described. The robot 12 is a humanoid type and autonomously moving type having various sensors, and can take a communication action using at least one of gesture and voice. As described above, the program for realizing the communication behavior (behavior) of the robot 12 is executed as a modularized “behavior module”. In addition, as will be described in detail later, the execution order of behavior modules is set as “state transition of behavior modules”, and communication behavior that maintains a consistent context or a harmonious situation in the long term is realized. Is done. “Behavior module state transition” indicates a short-term transition or transition of the behavior module. For example, it may be a connection or ordering of several behavior modules, but a long-term (for example, all day) or all behaviors. It does not define the transition of modules. That is, the robot 12 sequentially executes the “behavior modules”, and the execution order of the behavior modules is guided by “state transition of the behavior modules”.

  However, the “episode rule” can also be applied as a rule regarding the execution order of the behavior modules. This “episode rule” is disclosed in Japanese Patent Application Laid-Open No. 2004-114242 filed by the applicant in advance and already published.

  The hardware configuration of the robot 12 will be described in detail with reference to FIG. As shown in FIG. 4, the robot 12 includes a carriage 50, and two wheels 52 and one slave wheel 54 that autonomously move the robot 12 are provided on the lower surface of the carriage 50. The two wheels 52 are independently driven by a wheel motor 76 (see FIG. 5), and the carriage 50, that is, the robot 12 can be moved in any direction, front, back, left, and right. The slave wheel 54 is an auxiliary wheel that assists the wheel 52. Therefore, the robot 12 can move in the arranged space by autonomous control. However, the robot 12 may be fixedly arranged at a certain place.

  A cylindrical sensor mounting panel 58 is provided on the carriage 50, and a large number of infrared distance sensors 60 are mounted on the sensor mounting panel 58. These infrared distance sensors 60 measure the distance to the sensor mounting panel 58, that is, the object (such as a human being or an obstacle) around the robot 12.

  In this embodiment, an infrared distance sensor is used as the distance sensor, but an ultrasonic distance sensor, a millimeter wave radar, or the like can be used instead of the infrared distance sensor.

  A body 62 is provided on the sensor mounting panel 58 so as to stand upright. Further, the above-described infrared distance sensor 60 is further provided at the front upper center of the body 62 (a position corresponding to a person's chest), and measures the distance mainly to a person in front of the robot 12. Further, the body 62 is provided with a support column 64 extending from substantially the center of the upper end on the side surface side, and an omnidirectional camera 66 is provided on the support column 64. The omnidirectional camera 66 captures the surroundings of the robot 12 and is distinguished from an eye camera 90 described later. As the omnidirectional camera 66, for example, a camera using a solid-state imaging device such as a CCD or a CMOS can be adopted. In addition, the installation positions of the infrared distance sensor 60 and the omnidirectional camera 66 are not limited to the portions, and can be changed as appropriate.

  An upper arm 70R and an upper arm 70L are provided at upper end portions on both sides of the body 62 (a position corresponding to a human shoulder) by a shoulder joint 68R and a shoulder joint 68L, respectively. Although illustration is omitted, each of the shoulder joint 68R and the shoulder joint 68L has three orthogonal degrees of freedom. That is, the shoulder joint 68R can control the angle of the upper arm 70R around each of three orthogonal axes. A certain axis (yaw axis) of the shoulder joint 68R is an axis parallel to the longitudinal direction (or axis) of the upper arm 70R, and the other two axes (pitch axis and roll axis) are orthogonal to the axes from different directions. It is an axis to do. Similarly, the shoulder joint 68L can control the angle of the upper arm 70L around each of three orthogonal axes. A certain axis (yaw axis) of the shoulder joint 68L is an axis parallel to the longitudinal direction (or axis) of the upper arm 70L, and the other two axes (pitch axis and roll axis) are orthogonal to the axis from different directions. It is an axis to do.

  In addition, an elbow joint 72R and an elbow joint 72L are provided at the tips of the upper arm 70R and the upper arm 70L, respectively. Although illustration is omitted, each of the elbow joint 72R and the elbow joint 72L has one degree of freedom, and the angles of the forearm 74R and the forearm 74L can be controlled around the axis (pitch axis).

