JP4556024B2 - Robot and robot system - Google Patents

Description

  The present invention relates to a robot and a robot system for collecting garbage.

In recent years, the robot-related industry market has grown greatly. In particular, the robot market in the household and welfare fields such as housework and nursing care is expected to expand further in the future, and there is a life support robot having a function according to the application such as a cleaning robot (for example, Patent Document 1). Various proposals and developments have been made.
JP 2003-299601 A

  Development of a robot that collects garbage as a life support robot is also desired. In developing such life support robots, it has been one of the important issues to minimize human assistance as much as possible while the robot achieves its purpose. In other words, it is considered that the existence value of a robot is to make a person's life convenient, and development of an autonomous robot having various functions for increasing its value has been advanced.

  In a modern society where technological competition is intense, when a robot having an excellent function is developed, development of a robot having a function exceeding that is immediately advanced. When a new robot having a further superior function is developed, the interest of the person is directed to the new robot, and the existence value of the previous robot is faded and discarded.

  The technological development cycle as described above is inevitable in the modern society of mass production and consumption.

  However, in order to build a recycling-oriented society with a low environmental load, we must move away from the above cycle. To that end, it is important to develop robots that are not discarded. Therefore, the present inventor considered the problems of the conventional robot as follows.

  A conventional robot must be convenient for a person as described above, and the relationship between the person and the robot is that used by the user (person) and the person used (robot). In such a relationship, the robot exists only to respond to one-sided requests from a person, and the person evaluates the existence value of the robot based on the convenience, and therefore, a robot that is excellent in convenience is required. For this reason, when a robot having a new function is developed, a robot having a lower function that has been used until then is discarded.

  As a method for solving such a problem, for example, by providing the life support robot as described above with a function for performing various gestures as provided by a pet robot that is currently commercially available, It is possible to give people attachment to robots.

  However, when people interact with pet-type robots, they feel lonely for some reason. One of the reasons is that the robot's behavior is so autonomous that it is not intended for human existence. As a result, the social connection between people and robots will not be felt.

  The other is that humans recognize robots as social beings, but robots do not recognize humans as social beings. For robots, humans are one of the same obstacles as stones. Only. Thus, the relationship between the person and the robot has asymmetry.

  When considering coexistence between humans and robots, it is necessary to reexamine the premise that robots should be autonomous. In this way, unless the relationship between humans and robots is reexamined, even if the life support robot has a pet-type robot function, humans and robots are those used by people (robots). ) Can not get out of the relationship.

  An object of the present invention is to provide a robot and a robot system that can achieve the purpose of collecting garbage by a joint action with a person and can establish an interdependent and reciprocal relationship with the person.

According to a first aspect of the present invention, there is provided a main body having a dust container, an expression means for expressing intention by at least one of an utterance by voice output, display of a facial expression, and movement of the main body, A first detection means for detecting a person, a second detection means for detecting that dust has been put into the dust container , and a control means for receiving the detection result of the first detection means and controlling the expression means. And the control means executes an action for inducing a person's action of picking up trash by the expression means when a person is detected by the first detection means, and an action of the person picking up the trash by the expression means When the second detection means detects that the garbage has been thrown into the garbage container after performing the action for inducing the user, the expression means executes a response action to the person, and the expression means Picking up trash In order to induce the person's act of picking up trash by the expression means when the second detection means does not detect that the trash has been thrown into the trash container after performing the action for inducing the act of Other actions are selectively performed .

  In the robot according to the present invention, when a person is detected by the first detection means, an action for inducing a person's action of picking up trash is executed by the expression means. This triggers the person's act of picking up trash.

  In this way, the robot cannot pick up trash itself, but can only collect trash with the help of others. Therefore, the purpose of collecting trash as a result of joint and reciprocal actions between robots and people is achieved.

  In this case, the robot is valued as an entity that realizes a function of collecting garbage by obtaining human assistance. At the same time, people are valued as beings that help robots. As a result, humans and robots mutually value each other and compose the meaning of existence. That is, an interdependent and reciprocal relationship between humans and robots is established. As a result, attachment to robots is fostered and social connections are formed.

In the control unit, when after the action to induce the act of people who pick up trash is executed by means of expression, that garbage thrown into the garbage receiving portion detected by the second detecting means The responding action is executed on the person by the expression means.

  In this case, when the second detection means detects that the garbage has been thrown into the garbage container, a response action is performed on the person by the expression means. As a result, people can find value in the act of picking up trash.

