JP4555039B2 - Robot, robot control method, robot control program - Google Patents

Description

The present invention relates to a method for controlling a robot and a robot, are those concerning the program.

In recent years, robots that recognize a user, interact with the user, and act autonomously have been developed (for example, Patent Document 1, Patent Document 2, and Patent Document 3). Such robots are designed to behave in the same way as humans and organisms. By giving robots information on internal states such as personality information and emotion information, they can behave according to internal states. Control that behaves in the same way as humans and living things is performed.
JP 2000-326274 A WO00 / 015396 JP 2002-283261 A

  However, models of internal states such as actual human and biological characters and emotions have not been fully elucidated, and the external states of robots are currently designed and developed by the subjectivity of developers. For this reason, the behavior of the robot is not intuitive to the user, and it is often impossible to understand the robot's behavior. In addition, since the robot's recognition ability is not sufficient, the robot does not perform the user's desired action, and there is a problem that it is difficult for the user to communicate with the robot.

  The present invention has been made in view of the above circumstances, and the present invention provides a technique for allowing a user to understand a robot and to have familiarity with it.

According to the present invention, a robot that performs thinking and acts based on the thought, the external information acquisition unit that acquires external information related to the external state of the robot, and the external state management unit that holds the acquired external information And internal state management means holding internal information about the internal state of the robot that changes with the acquired external information as a trigger, and at least the external information as a trigger to develop thoughts , and actions to be taken based on the results of the thought comprising a thought means for determining, and an output means for outputting the information consisting of external information and internal information such as the basis for thinking over extent itself and thought process by thinking means, and thought processes are those over multistage , output means, robot provided, which comprises at least a portion the screen display of the multi-stage over thought process as at least a part of the output information It is.

In this robot, before Symbol thinking means it may be configured to perform thinking based on the external information and the internal information.

  The output means may output at least a part of the external information or the internal information as the information representing a thought process.

Furthermore, the output method determining means, for example, can be configured to determine an output method in consideration of the external information or internal information.

  The external information includes information issued from a user, and the output method determination means determines an output method of the information representing the thought process in consideration of the information issued from the user. Good.

  The output means includes a plurality of output units that output information via a robot action, robot utterance or display, and the output method determination means selects any of the plurality of output units, and the output method The information representing the thought process may be output via the output unit selected by the determining means.

Various modes can be adopted for the external information and the output method determining means. The following is an example.
(i) The external information includes information related to noise in the surrounding environment, and the output method determination means considers the information related to noise, and if there is noise in the surroundings, the screen display is set as the output method. If quiet, utterance or pronunciation is used as the output method.
(ii) The external information includes information regarding the presence or absence of a location where an image can be projected, and the output method determination means displays the screen display at the location as the output method when the location where the image is projected exists. To do.
(iii) The external information includes information related to the user's age group, and the output method determining means determines the output method in consideration of the user's age group.

Here, with respect to the presentation of prior Ki思 considered process can take various forms. The following is an example.
(i) The thought process presentation method management means presents the thought process by any one of a method of uttering with one's own voice, a method of uttering with a third party's voice, and a method of sounding with sound effects.
(ii) The thought process presentation method management means presents the thought process by any one of a method of moving hands, feet, ears, and the like, and a method of lighting lamps such as eyes and mouth.
(iii) The thinking process presentation method management means presents the thinking process by any one of a graph structure, text, table, time series chart, camera image, and robot diagram.

According to the present invention, there is provided a method for controlling a robot, the first step of acquiring external information related to the external state of the robot with respect to the robot, and the second step of storing the acquired external information, a third step of carrying internal information about the internal state of the robot which changes the external information as a trigger, performs thought triggered at least external information, a fourth scan based on the result of the thinking, determines the action to be taken a step, a fifth step of outputting the fourth information comprising an external information and internal information such as the basis for thinking over extent itself and thought process in step is executed, the thinking process, are those over multistage in the fifth step, the robot system, which comprises at least a portion the screen display of the multi-stage over thought process as at least a part of the output information A method is provided.

In this robot control method, before Symbol second step, thinking may be used as the configuration of the row Uyo urchin control based on the external information and the internal information.

According to the present invention, there is provided a program for controlling a robot, the first step for acquiring external information related to the external state of the robot with respect to the robot, and the second step for storing the acquired external information. a third step of carrying internal information about the internal state of the robot which changes the external information as a trigger, performs thought triggered at least external information, a fourth scan based on the result of the thinking, determines the action to be taken a step, a fifth step of outputting the fourth information comprising an external information and internal information such as the basis for thinking over extent itself and thought process in step is executed, the thinking process, are those over multistage in the fifth step, Robo, characterized in that the screen displaying at least a portion of the multi-stage over thought process as at least a part of the output information Door control program is provided.

In this robot control program, before Symbol second step, thinking may be configured to line Uyo urchin controlled based on the external information and the internal information.

  The present invention simulates the behavior of humans and organisms as in conventional robotic devices, and does not aim for the relationship between humans and humans and organisms as the relationship between humans and robots, but builds new relationships between humans and robots. To solve it.

  In particular, in the conventional robot apparatus, the fact that it is difficult for the user to understand what the robot thinks is an obstacle to the user's familiarity with the robot. For example, it can be seen from the tone and behavior of the robot that the robot is in a bad mood, but I did not know why it was bad.

  Therefore, in the present invention, a configuration for outputting information representing a thought process is employed. “Information representing a thinking process” refers to the thinking process itself, or external information or internal information that is the basis of the thinking process. According to the present invention, since the above configuration is adopted, the user can understand why the robot takes such action when the robot takes action.

 Further, when the robot does not perform the user's desired action, the reason can be known. As described above, according to the present invention, communication between the user and the robot can be improved, and the user can be made to understand the robot and become familiar.

  In this invention, it can be set as the structure which holds external information and internal information inside a robot, and thinks based on this. If it does in this way, it will become possible to express a thought process using external information or internal information, and the information showing a thought process can be outputted suitably. Note that “holding” means that the robot is held at least temporarily inside the robot, and includes a mode of holding until the point of completion of thinking and a mode of holding until the point of time when information indicating the thinking process is output.

  In the present invention, an output method of information representing a thought process may be determined in accordance with an internal state or the like of the external state of the robot, and output may be performed according to the determined method. In this way, since the information is output in a form that is easy for the user to understand, the communication between the robot and the user can be further improved.

  The output method can be determined in consideration of, for example, external information or internal information. For example, what kind of environment the robot is placed in, what kind of person (user) who took action against the robot is, and when is the current date and time, etc. Thus, the output method can be determined as appropriate. As an output method, there are action, utterance, display, and the like, and these may be appropriately combined.

  Although the configuration of the present invention has been described above, any combination of the above-described components, and a conversion of the expression of the present invention between methods and apparatuses are also effective as an aspect of the present invention.

  As described above, according to the present invention, the configuration that outputs information representing the thinking process is adopted, so that the user can understand the robot better and have a friendship. Can do.

  Next, embodiments of the present invention will be described in detail with reference to the drawings. In the following description, the configuration and function of the robot will be mainly described. However, the present invention discloses not only such a robot but also a method and a program for controlling the robot. In the following description, information indicating a thought process is appropriately expressed as a “thinking process”.