  A sphere 76R and a sphere 76L corresponding to a human hand are fixedly provided at the tips of the forearm 74R and the forearm 74L, respectively. However, when a finger or palm function is required, a “hand” in the shape of a human hand can be used. Although not shown, the front surface of the carriage 50, the portion corresponding to the shoulder including the shoulder joint 68R and the shoulder joint 68L, the upper arm 70R, the upper arm 70L, the forearm 74R, the forearm 74L, the sphere 76R, and the sphere 76L, , A contact sensor 78 (shown generically in FIG. 5) is provided. A contact sensor 78 on the front surface of the carriage 50 detects contact of a person or other obstacle with the carriage 50. Therefore, when the robot 12 is in contact with an obstacle during its movement, the robot 12 can detect the contact and immediately stop the driving of the wheel 52 to suddenly stop the movement of the robot 12. Further, the other contact sensors 78 detect whether or not the respective parts are touched. In addition, the installation position of the contact sensor 78 is not limited to the said site | part, and may be provided in an appropriate position (position corresponding to a person's chest, abdomen, side, back, and waist).

  A neck joint 80 is provided at the upper center of the body 62 (a position corresponding to a person's neck), and a head 82 is further provided thereon. Although illustration is omitted, the neck joint 80 has a degree of freedom of three axes, and the angle can be controlled around each of the three axes. A certain axis (yaw axis) is an axis directed directly above (vertically upward) of the robot 12, and the other two axes (pitch axis and roll axis) are axes orthogonal to each other in different directions.

  The head 82 is provided with a speaker 84 at a position corresponding to a human mouth. The speaker 84 is used for the robot 12 to communicate with a person around it by voice or sound. A microphone 86R and a microphone 86L are provided at a position corresponding to a human ear. Hereinafter, the right microphone 86R and the left microphone 86L may be collectively referred to as a microphone 86. The microphone 86 captures ambient sounds, in particular, the voices of humans who are subjects of communication. Furthermore, an eyeball part 88R and an eyeball part 88L are provided at positions corresponding to human eyes. The eyeball part 88R and the eyeball part 88L include an eye camera 90R and an eye camera 90L, respectively. Hereinafter, the right eyeball part 88R and the left eyeball part 88L may be collectively referred to as the eyeball part 88. Further, the right eye camera 90R and the left eye camera 90L may be collectively referred to as an eye camera 90.

  The eye camera 90 captures a human face approaching the robot 12, other parts or objects, and captures a corresponding video signal. The eye camera 90 can be the same camera as the omnidirectional camera 66 described above. For example, the eye camera 90 is fixed in the eyeball unit 88, and the eyeball unit 88 is attached to a predetermined position in the head 82 via an eyeball support unit (not shown). Although illustration is omitted, the eyeball support portion has two degrees of freedom, and the angle can be controlled around each of these axes. For example, one of the two axes is an axis (yaw axis) in a direction toward the top of the head 82, and the other is orthogonal to the one axis and goes straight in a direction in which the front side (face) of the head 82 faces. It is an axis (pitch axis) in the direction to be performed. By rotating the eyeball support portion around each of the two axes, the tip (front) side of the eyeball portion 88 or the eye camera 90 is displaced, and the camera axis, that is, the line-of-sight direction is moved. In addition, the installation positions of the above-described speaker 84, microphone 86, and eye camera 90 are not limited to the portions, and may be provided at appropriate positions.

  As described above, the robot 12 of this embodiment includes independent two-axis driving of the wheels 52, three degrees of freedom of the shoulder joint 68 (6 degrees of freedom on the left and right), one degree of freedom of the elbow joint 72 (2 degrees of freedom on the left and right), It has a total of 17 degrees of freedom: 3 degrees of freedom for the neck joint 80 and 2 degrees of freedom for the eyeball support (4 degrees of freedom on the left and right).

  FIG. 5 is a block diagram showing an electrical configuration of the robot 12. Referring to FIG. 5, the robot 12 includes a CPU 100. The CPU 100 is also called a microcomputer or a processor, and is connected to the memory 104, the motor control board 106, the sensor input / output board 108, and the audio input / output board 110 via the bus 102.

  Although not shown, the memory 104 includes a ROM, an HDD, and a RAM. In the ROM and the HDD, the same program (behavior module) for controlling the behavior of the robot 12 as that stored in the memory 24 of the development support apparatus 10 is stored in association with each behavior. Furthermore, the same ROM and HDD as those stored in the template DB 30 are templated for each type of behavior such as “Talk (greeting)”, “Guide”, “Navi (guidance)”. The action module is stored. The RAM is used as a work memory or a buffer memory.