When the control means, which after the act to elicit the act of people who pick up trash is executed by representable means that dust in the dust receiving portion is turned not detected by the second detecting means, other acts are performed selectively for inducing act of people who pick up trash by expression means.

  In this case, if a person does not pick up the trash, by performing various actions to induce the person's act of picking up the trash, trials and errors are performed to allow the person to pick up the trash. Can learn. As a result, the robot can acquire an action for reliably pulling out the intentional posture of putting the garbage into the garbage storage unit from the person. As a result, the interdependence and mutual constitutional relationship between the person and the robot gradually builds up as the person is involved with the robot, and the attachment to the robot increases and the social connection is strengthened.

According to a second aspect of the present invention, there is provided a main body having a dust container, an expression means for expressing an intention by at least one of an utterance by voice output, display of a facial expression, and movement of the main body, first detecting means for detecting a human, detection result receiving the first detection means, and control means for controlling the expression means, the control means for inducing the act of people who pick up trash A plurality of actions and selection probabilities for each action are stored, and when a person is detected by the first detection means, one of the plurality of actions is selected at random reflecting the selection probability, and the expression means After executing the selected action, the selection probability of the selected action is updated based on whether or not garbage is thrown into the garbage container .

In this robot, a plurality of actions and the selection probability of each action are stored. Any one of the plurality of stored actions is selected at random reflecting the selection probability, and the selected action is executed. The selection probability of the action selected based on whether or not garbage is thrown into the garbage container is updated. As a result, the robot can learn an action for a person to pick up trash. As a result, the interdependence and mutual constitutional relationship between the person and the robot gradually builds up as the person is involved with the robot, and the attachment to the robot increases and the social connection is strengthened.

  The expression means may include at least one of a voice output means for speaking by voice output, a display means for expressing a facial expression, and a drive means for moving the main body.

  In this case, the intention of the robot is expressed by the utterance by the voice output by the voice output means, the display of the facial expression by the display means, or the movement of the main body by the drive means.

A robot system according to a third aspect of the present invention includes a plurality of robots and a control unit that controls each of the plurality of robots, and each of the plurality of robots includes a main body unit having a dust container, a speech by voice output, Expression means for expressing intention by at least one action of display of facial expression and movement of main body, first detection means for detecting a person around the main body, and detection that garbage is thrown into the dust container and a second detecting means for the control means, when a person is detected by either the first detection means of the plurality of robots, the expression means of operation and a plurality of robots means of expression of each robot by at least one of operation of the federated operation running acts to induce the act of people who pick up trash, after performing the acts for inducing act of people who pick up trash, garbage When the second detecting means detects that dust has been thrown into the container, it is possible for a person to act by at least one of the actions of the expression means of each robot and the action of the expression means of a plurality of robots in cooperation. If the second detecting means does not detect that the trash has been thrown into the trash container after executing the action for inducing the person's action of picking up the trash in response to the robot, each robot The other action for inducing the person's action of picking up trash by at least one of the movement of the expression means and the coordinated action of the expression means of the plurality of robots is selectively executed .

  In the robot system according to the present invention, when a person is detected by the first detection means of any of the plurality of robots, the operation of the expression means of each robot and / or the operation of the expression means of the plurality of robots in cooperation with each other An action for inducing a person's action of picking up trash is performed. This triggers the person's act of picking up trash. In particular, the action of a person trying to pick up garbage is effectively induced by the coordinated movement of a plurality of robots.

  In this way, each robot cannot pick up trash itself and can collect trash only with the help of others. Therefore, the purpose of collecting trash as a result of joint and reciprocal actions between the robot system and people is achieved.

  In this case, the robot system is valued as an entity that realizes a function of collecting garbage by obtaining human assistance. At the same time, people are valued as beings that help robot systems. As a result, humans and robot systems mutually value each other and compose the meaning of existence. That is, an interdependent and reciprocal relationship between a person and a robot system is established. As a result, attachment to the robot system is fostered and social connection is formed.

In the control means, after an action for inducing a person's action to pick up garbage by at least one of the actions of the expression means of each robot and the linked actions of the expression means of a plurality of robots, If the be garbage garbage receiving portion is turned detected by the second detecting means, responsive to a person in cooperation with operation of the operation and / or a plurality of robots of expression means expression means of each robot act is performed.