FIG. 1 is an external configuration diagram illustrating an example of a robot apparatus according to the present embodiment. This robot apparatus is a robot that performs thinking and acts based on the thinking, and has the following configuration.
(i) External information acquisition means (external state input unit 102) for acquiring external information related to the external state of the robot
(ii) External state management means holding external information (external state management unit 103)
(iii) Internal state management means (internal state management unit 104) that holds internal information about the internal state of the robot that changes with the acquired external information as a trigger
(iv) Thinking means (thinking unit 105) that thinks based on external information and internal information and decides an action to be taken based on the result of thinking

  The robot apparatus 101 is configured, for example, by connecting the body 1 and the head 2. A wheel 3A and a wheel 3B are attached to the lower part of the body part 1 on the left and right, respectively, and these wheels can rotate back and forth independently.

  The head 2 can rotate within a predetermined range with respect to a vertical axis that is vertically attached to the body 1 and a horizontal axis that is installed at an angle of 90 degrees with respect to the vertical axis. The vertical axis is installed so as to pass through the center of the head 2, and the horizontal axis is installed horizontally through the center of the head 2 with the body 1 and the head 2 facing the front and in the left-right direction. . That is, the head 2 can rotate within a predetermined range with two degrees of freedom, left and right and up and down.

  A speaker 12, a projection unit 15, and the like are provided on the surface of the body unit 1. Further, on the surface of the head 2, a CCD camera 21A and a CCD camera 21B, a voice recognition microphone 13, a head switch 14, and the like are provided.

FIG. 2 is a block diagram illustrating a configuration of the robot apparatus 101 according to the present embodiment.
The robot apparatus 101 includes an external state input unit 102, an external state management unit 103, an internal state management unit 104, a thinking unit 105, a thought process presentation processing unit 106 (output method determining unit), and an output processing unit 110.

  The external state input unit 102 inputs an external state around the robot apparatus 101. The external state input unit 102 may be an input device such as a CCD camera 21A, a CCD camera 21B, a microphone such as the voice recognition microphone 13, an obstacle sensor, a motor, a switch, a cable, a gyro, or an external signal input unit. . The obstacle sensor includes a sensor such as an ultrasonic sensor or an infrared sensor. The external signal input unit can also input signals from external devices such as surveillance cameras, mobile phones, other robots, and media such as ID tags.

  The external state management unit 103 holds external information relating to the external state of the robot. In other words, the external state input unit 102 recognizes the external state of the robot apparatus 101 based on information acquired from the outside. The external state management unit 103 recognizes colors, shapes, faces, and the like based on image information acquired by the external state input unit 102, for example. Further, the external state management unit 103 recognizes a sound source direction, a sound, a human voice, a voice inflection, and the like based on the audio information acquired by the external state input unit 102, for example. In addition, the external state management unit 103 recognizes the presence / absence of an obstacle, a distance, and the like based on the obstacle information acquired by the external state input unit 102, for example. Also, the external state management unit 103 recognizes an event such as stroking, hitting, pressing, or hitting an obstacle based on the pressure information acquired by the external state input unit 102, for example. In addition, the external state management unit 103 recognizes information such as the posture and the current position during charging.

  The internal state management unit 104 includes an internal state storage unit 107 and an internal state acquisition unit 108. The internal state storage unit 107 stores information representing the internal state of the robot. Specifically, it stores internal information relating to the internal state such as operation mode (state) information, personality information, personality type information, mood information, emotion information, desire information, growth stage information, history DB information, etc. . Here, the operation mode (state) information includes a dialogue mode, an autonomous action mode, and the like. Personality information defines positiveness, cooperation, diligence, intelligence, emotional stability, and so on. Personality type information defines things such as things, dance enthusiasts, game enthusiasts, guttara, and the like. The mood information defines that the mood is good, the mood is normal, the mood is bad. Emotion information defines joy, anger, sadness, and the like. The desire information defines appetite, sleep desire, activity desire, affinity desire, and the like. Growth stage information defines childhood, boyhood, adulthood, and the like. The history DB information defines the number of executions for each function / user, the number of times given up for each function / user, and the like.

  FIG. 3 is a diagram illustrating an example of a part of the internal configuration of the internal state storage unit 107. Here, personality type information will be described as an example. The personality type information includes “monogatari”, “dance lover”, “game lover”, and “guuta”. Each personality type is associated with a numerical value indicating the tendency of the personality type and stored. For example, “Monoshiri” is “0.5”, “Dancing Love” is “1”, “Game Love” is “0.8”, and “Gutara” is “0.2”. The robot apparatus 101 has the highest personality type of “dance lover”, and the personality types are lower in the order of “game lover”, “monolith”, and “guuta”.

  Returning to FIG. 2, the internal state acquisition unit 108 acquires information on the necessary internal state from the internal state storage unit 107 as appropriate, and sends it to the thinking unit 105. Based on the external state recognized by the external state management unit 103, the internal state acquisition unit 108 can also read out information (internal information) related to the internal state from the internal state storage unit 107. For example, when the external state management unit 103 authenticates the user based on the image input by the external state input unit 102, the external state management unit 103 sends information identifying the authenticated user to the internal state acquisition unit 108. . Based on the user identification information sent from the external state management unit 103, the internal state acquisition unit 108 refers to the history DB information in the internal state storage unit 107 and reads the history related to the user.

  Further, the internal state acquisition unit 108 can update information in the internal state storage unit 107 based on the external state recognized by the external state management unit 103. For example, when the external state management unit 103 recognizes that the user has been praised, the internal state acquisition unit 108 sets the numerical value indicating the tendency of “joy” in the emotion information in the internal state storage unit 107 to be high. can do. Thus, by sequentially updating the information in the internal state storage unit 107, the current internal state of the robot apparatus 101 can be accurately stored in the internal state storage unit 107.

The thinking unit 105 develops thinking using external information as a trigger, and determines an action to be taken by the robot based on the result of the thinking. The thinking unit 105 includes an action determining unit 109 and a processing procedure storage unit 115. The action determination unit 109 determines the action to be taken next by the robot apparatus 101 based on the external state and the internal state. The processing procedure storage unit 115 stores a processing procedure for the behavior determination unit 109 to determine the behavior of the robot apparatus 101. The processing procedure storage unit 115 stores, for example, a neural network, a Bayesian network, a state transition model, a rule base, a flowchart, a programming language, and the like. In the present embodiment, the action determination unit 109 uses the internal state and external state (external information and internal information) of the robot apparatus 101 acquired from the external state management unit 103 and the internal state management unit 104 as parameters, and a processing procedure storage unit. In accordance with the processing procedure stored in 115, the action to be taken next by the robot apparatus 101 is determined.
Each example is shown below.

<Bayesian network>
A Bayesian network is a well-known method in the field of artificial intelligence, and will not be described in detail. However, a Bayesian network is a probability inference model in which a network is configured with the probability of an event as a unit. If input information is given as a non-ignition pattern and propagation calculation is performed, a non-ignition pattern can be obtained as output information. For example, it is possible to determine the action to be taken next by the robot apparatus 101 by assigning each of various kinds of information related to the internal state and the external state of the robot apparatus 101 to the node and performing propagation calculation of the Bayesian network.