  The motor control board 106 is configured by, for example, a DSP, and controls driving of motors of axes such as arms, neck joints, and eyeballs. That is, the motor control board 106 receives control data from the CPU 100 and controls two motors (in FIG. 5, collectively referred to as “right eyeball motor 112”) that control the angles of the two axes of the right eyeball portion 88R. Control the rotation angle. Similarly, the motor control board 106 receives the control data from the CPU 100, and shows two motors for controlling the respective angles of the two axes of the left eyeball portion 88L (in FIG. 5, collectively referred to as “left eyeball motor 114”). ) To control the rotation angle.

  The motor control board 106 receives control data from the CPU 100, and includes a total of four motors including three motors for controlling the angles of the three orthogonal axes of the shoulder joint 68R and one motor for controlling the angle of the elbow joint 72R. The rotation angle of two motors (collectively indicated as “right arm motor 116” in FIG. 5) is controlled. Similarly, the motor control board 106 receives control data from the CPU 100, and includes three motors for controlling the angles of the three orthogonal axes of the shoulder joint 68L and one motor for controlling the angle of the elbow joint 72L. The rotation angles of a total of four motors (collectively indicated as “left arm motor 118” in FIG. 5) are controlled.

  Further, the motor control board 106 receives control data from the CPU 100, and controls three motors for controlling the angles of the three orthogonal axes of the neck joint 80 (in FIG. 5, collectively indicated as “head motor 130”). Control the rotation angle. The motor control board 106 receives control data from the CPU 100 and controls the rotation angles of two motors (collectively referred to as “wheel motors 76” in FIG. 5) that drive the wheels 52. In this embodiment, a motor other than the wheel motor 76 uses a stepping motor (that is, a pulse motor) in order to simplify the control. However, a DC motor may be used similarly to the wheel motor 76. The actuator that drives the body part of the robot 12 is not limited to a motor that uses a current as a power source, and may be changed as appropriate. For example, in another embodiment, an air actuator or the like may be applied.

  Similar to the motor control board 106, the sensor input / output board 108 is configured by a DSP and takes in signals from each sensor and gives them to the CPU 100. That is, data relating to the reflection time from each of the infrared distance sensors 60 is input to the CPU 100 through the sensor input / output board 108. Further, the video signal from the omnidirectional camera 66 is input to the CPU 100 after being subjected to predetermined processing by the sensor input / output board 108 as necessary. Similarly, the video signal from the eye camera 90 is also input to the CPU 100. Further, signals from the plurality of contact sensors 78 described above (collectively referred to as “contact sensors 78” in FIG. 5) are given to the CPU 100 via the sensor input / output board 108. Similarly, the voice input / output board 110 is also configured by a DSP, and voice or voice in accordance with voice synthesis data provided from the CPU 100 is output from the speaker 84. Also, voice input from the microphone 86 is given to the CPU 100 via the voice input / output board 110.

  The CPU 100 is connected to the communication LAN board 132 via the bus 102. The communication LAN board 132 is configured by a DSP, for example, and sends transmission data given from the CPU 100 to the wireless communication device 134, and the wireless communication device 134 sends the transmission data to the development support device 10 via the network 200. The communication LAN board 132 receives data via the wireless communication device 124 and gives the received data to the CPU 100.

  The development support apparatus 10 is useful for the development of the robot 12 as described above. Specifically, the development support apparatus 10 has a function of creating a new behavior module based on a template and execution information given as an argument to the template. For example, the development support apparatus 10 is used. By sequentially adding action modules, the robot 12 having a large-scale action pattern can be realized.

  In the development support device 10, when the development support process is started, a composer screen 300 as shown in FIG. 6 is displayed on the display device 26, for example. However, each screen shown in this embodiment is an example and can be changed as appropriate.

  On this composer screen 300, the developer of the robot 12 can set the state transition of the behavior module. The composer screen 300 includes a visualization screen 302, and the state transition of the behavior module is graphically visualized and displayed on the visualization screen 302. Specifically, on the visualization screen 302, the behavior corresponding to the behavior module is displayed as a rectangular icon (behavior icon), and the name of the behavior is displayed in text (character) below the behavior icon. The Although detailed explanation is omitted, on the visualization screen 302, an arrow line (transition line) indicating a transition of the behavior module, a start icon, an end icon, and the like are displayed. Transitions or transitions can be easily grasped.

