JP2006178644A - Behavioral representation system and behavioral representation processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent a family member who watches for a robot representing the behavior of an elderly person from becoming too familiar or bored with the watching. <P>SOLUTION: Depending on the first result of detection by a first motion detecting means 401 that detects the motion of an elderly person, a robot 200 at the house of a family member of the elderly person is made to represent the behavior of the elderly person. The level of interest of the family member is estimated from second detection results obtained from second motion detecting means 105, 201 that detect the watching action of the family member. The result is fed back so as to vary the way in which the robot 200 represents the behavior. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a behavior expression system and a behavior expression processing device that are used when expressing the behavior of an object to be observed with respect to an observer, and in particular, when used to indirectly observe an object to be observed. Is preferred.

  With the aging society and the declining birthrate, life care for the elderly living alone has become a major social issue. Nowadays, it is said that life care support for the elderly living alone is necessary not only for separated families but also for the entire community including volunteers. Under such circumstances, expectation for a life monitoring system for elderly people using robots is increasing.

On the other hand, as a conventional monitoring system for elderly people, information on the presence, movement, environment, etc. of people is detected and transmitted, and this information is expressed as a specific expression on the receiving side. A communication device (patent document 1) for transmitting an emergency state to a close relative in the country, and detecting a living situation of a resident, and transmitting the detected result to a remote place via a communication network such as the Internet. There has been proposed a system (Patent Document 2) that transmits a living situation of a resident to a close relative in a remote place by operating a virtual flower or a robot.
JP 2003-67874 A JP200496630

  However, it is a tedious task for the watcher to constantly watch the life of the elderly, and perseverance is often required to keep watching. In addition, “habituation” and “boring” occur in the monotonous work of watching the movements of the above-mentioned virtual flowers and robots that reflect the living conditions of the elderly, and as a result, the original is to watch the lives of the elderly living alone. May defeat the purpose.

  Therefore, an object of the present invention is to provide an action expression system and an action expression processing device that reduce “habituation” and “satisfaction” with respect to monotonous work on the side of watching.

  The behavior expression system of the present invention includes a first motion detection unit that detects a motion of an object to be observed, a behavior expression unit that performs an operation corresponding to the motion of the object to be observed, and an observer's Second motion detection means for detecting motion, and control information for controlling the behavior expression unit, the first detection result obtained by the first motion detection means, and the second And a second detection result obtained by the motion detection means, and a control information generation means for obtaining from the second detection result.

  According to the present invention, when the action expression unit performs an action corresponding to the action of the observed object detected by the first action detecting means, the reaction of the observer who sees the action is detected by the second action detecting means. By detecting and feeding back to the control information for controlling the action expression unit, it is possible to reduce the feeling that the observer feels “familiar” or “tired” with respect to the monotonous work of watching.

  The control information generation unit may obtain the control information by adding changes to the original control information obtained from the first detection result based on the second detection result.

  Furthermore, the behavior expression system more preferably observes the motion of the behavior expression unit based on the second detection result, and motion estimation means for estimating the motion of the observed object based on the first detection result. Interest level estimation means for estimating the interest level of the observer, and the control information generation means estimates the motion of the observed object estimated by the motion estimation means and the interest level estimation means. The control information may be obtained from the degree of interest of the observer.

  In addition, the behavior expression system includes a transmission unit that transmits the first detection result, and a reception unit that receives the first detection result transmitted by the transmission unit, and the control information generation unit May obtain the control information from the first detection result received by the receiving means and the second detection result obtained by the second operation detecting means.

  Further, the control information generation means obtains the control information by adding a change to the original control information obtained from the first detection result received by the receiving means based on the second detection result. May be.

  In addition, the behavior expression system includes a transmission unit that transmits the first detection result, and a reception unit that receives the first detection result transmitted by the transmission unit. The operation of the object to be observed may be estimated based on the first detection result received by the receiving unit.

  Further, the behavior expression system more preferably includes transmission means for transmitting information on the motion of the observed object estimated by the motion estimation means, and information on the motion of the observed object transmitted by the transmitting means. Receiving the receiving means, the control information generating means from the information about the operation of the object received by the receiving means, and the interest level of the observer estimated by the interest level estimation means The control information may be obtained.

  As described above, the first detection result is transmitted from the observed object side by the transmission unit, so that it is possible to watch the remotely observed object.

  The behavior expression system includes storage means for storing the control information for controlling the behavior expression unit and the second detection result related to the observer detected by the second motion detection means in association with each other. The control information generation means may obtain the original control information from the first detection result and information stored in the storage means.