  In this case, when it is detected by the second detection means that the garbage has been put into the garbage container, the operation of the expression means of each robot and / or the cooperation of the expression means of a plurality of robots is performed on the person. The response action is executed. As a result, people can find value in the act of picking up trash.

In the control means, after the action for inducing an act of people who pick up garbage is executed, if that garbage thrown into the garbage receiving portion is not detected by the second detection means, for each robot other acts to induce the act of people who pick up trash in collaboration with operation of the operation and / or a plurality of robots of expression means the expression means is selectively executed.

  In this case, if a person does not pick up the trash, by performing various actions to induce the person's act of picking up the trash, trials and errors are performed to allow the person to pick up the trash. Can learn. As a result, the robot system can acquire an action for reliably pulling out the intentional posture of throwing trash into the trash storage unit from the person. As a result, the interdependence and mutual constitutional relationship between the person and the robot system is gradually established as the person repeats the relationship with the robot system, and the attachment to the person increases and the social connection is strengthened.

A robot system according to a fourth aspect of the present invention includes a plurality of robots and a control unit that controls each of the plurality of robots, and each of the plurality of robots includes a main body unit having a dust container, a speech by voice output, An expression means for expressing an intention by at least one of an action of displaying a facial expression and a movement of the main body, and a first detection means for detecting a person around the main body, and the control means is for a person who picks up trash A plurality of actions for inducing an action and a selection probability of each action are stored, and when a person is detected by any of the first detection means of the plurality of robots, any one of the plurality of actions is randomly selected. Reflecting the selection probability, and performing an action selected by at least one of the operations of the expression means of each robot and the operations of the expression means of the plurality of robots, In which dust or trash receiving portion bot updates the selection probability of action selected on the basis of whether it is turned on.
In this robot system, a plurality of actions and the selection probability of each action are stored. Any one of the plurality of stored actions is selected at random reflecting the selection probability, and the selected action is executed. The selection probability of the action selected based on whether or not garbage is thrown into the garbage container is updated. As a result, the robot can learn an action for a person to pick up trash. As a result, the interdependence and mutual constitutional relationship between the person and the robot gradually builds up as the person is involved with the robot, and the attachment to the robot increases and the social connection is strengthened.
The expression means may include at least one of a voice output means for speaking by voice output, a display means for expressing a facial expression, and a drive means for moving the main body.

  In this case, the intention of the robot is expressed by the utterance by the voice output by the voice output means, the display of the facial expression by the display means, or the movement of the main body by the drive means.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to achieve the objective of collect | recovering garbage by the joint action with a person, and to build the interdependence and mutual constitutional relationship with a person. As a result, attachment to robots is fostered and social connections are formed.

  FIG. 1 is a schematic diagram showing an appearance of a trash can robot according to an embodiment of the present invention.

  As shown in FIG. 1, the trash can robot 100 includes a cylindrical main body 1 and a disk-shaped support 2. The main body 1 is provided on the support 2 so as to be rotatable around a vertical axis. The inside of the main body 1 is a dust container 30.

  A display 11 for displaying an image is provided on the outer peripheral surface of the main body 1. The display 11 is a liquid crystal display, for example. A face is displayed on the display 11. The trash can robot 100 can convey various emotions to a person by changing the facial expression displayed on the display 11.

  Microphones 12 and 13 are provided on both sides of the display 11. These microphones 12 and 13 can input voices of people around the trash can robot 100. The microphones 12 and 13 function as stereo microphones. The sound source direction can be recognized based on the phase difference of the sound input to the microphones 12 and 13.

  A speaker 14 is provided at the center of the outer peripheral surface of the main body 1. The trash can robot can speak to a person by outputting sound from the speaker 14.

  A person discrimination camera 15 is provided at the upper part of the outer peripheral surface of the main body 1. The person discrimination camera 15 is, for example, a CCD (charge coupled device) camera. This person discrimination camera 15 images a person existing in front of the trash can robot 100. The position of the person in front of the trash can robot 100 can be detected by the face area search technique based on the image signal obtained by the person discrimination camera 15.

  A dust discrimination camera 16 is provided near the upper end of the inner peripheral surface of the main body 1. The dust discrimination camera 16 is, for example, a CCD camera. The dust discrimination camera 16 images an object such as dust thrown into the dust container 30 of the main body 1. The type of dust thrown into the dust container 30 can be determined from the image signal obtained by the dust determination camera 16 by image recognition technology.