FIG. 4 is a diagram illustrating an example of a Bayesian network for determining an action to be taken by the robot apparatus 101.
Here, each node has a probability value. Within each node, for example, the probability of an established event is “1”, the probability of an unsatisfied event is “0”, and the probability of an unobservable event is a value between “0” and “1”. Take. A link connecting nodes means that there is a dependency relationship between the nodes. For example, when the probability of “person identification (I met after a long time)” is high, the probability of “happy” increases. Probability dependency is managed by a table called CPT (Conditional Probability Table), and the probability is defined according to the true / false combination of nodes with input links. Here, “I was greeted”, “Person identification (I met for a long time)”, “Person identification (I met frequently)”, “I was struck by a head”, “He was struck”, etc. Whether or not it is established is determined based on the external state 101. Also, whether or not “I like dancing”, “Monodashiya”, etc. is established based on the internal state of the robot apparatus 101 is determined.

  The rightmost column of the Bayesian network represents the behavior of the robot apparatus 101. Based on the external state and internal state of the robot apparatus 101, the probability of each node is determined, and the action having the highest probability value is finally selected as the action to be taken next by the robot apparatus 101.

  FIG. 5 is a diagram illustrating a process in which an action to be taken next by the robot apparatus 101 is selected by the Bayesian network illustrated in FIG. 4. Here, for example, there is a high probability of “headed”, “favorite person”, “happy”, “dancing lover”, and finally “dancing” is selected as the action that the robot apparatus 101 should take next. Has been.

<Neural network>
A neural network is a well-known method in the field of artificial intelligence, and will not be described in detail. However, a neural network is a model that consists of neurons simulating the human brain, and the input information is not ignited. If the propagation calculation is given as a pattern, it is possible to obtain a non-firing pattern as output information. By assigning each of various types of information related to the internal state and external state of the robot apparatus 101 to a node and performing propagation calculation of the Bayesian network, it is possible to determine the action that the robot apparatus 101 should take next.

FIG. 6 is a diagram illustrating an example of a neural network for selecting an action to be taken by the robot apparatus 101.
Here, each node has an activation value. The activation value of each node can be calculated, for example, by a method of normalizing the sum of the activation value of each input node by a coefficient from 0 to 1 with a sigmoid function. Also in the neural network, as in the case of the Bayesian network described above, the action having the highest activity value is selected from the actions of the robot apparatus 101.

<State transition model>
Since the state transition model is a well-known method in the field of information processing, detailed explanations are omitted, but the states of robots are classified and enumerated, the meaning and behavior of each state are defined, and transitions between states The state transition model is defined by defining the conditions and timing to be performed. Here, what is shown as a network is called a state transition network.

<Rule base>
The rule base is a well-known method in the field of information processing, and will not be described in detail. However, there are many rules such as what to do under what conditions. This is realized by an if sentence or the like.

<Flowchart>
Since a flowchart is a method well known in the field of information processing, detailed description thereof is omitted, but a procedural language is mainly described by diagrams.

<Programming language>
The programming language is a well-known method in the field of information processing, and thus detailed description thereof is omitted, but there are a procedural language, an object-oriented language, a functional language, a logical language, and the like.

  Returning to FIG. 2, the output processing unit 110 performs a process of outputting the thinking process by the thinking unit 105. The output processing unit 110 includes an action unit 111, an audio output unit 112, a display unit 113, and an external output unit 114. The output processing unit 110 outputs the thinking process in the thinking unit 105 according to the method determined by the thinking process presentation processing unit 106. That is, the output processing unit 110 presents a thought process until the behavior determination unit 109 selects an action of the robot apparatus 101 based on an instruction from the thought process presentation processing unit 106. The thinking process (information indicating the thinking process) refers to information indicating the external state or internal state of the robot, thinking based on the information, and the like. The function of the thought process presentation processing unit 106 will be described later.

  The behavior unit 111 performs the behavior determined by the behavior determination unit 109. The action unit 111 is realized by an actuator or the like that drives the wheel 3A, the wheel 3B, or the like shown in FIG. In the present embodiment, the action unit 111 performs actions such as dancing, playing music, getting angry, searching for a person, approaching a person, and the like.

  The voice output unit 112 outputs the voice of the robot apparatus 101. The audio output means 112 is realized by the speaker 12 shown in FIG. The voice output unit 112 outputs a voice according to the action determined by the action determination unit 109. In the present embodiment, the audio output means 112 performs pronunciation and utterance such as greeting, giving a message, and conveying information.

  The display means 113 displays information. The display means 113 can be a monitor screen (not shown) built in the robot apparatus 101, for example. Further, the display means 113 can be configured by the projection unit 15 shown in FIG. Information can be projected from the projection unit 15 toward the screen. In the present embodiment, the display unit 113 performs display such as displaying a setting screen or displaying an image viewed by the robot.

  The external output unit 114 outputs information to an external terminal. The external terminal can be a display device such as a television, an audio output device, another robot device, a portable terminal, or the like.

The thought process presentation processing unit 106 determines a method for outputting the thought process of the robot apparatus 101 and performs a process of causing the output processing unit 110 to present the thought process. Examples of the output method include action, utterance, and display. The thought process presentation processing unit 106 selects one or more means constituting the output processing unit 110 and instructs the selected means to output the thought process.
Here, the thinking process of the robot apparatus 101 is, for example, a process of determining an action of the robot apparatus 101 in the thinking unit 105, an internal state or an external state of the robot apparatus 101, and the like. The thought process presentation processing unit 106 manages means, timing, contents, and the like for presenting the thought process of the robot apparatus 101.
The thought process presentation processing unit 106 refers to information (external information) on the external state temporarily held by the external state management unit 103, information on the internal state (internal information) stored in the internal state storage unit 107, and the like. Based on these, the output method of the thinking process is determined. Thereby, since an appropriate output method is selected according to the situation, communication between the robot and the user can be further improved.
Regarding the presentation of the thought process by the thought process presentation processing unit 106, various modes can be taken. The following is an example.
(i) The external information includes information related to the noise of the surrounding environment, and the thought process presentation processing unit 106 considers the information related to the noise, and uses the screen display as an output method if the surroundings are noisy. If so, utterance or pronunciation is used as the output method.
(ii) The external information includes information regarding the presence / absence of a location where an image can be projected, and the thinking process presentation processing unit 106 uses a screen display at the location as an output method when there is a location where an image is projected. For example, if there is a place where screen projection is possible, screen projection is performed.
(iii) External information includes information related to the user's age group, and the thought process presentation processing unit 106 determines an output method in consideration of the user's age group. For example, if the user is a child, the presentation form can be understood by the child.

  Below, the process of the thought process presentation process part 106 is demonstrated.

<Means>
Examples of a method for presenting the thinking process of the robot apparatus 101 include voice output and screen display.

  The thought process presentation processing unit 106 can cause the voice output device or the like connected via the voice output unit 112 or the external output unit 114 to output the thought process of the robot apparatus 101 by voice. At this time, in order to distinguish the thinking process of the robot apparatus 101 and the normal utterance of the robot apparatus 101, the thinking process presentation processing unit 106 sends a third party different from the voice of the robot apparatus 101 to the voice output unit 112. The voice of the thinking process of the robot apparatus 101 is output by voice.