  A list screen 304 is provided on the right side of the visualization screen 302 in FIG. The list screen 304 includes an area 306 and an area 308. On the list screen 304, a list of behavior names is displayed as a target object (behavior object) of a drag and drop operation by the input device 28 such as a mouse. The Specifically, when the developer of the robot 12 operates the input device 28 such as a mouse and drags a behavior object on the list screen 304 and drops it on the visualization screen 302, the behavior corresponding to the behavior is displayed. An icon is drawn on the visualization screen 302.

  For example, in the area 306, action modules that are not templated, such as “Akushu (handshake)”, “Hug”, “Jyan”, and the like, that is, creation / update processing based on a template are performed. Behavior objects corresponding to behavior modules that cannot be displayed are displayed.

  In the area 308, a behavior module that is made into a template such as “Talk (greeting)”, “Guide”, “Navi (guidance)”, that is, creation / update processing based on the template is performed. The behavior objects corresponding to the possible behavior modules are classified and displayed for each template to which the behavior belongs.

  Further, an “add” button is provided at the bottom of each template, and the input screens 400 and 500 corresponding to the template are displayed by selecting the “add” button with the input device 28 such as a mouse. be able to.

  The input screens 400 and 500 shown in FIGS. 7 and 8 are obtained by visualizing the templates. As described above, the “add” button displayed on the list screen 304 can be displayed using the input device 28 such as a mouse. By selecting, it is displayed in a separate window from the composer screen 300.

  In the input screens 400 and 500, the developer of the robot 12 creates a new behavior module based on the template and the execution information by inputting information (execution information) related to the execution of the behavior module as an argument. Can do. For example, FIG. 7 shows an input screen 400 visualizing a “Talk” template, and FIG. 8 shows an input screen 500 visualizing a “Guide” template.

  First, the “Talk (greeting)” input screen 400 will be described with reference to FIG. 7.

  The input screen 400 shown in FIG. 7 includes an area 402a, an area 402b, an area 402c (argument input area 404), and an area 402d. Below the areas 402a to 402d, an “OK” button 406 and a “CANCEL” button are displayed. 408, as well as a “Clear” button 410 are provided.

  The name of the template is displayed in the area 402a. The example shown in FIG. 7 indicates that this behavior belongs to the “Talk (greeting)” template.

  The name of the behavior is displayed in the area 402b. In the example shown in FIG. 7, that the name of this behavior is "Hello" is shown.

  In the area 402c, execution information input as an argument to this “Talk (greeting)” template is displayed. That is, the area 402 c functions as the argument input area 404. For example, the execution information is text (character) data corresponding to the utterance content of the robot, and a method in which the developer of the robot 12 directly inputs manually using the input device 28 such as a keyboard can be considered. In the example shown in FIG. 7, indicated that the robot 12 utters "Hello. Welcome.".

  A description of the behavior and the like are displayed in the area 402d. For example, in the area 402d, toward the "head in the face position of the dialogue partner," Hello. Welcome. A commentary (description) of the behavior, such as “shake the arm as appropriate while speaking” is displayed.

  Next, a template input screen 500 for “Guide” will be described with reference to FIG.

  The edit screen 500 shown in FIG. 8 includes an area 502a, an area 502b, an area 502c (argument input area 504a), an area 502d (argument input area 504b), an area 502e (argument input area 504c), and an area 502f (argument input area 504d). , An area 502g (argument input area 504e) and an area 502h, and an “OK” button 506, a “Cancel” button 508, and a “Clear” button 510 are provided below the areas 502a to 502h.

  The name of the template is displayed in the area 502a. The example shown in FIG. 8 indicates that this behavior belongs to the “Guide” template.

  The name of the behavior is displayed in the area 502b. In the example illustrated in FIG. 8, the behavior name is “To Miles”.

  In the area 502c, execution information given as an argument to the “Guide” template is displayed. That is, the area 502c functions as an argument input area 504a. For example, the execution information is text data corresponding to the utterance content of the robot, and a method in which the developer of the robot 12 directly inputs manually using the input device 28 such as a keyboard is conceivable. In the example shown in FIG. 8, it is shown that the robot 12 speaks “XX (guide point name)”.