  A plurality of object detection sensors 19 for detecting a person are provided below the outer peripheral surface of the main body 1. The object detection sensor 19 includes a pyroelectric sensor, an infrared distance measuring sensor, and the like. Based on the output signal of the object detection sensor 19, the presence of a person or an obstacle around the trash can robot 100 can be detected.

  Moving wheels 17 and 18 are provided on the lower surface of the support portion 2. The wheels 17 and 18 are driven by a moving motor 24 (FIG. 2), which will be described later, built in the support portion 2. The trash can robot 100 can move in all directions by the wheels 17 and 18 while avoiding an obstacle based on the output signal of the object detection sensor 19. Thereby, an autonomous movement mechanism is realized.

  In addition, the trash can robot 100 can rotate the main body 1 on the support 2 by a main body rotation motor 23 (FIG. 2) described later built in the support 2. Thereby, the trash can robot 100 can express an emotion or a request by the movement of the main body 1.

  FIG. 2 is a block diagram showing a control system of the trash can robot 100 of FIG.

  The trash can robot 100 includes a control computer 20, a speech recognition device 21, a speech synthesizer 22, a main body rotation motor 23, and a movement motor 24.

  Voice signals obtained by the microphones 12 and 13 are input to the voice recognition device 21. The voice recognition device 21 performs voice recognition based on the voice signals input from the microphones 12 and 13. The recognition result of the voice recognition device 21, the image signal obtained by the person discrimination camera 15, the image signal obtained by the dust discrimination camera 16 and the output signal of the object detection sensor 19 are given to the control computer 20.

  The control computer 20 includes a CPU (Central Processing Unit), a memory, and the like. The computer 20 for control is based on the recognition result of the voice recognition device 21, the image signal obtained by the person discrimination camera 15, and the output signal of the object detection sensor 19. Determine the position of. Further, the control computer 20 determines the type of dust thrown into the dust container 30 based on the image signal obtained by the dust determination camera 16. Further, the control computer 20 displays a face on the display 11. Further, the control computer 20 synthesizes a voice signal using the voice synthesizer 22 and outputs the voice signal from the speaker 14 as a voice.

  The control computer 20 also controls the main body rotation motor 23 and the movement motor 24. When the main body rotating motor 23 is operated, the main body 1 in FIG. 1 rotates with respect to the support 2. Thereby, the main body 1 performs a swinging motion. When the moving motor 24 operates, the wheels 17 and 18 in FIG. 1 rotate. The wheels 17 and 18 can be directed in any direction by the moving motor 24. As a result, the trash can robot 100 moves in all directions.

  A control program is stored in the memory of the control computer 20. The control computer 20 executes processing described later according to the control program.

  FIG. 3 is a flowchart showing the operation of the trash can robot 100 of FIG. The trash can robot 100 operates according to a control program stored in the memory of the control computer 20.

  Here, a plurality of actions of the trash can robot 100 are stored in the memory of the control computer 20 in advance. The actions of the trash can robot 100 are movement of the main body 1, utterance by the speaker 14, and facial expression displayed on the display 11. Specifically, a plurality of movement patterns of the main body 1, a plurality of utterance contents by the speaker 14, and a plurality of facial expressions displayed on the display 11 are stored in advance in the memory as a plurality of actions. Each of the plurality of actions is given a selection probability. In the initial state, the selection probabilities for a plurality of actions are equal.

  In addition, the control computer 20 uses the recognition result of the voice recognition device 21, the image signal obtained by the person discrimination camera 15, and the output signal of the object detection sensor 19, so that the control computer 20 Here, a person is detected. Here, a person around the trash can robot 100 is detected based on an output signal of the object detection sensor 19.

  First, the control computer 20 of the trash can robot 100 detects a person around the main body 1 by the object detection sensor 19 (step S1). At this time, for example, the main body rotation motor 23 may rotate the main body 1 to the left and right by a predetermined angle so that the entire periphery can be viewed. When no person is detected around the main body 1 (step S2), the trash can robot 100 is moved by the moving motor 24 (step S10), and the process returns to step S1. This movement is performed so that the direction changes at random like a person walking around.

  When a person is detected around the main body 1 (step S2), the main body 1 is directed toward the person by the main body rotation motor 23 (step S3).

  The control computer 20 randomly selects one of a plurality of prestored actions using the random number table (step S4). Here, a random selection reflecting the selection probability is performed, but as described above, the selection probabilities of a plurality of actions are equal in the initial state.