  The thinking process presentation processing unit 106 can display the thinking process of the robot apparatus 101 on the display unit 113 or a display device connected via the external output unit 114.

  Examples of screen displays include graph structures, text, tables, time series charts, camera images, robot diagrams, and the like. Details of these will be described later.

  The thinking process presentation processing unit 106 can also present the robot's thinking process by a combination of voice output and screen display.

<Timing>
The thought process presentation processing unit 106 can present the thought process of the robot apparatus 101 based on a user instruction, for example. The thinking process presentation processing unit 106 presents the thinking process of the robot apparatus 101 when the user asks, for example, “Why are you doing xx?” Or “What are you thinking?” In this case, the voice of the user is input as voice waveform information by the voice recognition microphone 13 (see FIG. 1) which is one of the external state input units 102. The external state management unit 103 recognizes that the user has given a thought presentation instruction based on the speech waveform information input by the external state input unit 102. The external state management unit 103 notifies the thinking process presentation processing unit 106 of an instruction from the user. The thinking process presentation processing unit 106 acquires the thinking process of the robot apparatus 101 from the thinking unit 105 based on the notification from the external state management unit 103 and causes the output processing unit 110 to present it.

  Further, the thinking process presentation processing unit 106 can always cause the output processing unit 110 to present the thinking process of the robot apparatus 101. The user can set in advance the setting for always presenting the thinking process of the robot apparatus 101. The thinking process presentation processing unit 106 can also cause the output processing unit 110 to present the thinking process of the robot apparatus 101 at a predetermined timing. Here, the predetermined timing is when an event occurs, when an action is selected, when an action is executed, or periodically.

  When an event occurs, for example, when a predetermined button is pressed, the battery is less than a specified remaining amount, the appetite exceeds a threshold, a noise is heard, a person is found, etc. It can be when the state or internal state changes.

  The action selection time is when the thinking unit 105 determines the action that the robot apparatus 101 should take next. The thinking process presentation processing unit 106 can present the thinking process of the robot device 101 when the thinking unit 105 determines, for example, “dancing” as the next action that the robot device 101 should take.

  The action execution time is when the robot apparatus 101 is executing an action determined by the thinking unit 105. The thinking process presentation processing unit 106 can cause the thinking process of the robot apparatus 101 to be presented while the robot apparatus 101 is actually dancing, for example.

  The thinking process presentation processing unit 106 can also present the thinking process of the robot apparatus 101 at a predetermined interval, for example, regardless of the occurrence of an event. The thinking process presentation processing unit 106 may change the interval according to the situation instead of the predetermined interval, or may cause the thinking process of the robot apparatus 101 to be presented at an uncertain interval by a fractal, a random number, or the like.

<Contents>
The thinking process presentation processing unit 106 may present each step of the robot when the thinking process of the robot has multiple stages, or may present them collectively.

  Further, the thinking process presentation processing unit 106 presents the thinking process of the robot apparatus 101 and what kind of content is presented according to the external state of the robot apparatus 101 acquired by the external state management unit 103. You can decide what to do. For example, when the external state management unit 103 recognizes that the user is a child, the thinking process presentation processing unit 106 can output the sound or output the thinking process with easy-to-understand expressions. When the surroundings are noisy, the thought process presentation processing unit 106 can display the thought process on the screen. On the other hand, when the surroundings are quiet, the thinking process presentation processing unit 106 can output the thinking process by voice. Also, for example, when the external state management unit 103 recognizes something that can be a screen such as a plain wall in the surroundings, the thinking process presentation processing unit 106 performs a thinking process on the screen by the projection unit 15 of the robot apparatus 101. It can also be made to present.

FIG. 17 is a flowchart illustrating a processing procedure for presenting the thinking process of the robot apparatus 101.
When an instruction to present a thought process is given from the user, a certain event is uttered, or a setting is made to always present a thought process (YES in S100), the external state input unit 102 inputs the external state management unit 103. The external state is recognized (S102). The thinking process presentation processing unit 106 determines the presentation contents and the presentation means according to the external state recognized by the external state input unit 102 (S104). Specifically, the thought process presentation processing unit 106 determines the presentation contents and the presentation means according to the type of the target user who presents the thought process of the robot apparatus 101, the surrounding environment, and the like. Subsequently, the thought process presentation processing unit 106 causes a predetermined means to present the content of the thought process determined in step S104 (S106). Next, it is determined whether or not to end the presentation (S108). When the presentation is to be ended (YES in S108), the process returns to step S100, and the process of waiting for the next presentation timing is ended.

Hereinafter, an example of the thinking process of the robot apparatus 101 presented to the output processing unit 110 will be shown.
(Example 1) Screen display: graph structure The graph structure is a structure composed of nodes and edges (or links).
When presenting the thinking process of the robot apparatus 101 in a graph structure, an activity diagram of UML (Unified Modeling Language), a UML class diagram, a UML state chart diagram, a UML sequence diagram, a UML collaboration diagram, It can be shown in a diagram only of nodes without links. Since UML is a well-known method in the field of information processing, only a brief description will be given here.

  The UML activity diagram is a diagram for representing the flow of activity, and is used when the dynamic aspects of the system are specifically described. A flowchart or data flow diagram is a kind of activity diagram. When presenting the thinking process of the robot apparatus 101, a neural network or a Bayesian network can also be expressed using this figure.

  FIG. 5 described above represents the Bayesian network as a graph by assigning each event of the Bayesian network to each node and associating the events having a dependency relationship with the edges. Referring to FIG. 5, each square is a node, and a line connecting the nodes is an edge. Strictly speaking, since the edge is an arrow in this figure, it is classified into a directed graph. Here, “dancing” is selected as the action to be taken next by the robot apparatus 101, but the thinking process of the robot apparatus 101 in which this action is selected is highlighted.

  In order to display the thinking process of the robot apparatus 101 in an easy-to-understand manner, it is possible to perform processing such as displaying colors in a coordinated manner or filtering so as to leave only high-value nodes according to the probability value of each node. . Below, the emphasis presentation method in the case of presenting the thinking process of the robot apparatus 101 by a screen display is illustrated. There is a method of emphasizing and displaying in an enlarged display, a method of emphasizing and displaying in a frame display, a method of emphasizing and emphasizing a frame, a method of emphasizing and highlighting a link line, and a method of emphasizing and displaying an annotation. .

(A) Emphasize presentation by color change: For example, the portion that is usually displayed in white is displayed in red only in the emphasized portion, the portion that is usually displayed in light blue is displayed in dark blue only in the emphasized portion, usually in white Only the highlighted part is displayed in a red, blue, green, red, and so on.

(B) Emphasize by blinking: For example, the part that is normally lit is displayed with only the highlighted part blinking, the part that is normally displayed with the unlit part blinks only, and the part that is normally highlighted is blinking. However, processing such as displaying only the highlighted portion with a fast blink is performed.

(C) Emphasize presentation with a change in brightness: For example, a place where normal brightness is displayed at a brightness of 10 is displayed only at an emphasized portion at a brightness of 20 brightness, and only the emphasized portion is brightened between 10 and 20 Repeat the process of darkening or darkening.