  In the area 502d, execution information given as an argument to the “Guide” template is displayed. That is, the area 502d functions as an argument input area 504b. For example, the execution information is the location information of the relay point, and the developer of the robot 12 or the like directly inputs the coordinates of the relay point manually using the input device 28 such as a keyboard, or a preset place name. A method of selecting by a so-called pull-down menu method using an input device 28 such as a mouse, a method of displaying a map screen (not shown) in a window different from the input screen 500, and selecting an arbitrary place on the map, etc. It is done. In the example shown in FIG. 8, the relay point is “Oriental Diner”.

  In the area 502e, execution information given as an argument to the “Guide” template is displayed. That is, the area 502e functions as an argument input area 504c. For example, the execution information is text data corresponding to the utterance content of the robot, and a method in which the developer of the robot 12 directly inputs manually using the input device 28 such as a keyboard is conceivable. In the example shown in FIG. 8, it is shown that the robot 12 utters “Oh (the name of the relay point) at the end of that road”.

  In the area 502f, execution information given as an argument to the “Guide” template is displayed. That is, the area 502f functions as an argument input area 504d. For example, the execution information is the position information of the guide point, and the developer of the robot 12 or the like directly inputs the coordinates of the guide point manually using the input device 28 such as a keyboard, or a preset place name. A method of selecting by a so-called pull-down menu method using an input device 28 such as a mouse, a method of displaying a map screen (not shown) in a window different from the input screen 500, and selecting an arbitrary place on the map, etc. It is done. In the example illustrated in FIG. 8, it is indicated that the guide point is “Miles”.

  In the area 502g, execution information given as an argument to the “Guide” template is displayed. That is, the area 502g functions as an argument input area 504e. For example, the execution information is text data corresponding to the utterance content of the robot, and a method in which the developer of the robot 12 directly inputs manually using the input device 28 such as a keyboard is conceivable. In the example shown in FIG. 8, it is shown that the robot 12 speaks “I'm over there, I'm out of the exit”.

  The content of the utterance of the robot 12 is displayed in the area 502h. Specifically, a sentence that summarizes the utterance contents of the robot 12 displayed in the area 502c, the area 502e, and the area 502g is displayed.

  In such input screens 400 and 500, the developer of the robot 12 or the like inputs execution information in the argument input areas 404 and 504, and then selects “OK” buttons 406 and 506 with the input device 28 such as a mouse. Can create a new behavior module based on the template. Further, the developer or the like of the robot 12 can close the input screens 400 and 500 by selecting the “cancel” buttons 408 and 508. Furthermore, the developer or the like of the robot 12 can initialize the execution information input in the argument input area by selecting the “clear” buttons 410 and 510.

  The input screens 400 and 500 select the behavior object displayed on the list screen 302 with the input device 28 such as a mouse, or select the behavior icon displayed on the visualization screen 302 with the input device 28 such as a mouse. For example, it can be displayed by selecting by double-clicking. In this case, the developer or the like of the robot 12 can update the existing behavior module by re-inputting the execution information already input in the argument input areas 404 and 504.

  Specifically, the CPU 20 of the development support apparatus 10 executes the development support process according to the flowchart shown in FIG.

  When the process is started, the CPU 20 first reads screen data from the memory 24 in step S1, and displays a composer screen 300 as shown in FIG.

  Next, in step S3, it is determined whether there is a stop command. For example, if a developer or the like of the robot 12 selects an end menu (not shown) on the composer screen 300, “YES” is determined, and in the subsequent step S5, an end process is executed to end the development support process. On the other hand, if “NO” in the step S3, that is, if there is no stop instruction, a behavior module creation / update process (see FIG. 10) described later is executed in a step S7, and the process proceeds to the step S9.

  In step S9, it is determined whether or not state transition of the behavior module is set. Specifically, after the state transition of the behavior module is set on the visualization screen 302 of the composer screen 300, it is determined whether or not a decision menu (not shown) on the composer screen 300 is selected. If the behavior module state transition is set, “YES” is determined, and in step S5, information (transition information) regarding the behavior module state transition is transmitted to the robot 12, and the process returns to step S3. The transition information also includes execution information input as an argument in the template. As will be described in detail later, the robot 12 that has received the transition information from the development support apparatus 10 is designated. Data of the template corresponding to the behavior module is read from the memory 104, and the behavior specified in the behavior module is performed by giving execution information as an argument referred to from the transition information to the template data.