  Next, the control computer 20 controls the display 11, the speech synthesizer 22, the speaker 14, the main body rotation motor 23, and the movement motor 24 so as to execute the selected action (step S5).

  For example, the main body rotating motor 23 moves the main body 1 to the left or right while moving the main body 1 so as to approach a person by the moving motor 24. In addition, the display 11 represents a familiar expression. Further, the speaker 14 utters a greeting and utters that he / she wants to pick up trash.

  Next, the control computer 20 determines whether or not desired dust has been thrown into the dust container 30 based on the image obtained by the dust discrimination camera 16 (step S6). For example, when the garbage container 30 is for empty cans, the control computer 20 determines whether or not the input garbage is empty cans. Whether or not the empty can has been inserted is determined by the control computer 20 storing in advance the shapes corresponding to various states such as the direction of the empty can, and the image of the object that has been input to the dust container 30 and the stored shape. This can be done by using a pattern recognition technique such as matching.

  When the desired garbage is thrown into the garbage container 30, the control computer 20 directs the main body 1 toward the person by the main body rotation motor 23 (step S7). Then, the control computer 20 expresses gratitude by the movement of the main body 1, the utterance by the speaker 14, and the facial expression displayed on the display 11 (step S8).

  For example, the main body 1 is rotated left and right by the main body rotation motor 23. In addition, the display 11 represents a joyful expression. Furthermore, a thank you word is uttered by the speaker 14.

  Thereafter, the control computer 20 increases the selection probability of the selected action (step S9) and returns to step S1.

  If no garbage is thrown into the garbage container 30 in step S6, the control computer 20 returns to step S1 and repeats the processes in steps S1 to S6. In this case, the control computer 20 randomly selects another action among the plurality of actions stored in advance in step S4.

  By repeating the processes of steps S1 to S10, the selection probabilities of a plurality of actions stored in the memory of the control computer 20 change. Thereby, an action having a high selection probability is easily selected in step S4.

  In this case, since the selection probability of an action when a desired garbage is thrown into the garbage container 30 is increased, the control computer 20 acquires an action for allowing a person to throw in the desired garbage by learning. .

  In the present embodiment, in order to efficiently select an action of the trash can robot 100 for picking up trash, a reward value is given to an action selected when a person puts a desired trash into the trash container 30. And the selection probability of each action is weighted according to the magnitude of the reward value acquired by each action. A so-called roulette selection method can be used to change the selection probability of a plurality of actions. Here, the concept of the roulette selection method will be described.

  The area of the roulette circumference is divided into a desired number of columns, and one of a plurality of columns is selected by a random number. The width of each column determines the probability of that column being selected.

  Assign one of several actions to each column. In the initial stage, by making the width of each column equal for all actions, any of a plurality of actions can be selected with the same probability (equal selection probability). Then, experience (learning) by trial and error is reflected in the selection probability.

  For example, when a person throws in desired garbage into the garbage container 30 by performing any of a plurality of actions, the width of the corresponding column of the roulette is increased, and the person removes the desired garbage into the garbage container 30. If it is not inserted into the box, the width of the corresponding column of the roulette is reduced. As a result, among the plurality of actions, the probability of selecting an action performed when a person throws the desired garbage into the garbage container 30 increases, and the person does not throw the desired garbage into the garbage container 30. In some cases, the probability of selecting an action performed is low.

  The degree to which the width of the corresponding column of the roulette is increased when a person throws in desired garbage into the garbage container 30 may be determined using a mathematical method called a Q-learning method, or an empirical rule You may decide by.

  As described above, in the trash can robot 100 according to the present embodiment, when a person is detected around the main body 1, the movement of the main body 1, the utterance by the speaker 14, and the expression on the display 11 are displayed. The act of inducing a person's act of picking up trash is performed.

  People try to read the mind of what moves in this way. Such a cognitive stance is called cognitive philosopher Dennett "Intentional stance". With this intentional attitude, the person recognizes that the trash can robot 100 wants to pick up trash or wants to pick up trash. The action of the trash can robot 100 as described above brings about the above recognition based on the person's intentional posture, and induces the person's action to pick up trash. That is, the above action of the trash can robot 100 is performed to induce a person's action of picking up trash.

  When desired garbage is thrown into the garbage container 30, a gratitude is conveyed to the person by the movement of the main body 1, the utterance by the speaker 14, and the expression of the expression by the display 11. As a result, people can find value in the act of picking up trash.