(D) Emphasize presentation by change of display non-display: For example, only the part to be emphasized is displayed and the other part is hidden, only the part to be emphasized is displayed and the other part is not displayed After that, a process of applying a filter is performed while leaving a portion to be highlighted and displayed, such as displaying the displayed item in accordance with the size of the screen.

(E) Enhancing and presenting in an enlarged display: For example, an enlarged part to be emphasized is displayed and the other parts are reduced and displayed, an enlarged display is performed around the part to be emphasized and the other parts disappear from the screen. The process of zooming the portion to be highlighted is performed, such as displaying in detail the portion to be highlighted and displaying the other portion in outline.
(F) Emphasizing by displaying the frame: For example, a process of marking the portion to be emphasized is performed by displaying a frame surrounding the portion to be emphasized.

(G) Highlight the frame and highlight it: for example, change the color or thickness so that the frame of the part you want to emphasize and highlight is displayed, especially if it is a graph display, the frame around the node The process of making the highlighted part stand out, such as changing the color or changing the thickness so as to make it stand out, is performed.

(H) Emphasize and highlight the link line: For example, the color is changed according to the data flow so that the line to be emphasized is displayed more prominently or with different thickness. The process of making the highlighted part stand out, such as changing the display to, etc., is performed.

(I) Annotate with emphasis: For example, display text to point to the part you want to emphasize, or display a star-like shape on the part you want to emphasize Process to mark the part to be presented.

  The various emphasis processes described above can be applied to the following other screen display examples. Moreover, the part to be highlighted can be, for example, a corresponding part, a part related to the corresponding part, and a route to the corresponding part. In the case of a graph structure, the corresponding part is a node, the part related to the relevant part is a related node connected to the relevant node or a related node connected to the relevant node, etc. The related nodes connected to each other, the related nodes connected to the related nodes, and the related edges connecting the nodes. At this time, for example, the direction of the thought path can be emphasized by changing the color so as to flow in consideration of the direction of the edge.

  The UML class diagram is a diagram for mainly expressing the structure between object-oriented classes, and is composed of classes, interfaces, and their relationships. A robot model or the like having a categorized hierarchical relationship can be expressed using this diagram.

  FIG. 7 is a diagram illustrating the thinking process of the robot apparatus 101 in a class diagram. Here, as nodes, “input”, “voice recognition”, “image recognition”, “command”, “greeting”, “person detection”, “person identification”, “message”, “game”, “news”, “Output”, “Dialogue”, “Autonomous”, “Function”, “Greeting”, “Find people”, “Walk”, “Message”, “Game”, “News” are prepared, and hierarchical relationship Is shown with an edge. In this figure, for example, “input” includes “voice recognition” and “image recognition”, “voice recognition” includes “command” and “greeting”, and “command” includes “message” and “game”. “News” is included. By using such a class diagram for presenting the thinking process of the robot apparatus 101, it is possible to abstract information necessary for presentation.

  For example, when the user speaks “message” and the robot apparatus 101 recognizes the voice, the thought process presentation processing unit 106 can highlight “message”. At this time, it is possible to grasp the thinking process of the robot apparatus 101 by simultaneously highlighting “instructions”, “speech recognition”, and “input” which are related parent levels.

  The UML state chart diagram is a so-called state transition diagram, and is used to express a state transition. A state transition type robot model or the like can be expressed using this diagram.

  FIG. 8 is a diagram showing a thinking process of the robot apparatus 101 in a state chart diagram. In this case, states such as “listening mode”, “walk mode”, “message mode”, “play mode”, and “person search mode” are prepared as nodes, and the state transition relationship is an edge. It is shown. In this figure, for example, when the robot apparatus 101 recognizes the voice “message” in the “listening mode” state, the robot apparatus 101 transitions to the “message mode” state, and when a person is detected in the “person search mode” state, Transition to “mode” is shown. Thus, by displaying the entire topology like a map and highlighting the current state of the robot apparatus 101, the user can grasp the current state of the robot apparatus 101. It can be used as a guide for how to contact the robot apparatus 101 when it is desired to be in a certain state.

  The UML sequence diagram is a diagram representing a sequence of messages flowing between objects, and is used to represent an interaction between objects from a temporal viewpoint.

  The UML collaboration diagram is used to express the interaction between objects from a spatial viewpoint. A robot model used in a distributed environment having components other than the robot and a robot that performs parallel operation processing inside the robot can be expressed using this diagram.

  Moreover, a robot model can be expressed even in a diagram with only nodes without links. For example, for a robot described in a logical language, a logical expression may be assigned as a node and an edge may not be used. In such a case, the thought process of the robot can be presented by displaying or highlighting only the igniting logical expression.

(Example 2) Screen display: Text Text is a character. When presenting a thought process in text, there are methods of displaying a large amount of details of events such as speech recognition results and sensor results, or displaying only the main points.

(Example 3) Screen display: table A table is a set of cells composed of rows and columns. For example, by summarizing the robot's thinking rules in a table, the list can be improved as compared with the case of text alone. Highlighting the cell corresponding to the current state helps in presenting the context.

(Example 4) Screen display: time-series chart A time-series chart is obtained by mapping the presence or absence of an operation and the type of operation along the flow of time. FIG. 9 is a diagram showing a time series chart. Here, “head orientation”, “right foot”, “left foot”, “mouth LED”, “cheek LED”, and “voice” are prepared as types of motion, and a time axis is arranged on the horizontal axis. The presence or absence of motion is expressed by a thick line or a thin line. The vertical line means the current time, and shifts to the right as time passes. When presenting a thought process in a time-series chart, it is possible to highlight an operation that is in operation, and to display the presence and characteristics of past and future operations in thickness and color.

(Example 5) Screen display: camera image A camera image is an image captured by the CCD camera 21A or the CCD camera 21B. FIG. 10 is a diagram illustrating the thinking process of the robot apparatus 101 as a camera image. Here, the camera image is displayed on the left side of the screen, and the text is displayed on the right side. The text indicates the speech recognition result of the user's utterance. In the camera image, along with the image acquired by the CCD camera 21A and the CCD camera 21B, the recognition result of the image and information indicating the internal state with respect to the recognition result are displayed in text. As a result of image recognition, “person name: Achan” and “favorability: high” are shown. In addition, as a recognition result of the speech, it has been shown that it is a "Hello", "dancing".

  When presenting a thought process with a camera image, it is preferable to present a recognition result of the robot apparatus 101 and information indicating an internal state with respect to the recognition result together with the camera image.

(Example 6) Screen display: robot diagram A robot diagram is an appearance diagram, a design diagram, a partial configuration diagram, or the like of a robot, and may be a still image or a moving image. FIG. 11 is a diagram illustrating the thinking process of the robot apparatus 101 as a robot diagram. The robot diagram is placed at the center, the description of each part is displayed beside it, and the state is displayed. For example, this robot is composed of “head switch (head tapping detection switch)”, “voice recognition microphone”, “camera”, “mouth / cheek / ear / eye LED”, “head switch (detection sensor with headlid)”, “ Components such as “sound source detection microphone”, “speaker”, “ultrasonic sensor”, “step sensor”, “wheel”, “remote control transmission”, and the like are indicated. In addition, the state is indicated as “On”, “Off”, “Activate”.