  The behavior representation system stores the control information for controlling the behavior representation unit and the interest level of the observer controlled by the interest level information estimated by the interest level estimation unit in association with each other. The control information includes the motion of the observed object estimated by the motion estimation means, the interest level of the observer estimated by the interest level estimation means, and the information stored in the storage means. The control information may be obtained.

  In this way, the control information for controlling the behavior expression unit, the second detection result detected by the second motion detection means, the interest level of the observer controlled by the interest level information estimated by the interest level estimation means , Etc., the past data can be used, and the system can be customized according to the individuality of the observer.

  The behavior expression processing device according to the present invention provides control information for controlling an action expression unit that performs an operation corresponding to the operation of the object to be observed with respect to the observer, and a first operation for detecting the operation of the object to be observed. Control information generating means for obtaining from the first detection result obtained by the detection means and the second detection result obtained by the second movement detection means for detecting the movement of the observer. It is characterized by.

  The behavior expression processing device may obtain the control information by changing the original control information obtained from the first detection result based on the second detection result.

  In addition, the behavior expression processing device observes the behavior of the behavior expression unit based on the second detection result and the motion estimation unit that estimates the motion of the observed object based on the first detection result. Interest level estimation means for estimating the interest level of the observer, and the motion of the observed object estimated by the motion estimation means and the interest level of the observer estimated by the interest level estimation means Control information may be obtained.

  Further, the behavior expression processing device includes an interest level estimation unit that estimates an interest level of the observer who observes the operation of the behavior expression unit based on the second detection result, and the first detection result includes The control information is obtained from the motion of the observed object estimated by the motion estimating means for estimating the motion of the observed object based on the interest level of the observer estimated by the interest level estimating means. May be.

  In addition, the behavior expression processing device includes a reception unit that receives the first detection result transmitted by the transmission unit that transmits the first detection result, and the first detection received by the reception unit. The control information may be obtained from the result and the second detection result obtained by the second motion detection means.

  Further, the behavior expression processing device obtains the control information by adding a change based on the second detection result to the original control information obtained from the first detection result received by the receiving unit. Also good.

  In addition, the behavior expression processing device includes a reception unit that receives the first detection result transmitted by the transmission unit that transmits the first detection result, and the motion estimation unit is received by the reception unit. The motion of the object to be observed may be estimated based on the first detection result.

  In addition, the behavior expression processing device includes a receiving unit that receives information on the motion of the observed object transmitted by the transmitting unit that transmits information on the motion of the observed object estimated by the motion estimating unit. The control information may be obtained from information related to the movement of the observed object received by the receiving unit and the interest level of the observer estimated by the interest level estimation unit.

  In addition, the behavior expression processing device has storage means for storing the control information for controlling the behavior expression unit and the second detection result related to the observer detected by the second motion detection means in association with each other. The original control information may be obtained from the first detection result and the information stored in the storage means.

  In addition, the behavior expression processing device stores the control information for controlling the behavior expression unit and the interest level of the observer controlled by the interest level information estimated by the interest level estimation unit in association with each other. And the control information is obtained from the motion of the observed object estimated by the motion estimation means, the interest level of the observer estimated by the interest level estimation means, and information stored in the storage means. It may be what you want.

  The behavior expression unit of the present invention is obtained by the first detection result obtained by the first motion detection means for detecting the motion of the object to be observed and the second motion detection means for detecting the motion of the observer. According to the second detection result and the control of the control information obtained from the second detection result, an operation corresponding to the operation of the object to be observed is performed on the observer.

  Further, the behavior expression unit may more preferably include the second motion detection means for detecting the motion of the observer.

  By providing the action expression unit with the second motion detection means, it is possible to perform more detailed motion detection of the observer.

  According to the present invention, it is possible to provide an action expression system and an action expression processing device that reduce “habituation” and “satisfaction” with respect to monotonous work on the watching side.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of an action expression system 1 according to the present embodiment. According to this system, for example, it is possible to watch an elderly person who lives away.

  Referring to the figure, 100 is an action expression processing device (hereinafter referred to as a server) installed on the observer (hereinafter referred to as a close relative) side, and 200 performs an operation corresponding to the operation of the object to be observed on the close relative. Therefore, an action expression unit (hereinafter referred to as a robot) installed on the close relative side, 300 is a telecommunication line (hereinafter referred to as the Internet), and 400 is a subunit installed on the observed object (hereinafter referred to as an elderly person) side.

  In the sub-unit 400, 401 is a motion detection sensor for motion detection of an elderly person, which is an example of the first motion detection means in the present invention. It is installed in household appliances. Reference numeral 402 denotes a motion estimation unit (hereinafter referred to as a behavior estimation unit) that estimates the behavior of an elderly person based on the detection result of the motion detection sensor 401. Reference numeral 403 denotes transmission means (hereinafter referred to as a transmission unit) that transmits the estimation result of the behavior estimation unit 402 to the reception unit 101 described later via the Internet 300.