  Thus, the trash can robot 100 cannot pick up trash itself, but can collect trash only with the help of others. Therefore, the purpose of collecting trash as a result of the joint action and reciprocal action between the trash can robot 100 and the person is achieved.

  In this case, the trash can robot 100 is valued as an entity that realizes a function of collecting trash by obtaining human assistance. At the same time, a person is valued as a person who helps the trash can robot 100. Thereby, the person and the trash can robot 100 mutually value each other and compose the meaning of existence. That is, an interdependent and reciprocal relationship between the person and the trash can robot 100 is established. As a result, attachment to the trash can robot 100 is fostered and a social connection is formed.

  Further, the control computer 20 randomly selects any one of the plurality of stored actions reflecting the selection probability, and executes the selected action. When the desired garbage is not thrown into the garbage container 30, another action is selectively executed, and when the desired garbage is thrown into the garbage container 30, the selection probability of the selected action is increased. Let Thereby, the trash can robot 100 can learn an action for having a person pick up trash. As a result, an interdependent and reciprocal relationship between the person and the trash can robot 100 gradually builds up as the person becomes more involved with the trash can robot 100, and the attachment to the trash can robot 100 increases and social connection. Will be strengthened.

  In the above embodiment, the roulette selection method has been described as a learning method for efficiently selecting an action for allowing a person to pick up garbage. However, the present invention is not limited to this, and various learning methods are used. be able to. For example, a general genetic algorithm, a Q-learning method, or the like can be used as a method of learning using a teacher signal that a person has picked up garbage.

  It is also possible to obtain in advance data obtained by weighting the action selection probability by such a learning method, and select one of a plurality of actions by combining this data with a random number or priority.

  In the above embodiment, the trash can robot 100 moves the main body 1, speaks with the speaker 14, and displays the facial expression on the display 11 as an action for the trash robot 100 to pick up trash. If it is an action that can lead to a person's action of picking up trash by recognizing that the person wants to pick up trash or wants to pick up trash, the movement of the main body 1, the speaker Any one or two of the utterance by 14 and the display of the facial expression by the display 11 may be performed. For example, only the utterance by the speaker 14 and the display of the facial expression by the display 11 may be performed.

  Furthermore, in the above-described embodiment, when desired garbage is thrown into the garbage container 30, the movement of the main body 1, the utterance by the speaker 14, and the expression on the display 11 are displayed as a response action. The act of expressing gratitude to the person, the act of expressing joy, or the act of expressing other intentions by any one or two of the movement of the main body 1, the utterance by the speaker 14 and the display of the expression on the display 11 Also good.

  For example, one or two of the act of rotating the main body 1 left and right by the main body rotation motor 23, the act of expressing a joyful expression by the display 11, and the act of uttering a thank-you word by the speaker 14 are performed. Also good.

  Next, a robot system according to another embodiment of the present invention will be described. FIG. 4 is a block diagram showing a configuration of a trash can robot in a robot system according to another embodiment of the present invention. 5 to 9 are schematic diagrams for explaining an example of the configuration and operation of a robot system according to another embodiment of the present invention.

  The robot system shown in FIGS. 5 to 9 includes a plurality of trash can robots 100a, 100b, and 100c and a general control computer 200. The robot system of this example includes a trash can robot 100a for empty cans, a trash can robot 100b for burning trash, and a trash can robot 100c for non-burnable trash.

  The trash can robot 100a in FIG. 4 differs from the trash can robot 100 in FIG. 2 in that a communication device 25 connected to the control computer 20 is further provided. The configuration of each of the other trash can robots 100b and 100c is the same as the configuration of the trash can robot 100a of FIG.

  The plurality of trash can robots 100a, 100b, and 100c in FIGS. 5 to 9 are connected to the overall control computer 20 by a wireless LAN (local area network).

  The plurality of trash can robots 100a, 100b, and 100c collect trash by operating in cooperation with the overall control of the overall control computer 200. Note that the control computer 20 for one or more of the trash can robots 100a, 100b, and 100c may control the plurality of trash can robots 100a, 100b, and 100c in an integrated manner.

  The overall control computer 200 communicates with the trash can robots 100a, 100b, and 100c, and maintains the relative arrangement between the plurality of trash can robots 100a, 100b, and 100c, for example, at a distance that does not collide with each other, and each trash can robot 100a. , 100b, 100c, for example, the orientation is controlled. In this case, the trash can robots 100a, 100b, and 100c measure the mutual distance based on the output signal of the object detection sensor 19 of FIG. The plurality of trash can robots 100 a, 100 b, and 100 c express a conversation between them by movement of the main body 1, utterance by the speaker 14, and facial expression displayed on the display 11.