  When presenting a thought process with a robot diagram, the user is further understood by highlighting a place to be noted. For example, here, a point of interest is highlighted by marking an explosion. There are two methods of highlighting: attaching to the text side and attaching to the robot diagram. When attaching to the robot diagram, change the color of the corresponding part or change the corresponding part to a square, etc. Highlighting can be performed by surrounding with a mark. For example, when the robot apparatus 101 is performing some action because an obstacle is detected by the right ultrasonic sensor, the right ultrasonic sensor can be highlighted. In addition, the ear lamp can be blinked when the surrounding sounds are loud and the voice cannot be heard.

(Example 7) Screen display: combination FIG. 12 is a diagram showing a combination of the above-described image display methods. Here, a camera image, a robot diagram, a graph structure, and a time series chart are displayed on one screen. By combining a plurality of display methods in this way, a large amount of information can be provided to the user in an easy-to-understand manner.

(Example 8) Screen Display: Projection FIG. 13 is a diagram schematically illustrating an example in which an image 201 indicating the thinking process of the robot apparatus 101 is projected from the projection unit 15 of the robot apparatus 101 toward the screen 200. Here, a plain wall or the like can be used as the screen 200. When the thinking process presentation processing unit 106 detects a screen on which an image can be projected, the thinking process presentation processing unit 106 can cause the projecting unit 15 to display an image indicating the thinking process of the robot apparatus 101 toward the screen.

(Example 9) Voice output For example, when the robot apparatus 101 “plays” to the user, if the user responds “because it is busy”, the emotion information of the robot apparatus 101 in the internal state storage unit 107 is displayed. Rewritten. For example, the thought process presentation processing unit 106 causes the voice output unit 112 to output a voice that is different from the normal voice of the robot apparatus 101 and that the voice is “I feel bad because I can't cover it”. Thereby, the user can grasp that the mood of the robot apparatus 101 has deteriorated. At this time, the thought process presentation processing unit 106 outputs a voice such as “It seems that he wanted to press the head button” in order to better indicate that the robot apparatus 101 wanted the user. Can do. In addition, by flashing the head lamp, it is possible to indicate that the head button was pushed and wanted to be bitten. Thereby, it is possible to grasp how the robot apparatus 101 is pleased from the next time, and communication between the user and the robot apparatus 101 can be promoted.

(Example 10) Audio output For example, when the robot device 101 is dancing, if the user gives up “good”, the robot device 101 recognizes that the user has given up, and the internal state storage unit 107 The number associated with “I like dance” in the personality type information increases. At this time, the thought process presentation processing unit 106 causes, for example, the voice output unit 112 to output a voice “I like dance more” with a voice different from the normal voice of the robot apparatus 101. Thereby, the user can grasp | ascertain that the robot apparatus 101 will like dancing by giving up dancing.

(Example 11) Voice output For example, when the user requests “message”, and the robot apparatus 101 likes a message, for example, a “pip-pip sound” or the like may be output. it can. At this time, the thought process presentation processing unit 106 outputs a voice such as “I am delighted” with a voice different from the normal voice of the robot apparatus 101. Thereby, the user can grasp that the robot apparatus 101 is pleased and that the beep sound is output when the robot apparatus 101 is pleased. Thus, from the next time, it is possible to grasp that the robot apparatus 101 is pleased to ask for a message even if there is no voice output of “I am happy”. The thought process presentation processing unit 106 recognizes a user, determines whether or not an explanation has been given to the user in the past, and is said to be “happy” only for a user who has not explained in the past. Audio can be output.

(Example 12) History The action determination unit 109 can determine the next action of the robot apparatus 101 in consideration of the user's past history stored in the internal state storage unit 107. For example, when the external state management unit 103 recognizes the user, the internal state acquisition unit 108 causes the internal state storage unit 107 to talk to the robot device 101 in the past and the actions performed with the robot device 101. Etc. are read. The behavior determination unit 109 determines the behavior of the robot device 101 based on the information read by the internal state acquisition unit 108 and causes the robot device 101 to act. The thinking process presentation processing unit 106 presents a series of processing procedures as a thinking process of the robot apparatus 101.

FIG. 14 is a screen showing an example of a thought process of the robot apparatus 101 presented by the thought process presentation processing unit 106 when the robot apparatus 101 is dancing.
Here, the recognition result of the external state indicates that the user is “Achan” and there is no utterance. As the history of the internal state of “Achan”, the information “I gave up after dancing” is shown. Based on these external state and internal state, the thinking unit 105 has an impression that “I want you to give up again”. The action of “dancing” is executed. Since these pieces of information are presented to the user, the user can know the impression and expectation of the robot, for example, can give up the robot apparatus 101 again. As a result, the favorable sensitivity of the robot apparatus 101 to “Achan” is increased, and the relationship between the user and the robot can be improved.

FIG. 15 is a screen showing an example of the thought process of the robot apparatus 101 presented by the thought process presentation processing unit 106 when the robot apparatus 101 asks the user “Can I add one channel?”.
Here, the recognition result of the external state indicates that the user is “Achan”, there is no utterance, and the current time is 20:55 on Sunday. In addition, as the history of the internal state regarding “Achan”, information that “I often watch TV of 1 channel from 21:00 on Sunday” is shown. Based on these external state and internal state, the thinking unit 105 causes the robot apparatus 101 to speak "Can I have 1 channel?" Since these pieces of information are presented to the user, the user can know the reason why the robot made such an utterance, and the relationship between the user and the robot apparatus 101 can be improved.

  FIG. 16 is a screen showing an example of the thinking process of the robot apparatus 101 presented by the thinking process presentation processing unit 106 when the robot apparatus 101 asks the user “Let's play a nervous breakdown game”.

  Here, the recognition result of the external state indicates that the user is “Achan” and there is no utterance. In addition, as the history of the internal state regarding “Achan”, information that “I lost the nerve weakness game yesterday” is shown. Based on the external state and the internal state, the thinking unit 105 causes the robot apparatus 101 to speak “Let's play a nervous breakdown game” as a result of having an impression that it is “regrettable and wants to do it again”. Thus, the user can know the emotion of the robot apparatus 101 and can become familiar with the robot apparatus 101.

(Example 13) When the robot apparatus 101 does not respond The robot apparatus 101 is set to recognize the user or perform an action according to the instruction from the user. However, if the user cannot recognize or the instruction from the user cannot be recognized, the robot apparatus 101 may not respond even if the user gives an instruction. In such a case, the user may not understand why the robot apparatus 101 does not respond, and may feel troublesome. The thought process presentation processing unit 106 presents the recognition result of the external state management unit 103, information about the internal state read by the internal state acquisition unit 108, and the like as the thought process of the robot apparatus 101.

  For example, when the robot apparatus 101 does not respond even if the user asks the robot apparatus 101 “Tell me the time,” the thought process presentation processing unit 106 displays the voice and the like recognized by the external state management unit 103 on the screen.

  For example, when the information “recognition result:“ it is right ”, the voice level is OK, and there is no corresponding word” is displayed, the user correctly recognizes the word “please tell me”. Therefore, it can be understood that the robot apparatus 101 does not respond. In addition, since the voice level is OK, it is possible to grasp that the loudness of the voice is good, but the way of speaking is bad, and then the robot says by saying “tell time” with a clear tone. It can be expected that the device 101 will respond.