  On the other hand, if “NO” in the step S9, that is, if the behavior module state transition is not set, the process returns to the step S3 as it is.

  FIG. 10 is a flowchart of the action module creation / update process of step S7 shown in FIG. As illustrated in FIG. 10, when the CPU 20 of the development support apparatus 10 starts the behavior module creation / update process, in step S31, the CPU 20 determines whether to create a new behavior module. For example, when the “add” button displayed on the list screen 304 is selected by the developer of the robot 12 using the input device 28 such as a mouse, “YES” is determined in the step S31, and the next step S33. Thus, the template data corresponding to the “add” button is read from the template DB 30.

  Subsequently, in step S35, the template read from the template DB 30 in step S33 is visualized, and the input screen 400 as shown in FIG. 7 (or the input screen 500 as shown in FIG. 8) is displayed as the composer screen 300. The data is displayed on the display device 26 in a separate window.

  In step S37, action module creation processing is executed. Specifically, a new behavior module is created by taking execution information input in the argument input area 404 of the input screen 400 into template data read from the template DB 30. Here, the developer or the like of the robot 12 inputs execution information into the argument area 404 of the input screen 400, and then selects an “OK” button 406 on the input screen 400.

  Next, in step S39, the created new behavior module is stored in the behavior module DB 32. Then, a behavior name (behavior object) corresponding to the created behavior module is newly added to the list screen 304.

  When step S39 ends, the process returns to step S31, and it is repeatedly determined whether or not a new behavior module is to be created. If “NO” in the step S31, that is, if a new behavior module is not created, the process proceeds to a step S41 as it is.

  In step S41, it is determined whether to update an existing behavior module. For example, when the behavior object displayed on the list screen 304 is selected by the developer of the robot 12 or the like, “YES” is determined in step S41, and the action corresponding to the behavior object is determined in the next step S43. Module data is read from the behavior module DB 32.

  Subsequently, in step S45, the input screen 400 as shown in FIG. 7 (or the input screen 500 as shown in FIG. 8) is displayed on the display device 26 as a separate window from the composer screen 300.

  In step S47, an existing behavior module update process is executed. Here, for example, the developer of the robot 12 corrects the execution information already input in the argument area 404 of the input screen 400. Thereafter, by selecting the “OK” button 406 on the input screen 400, the behavior module is updated based on the execution information as the argument re-input in the argument area 404.

  Next, in step S49, the updated behavior module is stored in the behavior module DB 32. Here, the updated behavior module is stored in the behavior module DB 32 in place of the existing behavior module. When step S49 is completed, the process returns to step S31 to repeatedly determine whether or not to create a new behavior module.

  If “NO” in the step S41, that is, if the existing behavior module is not updated, the process proceeds to a step S51 as it is, and it is determined whether or not the behavior module creating / updating process is terminated in a step S51. . If “YES” in the step S51, the behavior module creating / updating process is ended, and the process proceeds to a step S9 (see FIG. 9).

  FIG. 11 is a flowchart showing the overall processing of the CPU 100 of the robot 12. As shown in FIG. 11, when the CPU 100 of the robot 12 executes the entire process, in step S71, the CPU 100 executes a predetermined behavior to be performed first. That is, an action module corresponding to a predetermined behavior to be performed first is read from the memory 104, and an action defined by the action module is performed. In addition, the behavior to be executed first may be set to investigate the surrounding environment such as “EXPLORE (surrounding the surrounding environment)”.

  Subsequently, in step S73, it is determined whether or not there is a stop command. Here, for example, if development is in progress, it is determined whether or not there has been an end instruction from the development support apparatus 10 or whether or not an end button for stopping the robot 12 has been pressed. If “YES” in the step S73, an end process is executed in a subsequent step S75, and the operation process of the robot 12 is ended. In this end process, each part of the body of the robot 12 may be returned to the home position.

  On the other hand, if “NO” in the step S73, that is, if there is no stop command, it is determined whether or not transition information regarding the state transition of the behavior module is transmitted from the development support apparatus 10 via the network 200 in a step S77. to decide. If “YES” in the step S 77, the transition information is received and written in the memory 104 in a succeeding step S 79.