  Here, an example of the operation of the robot system will be described with reference to FIGS.

  As shown in FIG. 5, a plurality of trash can robots 100a, 100b, and 100c have a conversation. When there are no people around the plurality of trash can robots 100a, 100b, 100c, as shown in FIG. 6, the plurality of trash can robots 100a, 100b, 100c move together while continuing the conversation.

  As shown in FIG. 7, when a plurality of trash can robots 100a, 100b, and 100c find a person, the plurality of trash can robots 100a, 100b, and 100c select one of a plurality of actions, and the selected main unit 1 The intention to have the user pick up trash is expressed by the movement, the utterance by the speaker 14, and the facial expression displayed on the display 11. In this case, the plurality of trash can robots 100a, 100b, and 100c want to pick up trash, but do not pick up the trash, but perform an act of recognizing that they need human help.

  As shown in FIG. 8, when a person throws in desired garbage into the garbage container 30 of the garbage box robot 100, the garbage box robot 100a determines whether or not the desired garbage has been thrown in. In this example, the trash can robot 100a determines whether or not the input trash is an empty can.

  When desired trash is thrown in, the plurality of trash can robots 100a, 100b, and 100c express gratitude by the movement of the main body unit 1, the utterance by the speaker 14, and the facial expression displayed on the display 11. As a result, as shown in FIG. 9, a person can find value in the act of picking up trash.

  As described above, in the robot system according to the present embodiment, when a person is detected around any of the plurality of trash can robots 100a, 100b, 100c, the main body of each trash can robot 100a, 100b, 100c. 1 action, the utterance by the speaker 14 and the expression of the expression on the display 11 and the action of pulling out the intentional attitude of putting trash into the trash storage unit 30 by the coordinated movement of the plurality of trash can robots 100a, 100b, 100c. Executed. This triggers a person's action to pick up trash. In particular, an action of a person trying to pick up trash is effectively induced by the coordinated movement of the plurality of trash can robots 100a, 100b, and 100c.

  When desired trash is put into the trash container 300, the movement of the main body 1 of each trash can robot 100a, 100b, 100c, the utterance by the speaker 14, the display of the expression on the display 11, and the plurality of trash can robots 100a, 100b. , 100c, an action to convey thanks to the person is executed. As a result, people can find value in the act of picking up trash.

  Thus, each trash can robot 100a, 100b, 100c cannot pick up trash itself, but can collect trash only with the help of others. Therefore, the purpose of collecting trash as a result of joint and reciprocal actions between the robot system and people is achieved.

  In this case, the robot system is valued as an entity that realizes a function of collecting garbage by obtaining human assistance. At the same time, people are valued as beings that help robot systems. As a result, humans and robot systems mutually value each other and compose the meaning of existence. That is, an interdependent and reciprocal relationship between a person and a robot system is established. As a result, attachment to the robot system is fostered and social connection is formed.

  In the above embodiment, the main body 1 corresponds to the main body, and the speech synthesizer 22 and the speaker 14, the display 11, or the main body rotation motor 23, the movement motor 24, and the wheels 17 and 18 constitute expression means. The voice synthesizer 22 and the speaker 14 correspond to speech means, the display 11 corresponds to display means, and the main body rotation motor 23, the movement motor 24, and the wheels 17, 18 correspond to drive means. Further, the microphones 11 and 12, the person discrimination camera 15 or the object detection sensor 19 correspond to the first detection means, and the dust discrimination camera 16 corresponds to the second detection means. The computer 200 corresponds to the control means.

  By efficiently placing the trash can robot 100 or robot system according to the present embodiment in public places, parks, kindergartens, schools, etc. where people, kindergarteners, schoolchildren, etc. frequently pass, the intended purpose is made efficient. Can be achieved well.

  The present invention can be used to collect garbage indoors and outdoors.