  In addition, when the information “Recognition result:“ Jigakute ”, voice level is low, and there is no corresponding word” is displayed, the user correctly recognizes the word “Give me the word”. The robot device 101 will respond by re-phrasing “tell time” in a clear tone with the next loudest voice. can do.

  In addition, when the information “Recognition result:“ Give me a message ”, voice level OK, no corresponding word” is displayed, the user understands that the robot apparatus 101 does not have a function of teaching time. Can do.

  In this way, when the robot apparatus 101 does not respond, whether the robot apparatus 101 does not respond because the robot apparatus 101 does not recognize the user's voice correctly, or whether the robot apparatus 101 does not respond because the robot apparatus 101 has no function. Etc., and even when the robot apparatus 101 does not respond, the next effective countermeasure can be taken.

  In addition, for example, when the user says “call my name” and the robot does not respond, the thought process presentation processing unit 106 uses the voice recognized by the external state management unit 103, the external state management unit Information on the user recognized by the user 103, information in the internal state storage unit 107, and the like are displayed on the screen.

  For example, when the information “speech recognition“ call my name ”, image recognition“ no corresponding face image data ”” is displayed, the user says that the robot apparatus 101 reads “call my name”. Is recognized correctly, but it can be understood that the robot apparatus 101 does not respond because its face image data is not collated. In this case, the user asks the robot apparatus 101 for an instruction such as “My name is Taro, register my face and name”, or “I should have been registered. Can be asked.

  Further, for example, when information such as “voice recognition“ call my name ”, image recognition“ with corresponding face image data ”, no name registered” is displayed, the user can confirm that the robot apparatus 101 is “my name”. It is possible to grasp that the robot device 101 does not respond because the name “has been called” is correctly recognized and its own face image data is also collated, but the name is not registered. In this case, the user can instruct the robot such as “My name is Taro, register”.

  Further, for example, when the information “speech recognition“ wananoname and voice level OK, no corresponding word ”is displayed, the user has not correctly recognized the user's instruction by the robot apparatus 101. You can see that you are not responding. In this case, the user can predict that the robot apparatus 101 will respond by rephrasing “call the name” with a clear tone.

  Each component of the robot apparatus 101 described in the above embodiments includes a CPU, a memory of any computer, a program that realizes the components shown in the figure loaded in the memory, and a storage unit such as a hard disk that stores the program. It is realized by an arbitrary combination of hardware and software, centering on a network connection interface. It will be understood by those skilled in the art that there are various modifications to the implementation method and apparatus. Each of the drawings described in the embodiments shows a functional unit block, not a hardware unit configuration.

  As mentioned above, although embodiment of this invention was described with reference to drawings, these are the illustrations of this invention, Various structures other than the above are also employable.

  In the above embodiment, as shown in FIG. 1, an example in which the projection unit 15 is provided on the front surface of the robot apparatus 101 has been described. However, in another example, for example, the projection unit 15 is disposed on the back surface of the robot apparatus 101. The thinking process of the robot apparatus 101 can be displayed behind the robot apparatus 101. In this case, a CCD camera or the like for screen detection may be provided on the back surface of the robot apparatus 101. Accordingly, the user can see the thinking process of the robot apparatus 101 on the screen while facing the robot apparatus 101 and interacting with the robot apparatus 101, and can smoothly communicate with the robot apparatus 101. The location of the projection unit 15 can be set as appropriate. For example, the projection unit 15 can be provided at a location where the CCD camera 21A or the CCD camera 21B is provided in FIG.

The present invention also includes the following aspects.
(1) A robot comprising a thought process presentation method management means for managing a method for presenting a process for determining an action by means such as action, utterance, and display.

(2) an external state input means for inputting an external state;
External state management means for managing information obtained by the external state input means;
Internal state management means for managing an internal state that changes based on at least the external state;
A thinking means for determining an action based on the information on the external state and the internal state;
A thought process presentation method managing means for managing a method for presenting a process for determining an action by means such as action, utterance, and display;
Robot output means for outputting to the outside of the robot such as action, utterance, display,
A robot characterized by comprising

(3) The thought process presentation method management means includes:
The method of presenting a thought process by any one of the method of uttering with one's own voice, the method of uttering with the voice of a third party, or the method of sounding with a sound effect, that is, pronunciation and utterance (above) The robot according to 1) or (2).

(4) The thought process presentation method management means includes:
(1) or (2), wherein the thinking process is presented by any one of a method of moving a hand, a foot, an ear, or the like, or a method of turning on a lamp such as an eye or a mouth, that is, a robot operation. The robot described in).

(5) The thought process presentation method management means includes:
(1) or (2), wherein the thought process is presented by any one of a graph structure, text, table, time series chart, camera image, and robot diagram, that is, a screen display. robot.

(6) The thought process presentation method management means includes:
The robot according to (1) or (2), wherein a thinking process is presented by the pronunciation / speech and the robot motion.

(7) The thought process presentation method management means includes:
The robot according to (1) or (2), wherein a thinking process is presented by the pronunciation / speech and the screen display.

(8) The thought process presentation method management means includes:
The robot according to (1) or (2), wherein a thinking process is presented by the robot operation and the screen display.

(9) The thought process presentation method management means includes:
The robot according to any one of (1) to (8) above, wherein a thinking process is presented when a question is asked.

(10) A robot control method comprising: a thought process presentation method management step which is a means for managing a method for presenting a process for determining an action by means such as action, utterance, and display.

(11) an external state input step in which the robot apparatus inputs an external state;
An external state management step for managing information obtained by the external state input means;
An internal state management step for managing an internal state that changes based on at least the external state;
Thinking steps that determine actions based on external and internal information,
A thought process presentation method management step for managing a method for presenting a process for determining an action by means such as action, utterance, and display;
A robot output step for outputting to the outside of the robot such as actions, utterances, and displays;
A method for controlling a robot, comprising:

(12) The thought process presentation method management step includes:
The method of presenting a thought process by any one of the method of uttering with one's own voice, the method of uttering with the voice of a third party, or the method of sounding with a sound effect, that is, pronunciation and utterance (above) The robot control method according to 10) or (11).

(13) The thought process presentation method management step includes:
(10) or (11), wherein the thinking process is presented by any one of a method of moving a hand, a foot, an ear, or the like, or a method of turning on a lamp such as an eye or a mouth, that is, a robot operation. The robot control method described in (1).

(14) The thinking process presentation method management step is characterized by presenting a thinking process by any one method of graph structure, text, table, time series chart, camera image, and robot diagram, that is, screen display. The robot control method according to (10) or (11) above.

(15) The thought process presentation method management step includes:
The robot control method according to (10) or (11), wherein a thought process is presented by the pronunciation / speech and the robot motion.

(16) The thought process presentation method management step includes:
The robot control method according to (10) or (11), wherein a thought process is presented by the pronunciation / speech and the screen display.

(17) The thought process presentation method management step includes:
The robot control method according to (10) or (11), wherein a thought process is presented by the robot operation and the screen display.

(18) The thought process presentation method management step includes:
The robot control method according to any one of (10) to (17), wherein a thought process is presented when a question is asked.