  In the robot 200, reference numeral 201 denotes a robot-side mounted sensor for detecting the operation of a close relative installed in the robot, which is an example of the second operation detecting means in the present invention.

  In the server 100, reference numeral 101 denotes a receiving unit (hereinafter referred to as a receiving unit) that receives data transmitted from the transmitting unit 403 via the Internet 300. 102 is an action expression determination unit that determines action expression data to be used in the next stage according to the estimation result of the elderly person's action received by the receiving unit 101 and the estimation result of the degree of interest of the close relative in the interest degree estimation unit 106 described later. Reference numeral 103 denotes an action expression controller that issues control information (hereinafter referred to as an action expression command signal) for causing the robot 200 to perform an action corresponding to the action of the elderly person specifically according to the action expression data. The behavior expression determination unit 102 and the behavior expression controller 103 are examples of control information generation means in the present invention.

  Reference numeral 104 denotes a communication unit that exchanges the behavior expression command signal and the detection result of the robot-side mounted sensor 201 with the robot 200, and reference numeral 105 denotes another example of the second motion detection means in the present invention for detecting the motion of a close relative. It is a motion detection sensor and is installed on the environmental side of the close relative, for example, the wall or ceiling of the close relative's house, home appliances and the like.

  106 is an interest level estimation unit that estimates the degree of interest of the close relative from the motion of the close relative based on the detection result of the motion detection sensor 105 and / or the robot-side sensor 201, and 107 is the behavior expression An estimation result storage unit (hereinafter, interest level DB [database]) that stores estimation results of the determination unit 102 and the interest level estimation unit 106. It is also possible for the behavioral expression determination unit 102 to determine the behavioral expression data in consideration of the data of the interest level DB 107.

  The motion detection sensor 105, the robot-side mounted sensor 201, and the motion detection sensor 401 include, for example, an image sensor, a sound sensor, a pressure sensor, an infrared (pyroelectric) sensor, a distance sensor, an acceleration sensor, and the like. Further, the motion detection sensor 105 and the motion detection sensor 401 may be, for example, a sensor mounted on a switch or door of a household electrical appliance, or a wearable sensor mounted on a close relative or an elderly person (not shown). It may be.

  FIG. 2 is a schematic external view for explaining an example of the robot 200 described above. The robot 200 shown in the figure is a robot capable of outputting voices by LED displays (eyes, cheeks, etc.) of the face portion and the speaker of the mouth portion, in addition to the operating portion having the total number of axes of 6, but is not limited thereto. Further, a humanoid robot that can express more detailed behavior and emotion by increasing the number of axes may be used. Further, when the object to be observed is a dog or the like, it may be a dog type robot, or may be a virtual flower or the like that can perform behavioral expression in accordance with the motion of an elderly person other than the robot.

  With the configuration of FIG. 1, in the behavior expression system 1 according to the present embodiment, the motion detection sensor 105 detects the motion of the close relative who saw the behavior expression performed by the robot 200, and the close relative's robot is based on the result. The interest level with respect to 200 is estimated by the interest level estimation unit 106, and the estimation result is fed back to the behavior expression determination unit 102. In other words, the degree of interest of the close relative to the behavioral expression of the robot 200 that is the output of the server 100 is fed back to the server 100 and reflected in the operation of the robot 200. As a result, the robot 200 can attract the attention of close relatives, and can reduce the “familiarity” and “satisfaction” of close relatives.

  Next, an operation flow in the series of behavior expression system 1 for watching the elderly will be described with reference to the flowchart of FIG.

  Referring to the figure, the right side of the figure is an operation flow on the subunit 400 side of an elderly person, and the left side is an operation flow on the server 100 side of a close relative. Both flows operate independently of each other.

  First, an operation flow on the right elderly home subunit 400 side will be described.

  When the power of the subunit 400 in the elderly home is turned on, the operation transitions from the start to step S101.

  In step S101, elderly person motion detection (hereinafter referred to as sensing operation) using the motion detection sensor 401 is started. Sensing data detected by each sensor is stored in a memory (not shown) in the subunit 400 in time series.

  In step S102, sensing data for an arbitrary time is acquired from the memory. That is, since the behavior is not an instantaneous character but rather has temporal continuity, a useful amount of sensing data is read when performing behavior estimation in the next step. Note that this time may depend on the time period for transmitting the estimation result from the transmission unit 403 to the reception buffer in step S104.

  In step S <b> 103, the behavior estimation unit 402 estimates the specific behavior of the elderly within the time using the sensing data. As a result, for example, the behavior of elderly people such as sleeping, eating, cleaning, watching TV, and walking is estimated from the sensing result.