It is a schematic diagram which shows the external appearance of the trash can robot which concerns on one embodiment of this invention. It is a block diagram which shows the control system of the trash box robot of FIG. It is a flowchart which shows operation | movement of the trash can robot of FIG. It is a block diagram which shows the structure of the trash can robot in the robot system which concerns on other embodiment of this invention. It is a schematic diagram for demonstrating an example of a structure and operation | movement of the robot system which concerns on other embodiment of this invention. It is a schematic diagram for demonstrating an example of a structure and operation | movement of the robot system which concerns on other embodiment of this invention. It is a schematic diagram for demonstrating an example of a structure and operation | movement of the robot system which concerns on other embodiment of this invention. It is a schematic diagram for demonstrating an example of a structure and operation | movement of the robot system which concerns on other embodiment of this invention. It is a schematic diagram for demonstrating an example of a structure and operation | movement of the robot system which concerns on other embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Main part 2 Support part 11 Display 12, 13 Microphone 14 Speaker 15 Person discrimination camera 16 Dust discrimination camera 17, 18 Wheel 19 Object detection sensor 20 Control computer 21 Voice recognition device 22 Speech synthesizer 23 Main body rotation motor 24 Moving motor 100, 100a, 100b, 100c Trash can robot 200 Computer for overall control

Claims (6)

A main body having a garbage container;
Expression means for expressing intention by at least one action of speech output, facial expression display and movement of the main body,
First detection means for detecting a person around the main body,
Second detecting means for detecting that dust is thrown into the dust container;
Control means for receiving the detection result of the first detection means and controlling the expression means;
The control means includes
An action for inducing a person's action of picking up trash by the expression means when a person is detected by the first detection means ;
The expression means when the second detection means detects that the garbage has been thrown into the garbage container after performing the action for inducing the person's action of picking up the garbage by the expression means. To respond to people by
When the second detection means does not detect that the garbage has been thrown into the garbage container after performing the action for inducing the person's action of picking up the garbage by the expression means, the expression means A robot that selectively performs other actions to induce a person's action to pick up garbage . A main body having a garbage container;
Expression means for expressing intention by at least one action of speech output, facial expression display and movement of the main body,
First detection means for detecting a person around the main body,
Control means for receiving the detection result of the first detection means and controlling the expression means;
The control means includes
Storing the selection probability of a plurality of behaviors and the action to induce the act of people who pick up garbage, if a person is detected by said first detection means, one of said plurality of behaviors Is selected reflecting the selection probability at random, and after the action selected by the expression means is executed, the selection probability of the selected action is updated based on whether or not garbage is thrown into the garbage container. features and to Carlo bot that. 3. The robot according to claim 1, wherein the expression means includes at least one of voice output means for uttering speech output, display means for expressing an expression, and drive means for moving the main body. With multiple robots,
Control means for controlling each of the plurality of robots,
Each of the plurality of robots is
A main body having a garbage container;
Expression means for expressing intention by at least one action of speech output, facial expression display and movement of the main body,
First detection means for detecting a person around the main body ,
Second detection means for detecting that dust is thrown into the dust container ,
The control means includes
When a person is detected by the first detection means of any of the plurality of robots , at least one of the operation of the expression means of each robot and the operation of the expression means of the plurality of robots in cooperation with each other Perform actions to induce the person to pick up trash by movement ,
After performing the act of inducing the person's act of picking up trash, when the second detection means detects that the trash has been thrown into the trash container, the expression means of each robot Performing a response action on a person by at least one of an operation and a coordinated operation of the expression means of the plurality of robots;
The operation of the expression means of each robot when the second detection means does not detect that garbage has been thrown into the garbage container after performing an action for inducing the person's action of picking up garbage And a robot system that selectively executes another action for inducing a person's action of picking up trash by at least one of the coordinated actions of the expression means of the plurality of robots . With multiple robots,
  Control means for controlling each of the plurality of robots,
  Each of the plurality of robots is
  A main body having a garbage container;
  Expression means for expressing intention by at least one action of speech output, facial expression display and movement of the main body,
  First detection means for detecting a person around the main body,
  The control means stores a plurality of actions for inducing a person's action to pick up garbage and a selection probability of each action, and a person is detected by the first detection means of any of the plurality of robots. And selecting one of the plurality of actions at random reflecting the selection probability, and at least one of the operation of the expression means of each robot and the operation of the expression means of the plurality of robots in cooperation with each other After the action selected by the operation is performed, the selection probability of the selected action is updated based on whether or not garbage is thrown into any one of the plurality of robots. Robot system. 6. The robot system according to claim 4 , wherein the expression means includes at least one of a voice output means for speaking by voice output, a display means for expressing an expression, and a drive means for moving the main body.

Images