(19) A computer software program written to execute control of a robot on a computer system,
A program for controlling a robot, comprising: a thought process presentation method management step for managing a method for presenting a process for determining an action by means such as action, utterance, and display.

(20) A thinking device comprising a thinking process presentation method managing means that is a means for managing a method for presenting a thinking process.

(21) state management means that is means for managing the state;
Based on the state, a thinking means that is a means for determining an action;
A thought process presentation method management means which is a means for managing the method of presenting the thought process;
A thinking device characterized by comprising:

(22) The thought process presentation method management means includes:
The method of presenting a thought process by any one of the method of uttering with one's own voice, the method of uttering with the voice of a third party, or the method of sounding with a sound effect, that is, pronunciation and utterance (above) The thinking device according to 20) or (21).

(23) The thought process presentation method management means includes:
(20) or (21), wherein the thinking process is presented by any one of a method of moving a hand, a foot, an ear, or the like, or a method of turning on a lamp such as an eye or a mouth, that is, a robot operation. ) Thinking device.

(24) The thought process presentation method management means includes:
The thinking device according to (20) or (21), wherein the thinking process is presented by any one of a graph structure, a text, a table, a time series chart, a camera image, and a robot diagram, that is, a screen display.

(25) The thought process presentation method management means includes:
The thinking device according to (20) or (21), wherein a thinking process is presented by the pronunciation / speech and the robot motion.

(26) The thought process presentation method management means includes:
The thinking device according to claim 20 or (21), wherein a thinking process is presented by the pronunciation / speech and the screen display.

(27) The thought process presentation method management means includes:
The thinking device according to claim 20 or (21), wherein a thinking process is presented by the robot operation and the screen display.

(28) The thought process presentation method management means includes:
The thinking device according to any one of (20) to (27), wherein a thinking process is presented when a question is asked.

It is an external appearance block diagram which shows an example of the robot apparatus in embodiment. It is a block diagram which shows the structure of the robot apparatus in embodiment. It is a figure which shows an example of a part of internal structure of an internal state memory | storage part. It is a figure which shows an example of the Bayesian network for determining the action which a robot apparatus should take. FIG. 5 is a diagram illustrating a process in which an action to be taken next by the robot apparatus is selected by the Bayesian network illustrated in FIG. 4. It is a figure which shows an example of the neural network for selecting the action which a robot apparatus should take. It is a figure which shows the thinking process of a robot apparatus with a class diagram. It is a figure which shows the thinking process of a robot apparatus with a state chart figure. It is a figure which shows the thinking process of a robot apparatus with a time series chart. It is a figure which shows the thinking process of a robot apparatus with a camera image. It is a figure which shows the thinking process of a robot apparatus with a robot figure. It is a figure which shows combining the image display method mentioned above. It is a figure which shows typically the example which projects the image which shows the thinking process of a robot apparatus toward the screen from the projection part of a robot apparatus. It is a screen which shows the screen which shows an example of the thought process of a robot apparatus. It is a screen which shows the screen which shows an example of the thought process of a robot apparatus. It is a screen which shows the screen which shows an example of the thought process of a robot apparatus. It is a flowchart which shows the process sequence which presents the thinking process of a robot apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Body part 2 Head 3A Wheel 3B Wheel 12 Speaker 13 Voice recognition microphone 14 Head switch 15 Projection part 21A CCD camera 21B CCD camera 101 Robot apparatus 102 External state input part 103 External state management part 104 Internal state management part 105 Thinking part 106 Thought process presentation processing unit 107 internal state storage unit 108 internal state acquisition unit 109 action determination unit 110 output processing unit 111 action unit 112 voice output unit 113 display unit 114 external output unit 115 processing procedure storage unit 200 screen 201 image

Claims (13)

A robot that performs thoughts and acts based on the thoughts,
External information acquisition means for acquiring external information related to the external state of the robot;
An external state management means for holding the acquired external information;
Internal state management means for holding internal information about the internal state of the robot that changes with the acquired external information as a trigger;
Thinking means that develops thinking with at least the external information as a trigger , and determines an action to be taken based on the result of the thinking;
And an output means for outputting the information consisting of the external information and the internal information such as the basis for thinking over extent itself and the thought process by the thinking means,
The thought process is multistage,
The output means displays on the screen at least a part of the thinking process in multiple stages as at least a part of the information to be output . The robot according to claim 1, wherein
The output means includes a plurality of output units for outputting the information,
An output method determining means for determining an output method of the information representing the thought process;
The output method determining means determines the output method of the thinking process in consideration of the external information or the internal information, and selects at least one from the plurality of output units,
A robot in which the information is output by the selected output unit. The robot according to claim 1 or 2 ,
Before SL thinking means, robot and performs thinking based on the external information and the internal information. The robot according to any one of claims 1 to 3 ,
The output means outputs at least a part of the external information or the internal information as the information representing a thought process. The robot according to any one of claims 1 to 4 , wherein
The external information includes information issued from a user, and the output method determining means determines an output method of the information representing a thought process in consideration of the information issued from the user. The robot according to any one of claims 1 to 5 ,
The output means includes a plurality of output units that output information via robot behavior, robot utterance or display,
The output method determining means selects any of the plurality of output units,
A robot configured to output the information representing a thought process through an output unit selected by the output method determining means. The robot according to any one of claims 1 to 6 ,
The external information includes information related to noise in the surrounding environment, and the output method determining means considers the information related to noise, and if the surroundings are noisy, the screen display is used as the output method, and the surroundings should be quiet. A robot characterized in that speech or pronunciation is used as the output method. The robot according to any one of claims 1 to 7 ,
The external information includes information regarding the presence or absence of a location where an image can be projected,
The said output method determination means uses the screen display to the said location as the said output method, when the said location which projects an image exists. The robot according to any one of claims 1 to 8 ,
The external information includes information regarding the user's age group,
The output method determining means determines the output method in consideration of a user's age group. A method of controlling a robot, wherein the robot
A first step of acquiring external information relating to the external state of the robot;
A second step of holding the acquired external information;
A third step of holding internal information about the internal state of the robot that changes with the acquired external information as a trigger;
It performed Thinking in response at least the external information, a fourth step based on the result of the thinking, determines the action to be taken,
A fifth step of outputting the information consisting of the external information and the internal information such as the basis of the fourth degree thinking over in step itself and the thought process, is executed,
The thought process is multistage,
In the fifth step, at least a part of the thinking process in multiple stages is displayed on a screen as at least a part of the information to be output . The robot control method according to claim 10 , wherein
The robot control method according to the fourth step , wherein control is performed so as to think based on the external information and the internal information. A program for controlling a robot,
A first step of acquiring external information relating to the external state of the robot;
A second step of holding the acquired external information;
A third step of holding internal information about the internal state of the robot that changes with the acquired external information as a trigger;
Performed thought as a trigger at least said external information, a fourth step based on the result of the thinking, determines the action to be taken,
A fifth step of outputting the information consisting of the external information and the internal information such as the basis of the fourth degree thinking over in step itself and the thought process, is executed,
The thought process is multistage,
In the fifth step, at least a part of the thinking process in multiple stages is displayed on a screen as at least a part of the information to be output . The robot control program according to claim 12 ,
The robot control program according to the fourth step , wherein control is performed so as to think based on the external information and the internal information.

Images