  Subsequently, in step S81, the first behavior specified in the transition information is selected, and in step S83, it is determined whether or not the behavior belongs to the template. Specifically, it is determined whether or not the behavior module corresponding to the behavior is a behavior module created by taking execution information input in the argument input areas 404 and 504 of the input screens 400 and 500 into template data. To do.

  If “YES” in the step S83, that is, if the behavior belongs to the template, an execution process of the behavior belonging to the template (see FIG. 12-14) is started in a step S85, and the process proceeds to the step S89.

  On the other hand, if “NO” in the step S83, that is, if the behavior does not belong to the template, the execution process of the behavior not belonging to the template is started in a step S87. That is, the behavior module corresponding to the behavior is read from the memory 104, the behavior prescribed in the behavior module is performed, and the process proceeds to step S89.

  In a succeeding step S89, it is determined whether or not execution of all the behaviors specified by the transition information is completed. If “YES” in the step S89, that is, if execution of all the behaviors specified by the transition information is completed, the process returns to the step S73 and the process is repeated.

  On the other hand, if “NO” in the step S89, that is, if the execution of all the behaviors is not yet completed, the next behavior specified by the transition information is selected in a step S91. When the process of step S91 is completed, the process returns to step S83, and it is repeatedly determined whether or not the behavior belongs to the template.

  12-14 is a flowchart of behavior execution processing based on the template in step S85 shown in FIG. When the CPU 100 of the robot 12 starts the behavior execution process based on the template, the CPU 100 reads the template data corresponding to the behavior from the memory 104 and gives execution information as an argument referred to from the transition information to the template data. Perform the actions specified in the Action Module.

  FIG. 12 shows an example of behavior execution processing when the behavior belongs to the “Talk (greeting)” template.

  In step S111, the CPU 100 of the robot 12 takes the video signal from the eye camera 90 into the sensor input / output board 108 and processes the video signal, thereby obtaining the coordinates (x, y, z) of the face of the conversation partner. calculate.

  In the following step S113, the calculated coordinate data is given to the motor control board 106 so that the body direction of the robot 12 is directed to the direction (x, y) of the conversation partner, and the face and eyeball part of the head 82 of the robot 12 Point 88 to the position (x, y, z) of the person's face.

  Next, in step S115, the execution information given to the template as the argument arg1 is referred to from the transition information, and the execution information is transmitted to the voice input / output board 110 so that the voice or voice is output from the speaker 84. To control. The execution information given as the argument arg1 is, for example, text data entered as text (characters) in the argument input area 404 of the “Talk (greeting)” input screen 400 shown in FIG. Speaks according to the contents of the text data.

  In step S117, physical movements (justures) such as shaking arms are executed in parallel until the utterance in step S115 is completed. When the process of step S115 is terminated, the behavior execution process based on the template is terminated, and the process proceeds to step S89 (see FIG. 11).

  FIG. 13 shows an example of behavior execution processing when the behavior belongs to the “Guide (direction guide)” template.

  As shown in FIG. 13, in step S 121, the execution information given as the argument arg 1 to the template is referred to from the transition information, the execution information is transmitted to the voice input / output board 110, and the voice or voice is sent from the speaker 84. Control to output. The execution information given as the argument arg1 is, for example, text data inputted as text (characters) in the argument input area 504a of the “Guide” input screen 500 shown in FIG. , Speak as the contents of the text data, and present the place name of the guidance point to be guided from now.

  Next, in step S123, the execution information given as the argument arg2 to this template is referred to from the transition information, and the relay point coordinates (x, y) are calculated from the execution information. The execution information given as the argument arg2 is, for example, the position information input in the argument input area 504b of the “Guide” input screen 500 shown in FIG. In step S125, the coordinates (x, y) of the relay point are given to the motor control board 106, and the relay point is indicated by a pointing operation or the like.

  In subsequent step S127, the execution information given as the argument arg3 to the template is referred to from the transition information, and the execution information is transmitted to the voice input / output board 110 so that the voice or voice is output from the speaker 84. To do. The execution information given as the argument arg3 is, for example, text data inputted as text (characters) in the argument input area 504c of the “Guide” input screen 500 shown in FIG. Speak the content of the text data and explain the route to the relay point.

  Next, in step S129, the execution information given as the argument arg4 to this template is referred to from the transition information, and the coordinates (x, y) of the guide point are calculated from the execution information. The execution information given as the argument arg4 is, for example, the position information input in the argument input area 504d of the “Guide” input screen 500 shown in FIG. In step S131, the coordinates (x, y) of the guide point are given to the motor control board 106, and the guide point is indicated by a pointing operation or the like.

  In the subsequent step S133, the execution information given to the template as the argument arg5 is referred to from the transition information, the execution information is given to the voice input / output board 110, and the voice or voice according to the execution information is output from the speaker 84. To be controlled. The execution information given as the argument arg5 is, for example, text data input as text (characters) in the argument input area 504e of the “Guide” input screen 500 shown in FIG. Speak the content of the text data and explain the route from the relay point to the guide point.

  When the process of step S133 is terminated, the behavior execution process based on the template is terminated, and the process proceeds to step S89 (see FIG. 11).

  FIG. 14 shows an example of behavior execution processing when the behavior belongs to the “Navi (guidance)” template.

  As shown in FIG. 14, in step S141, referring to the execution information given as the argument arg1 to this template from the transition information, the execution information is given to the voice input / output board 110, and the voice or Control is performed so that the voice is output from the speaker 84. The execution information given as the argument arg1 is, for example, text data including the name of the destination, and the robot 12 speaks according to the contents of the text data and presents the destination.

  Next, in step S143, the execution information given as the argument arg2 to this template is referred to from the transition information, and the coordinates (x, y) of the destination are calculated from the execution information. In step S145, the coordinates (x, y) of the destination are given to the motor control board 106, and the person is moved (guided) to the destination.

  When step S145 is completed, that is, when the vehicle arrives at the destination, in step S147, for example, “I got it” is spoken, the behavior execution processing based on the template is terminated, and the process proceeds to step S89 (see FIG. 11). .

  Thus, in this embodiment, the behavior module is templated for each type of behavior and stored in the template DB 30, and the input screen 400 that visualizes the template by reading the data of the desired template from the template DB 30. , 500 can be displayed. A new behavior module can be created by taking the execution information input in the argument input areas 404 and 504 of the input screens 400 and 500 into the template data. Therefore, according to this embodiment, a new behavior can be easily created.

  Further, in this embodiment, the behavior module created based on the template and the execution information is stored in the behavior module DB 32 in association with the behavior of the robot 12, and a desired behavior is read from the behavior module DB 32. The input screens 400 and 500 can be displayed. The action module is updated based on the execution information re-entered in the argument areas 404 and 504 by re-inputting the execution information already input in the argument input areas 404 and 504 of the input screens 400 and 500. can do. Therefore, according to this embodiment, a similar behavior can be easily created simply by partially re-entering execution information, for example.

10 ... Development support device 12 ... Communication robot 20 ... CPU
22 ... Bus 24 ... Memory 26 ... Display device 28 ... Input device 30 ... Template DB
32 ... Behavior module DB
60 ... Infrared distance sensor 66 ... Omnidirectional camera 78 ... Contact sensor 84 ... Speaker 86 ... Microphone 90 ... Eye camera 100 ... CPU
DESCRIPTION OF SYMBOLS 102 ... Bus 104 ... Memory 106 ... Motor control board 108 ... Sensor input / output board 110 ... Voice input / output board 200 ... Network 300 ... Composer screen 400, 500 ... Input screen 404, 504 ... Argument input area

Claims (2)

A communication robot development support device for developing a communication robot that takes a communication action by executing an action module,
A template storage means for classifying and storing the behavior module consisting of a series of behavior programs for each type of communication behavior and making a template other than the execution information ;
Template reading means for reading a template of the behavior module corresponding to a desired communication behavior from the template storage means;
Visualizing the template read by the template reading means, a first display means for displaying a first input screen including an argument input area for inputting the execution information of the behavior module , and the template reading means A communication robot development support apparatus comprising behavior module creation means for creating a new behavior module based on the read template and the execution information input to the argument input area of the first input screen. A behavior module storage that stores the template read by the template reading unit and the execution information input to the argument input area of the first input screen in association with the behavior module created by the behavior module creation unit. means,
Behavior module reading means for reading a desired behavior module from the behavior module storage means;
Second display means for displaying a second input screen corresponding to the behavior module read by the behavior module reading means and including an argument input area in which execution information has already been input; and the argument of the second input screen The communication robot development support apparatus according to claim 1, further comprising behavior module updating means for updating the behavior module read by the behavior module reading means based on execution information re-input to the input area.

Images