  In step S <b> 104, the estimation result is output from the transmission unit 403. The output estimation result is output to the reception buffer of the server 100 via the reception unit 101. This transmission operation is performed at intervals of 1 minute, for example.

  In step S105, if there is an operation end instruction or the like to the subunit 400, an end process is performed, and a series of flows is ended. Otherwise, the process moves to step S102, and a series of flows is repeated.

  Next, the operation flow on the left side relatives' home server 100 side will be described. The description will be made on the assumption that the power of the server 100 is turned on after a lapse of a certain time since the power of the subunit 400 is turned on and data is accumulated in the reception buffer in the server 100 to some extent.

  When the power of the server 100 in the close relative's house is turned on, the operation transitions from start to step S201 in the flow in the left diagram.

  In step S201, a close relative sensing operation using the motion detection sensor 105 is started. Sensing data detected by each sensor is stored in a memory or the like in the server 100 in time series. Note that the robot-side sensor 201 may be used for the sensing operation. In the present embodiment, the case where the robot-side sensor 201 is also used will be described as an example.

  In step S202, sensing data for an arbitrary time is acquired from the memory. That is, a useful amount of sensing data is read when performing behavior estimation in the next step.

  In step S <b> 203, the degree of interest (degree of interest) of the close relatives robot 200 (including the behavioral expression of the robot 200) is estimated by the interest level estimation unit 106 using the sensing data. In this estimation, when only the motion detection sensor 105 is used, the degree of interest of the close relative to the robot 200 is estimated from only the sensing result of the sensor 105. When the robot-side sensor 201 is used as in the present embodiment, for example, the motion detection sensor 105 is used to specify the location of the close relative and move the robot 200 toward the close relative. It is also possible to properly use the sensing results, such as estimating whether the person is interested in the robot 200 using the sensing results of the robot-side sensor 201.

  In step S <b> 204, the degree of interest of the above-mentioned relatives in the robot 200 is stored in the degree-of-interest DB 107. Note that if the data read from the interest level DB 107 in step S207 described later is not referred to, step S204 may not be executed.

  In step S205, the behavior expression determination unit 102 reads the estimation result of the elderly's behavior from the reception buffer. The amount of data at the time of reading and the reading cycle may be arbitrarily set, but the amount of data and the cycle should not be such that the buffer overflows.

  In step S206, the degree of interest of the close relative stored in the past from the interest degree DB 107 is read out. For example, the degree-of-interest DB 107 stores the degree of interest, profile, preference, pre-existing condition, etc. of a close relative. Note that if the data read from the interest level DB 107 in the next step S207 is not referred to, the step S206 need not be executed.

  In step S207, action expression data as shown in FIG. 4 is determined. The behavior expression data may be, for example, a gesture for an elderly person. The robot action output is uniquely determined according to the elderly person action estimation result read in step S205. Further, the motion level in the robot action output is determined according to the degree of interest of the close relative evaluated in step S203. The series of contents will be described later with reference to FIG.

  In step S208, an action expression command signal determined using the matrix in FIG. 4 according to the robot action output and action level determined by the action expression determining unit 102 is transmitted from the action expression controller 103 to the robot 200. Is done. In accordance with this command, the robot 200 performs behavioral expression.

  In step S209, if there is an operation end instruction or the like to the server 100, end processing is performed, and a series of flows is ended. Otherwise, the process moves to step S202, and a series of flows is repeated.

  Next, a method for determining the robot action output in step S207 will be described.

  Referring to the table of FIG. 4, the behavioral expression of the robot 200 is determined from the table as shown in the figure in accordance with the state level (described later) of the elderly and the degree of interest of the close relative in the robot 200. A plurality of such tables may exist as shown in the figure, and the tables may be determined depending on the contents of the results of the elderly person's behavior estimation. The figure shows an example of selecting a table when it is estimated from the result of behavior estimation that an elderly person enjoys watching a baseball game on television. Although the figure is described as a 5 × 5 matrix for the sake of explanation, it is not limited to this and can be set freely.

  Now, in the table of the figure, the vertical axis direction specifically represents the robot action output. The robot action output is selected depending on the state level of the elderly person described later obtained as a result of the elderly person's action estimation. On the other hand, the horizontal axis direction specifically represents the motion level of the robot operation corresponding to the degree of interest of the close relative in the robot 200. The larger the number, the higher the degree of interest in the robot, and the movement level of the robot decreases accordingly. The operation level of the robot is the degree of exaggeration of the robot operation. For example, as seen in the table, the operation level of the robot represents the LED light emission period, the sound volume, and the degree of limb swing. This is determined depending on the degree of interest estimated in step S203.

  As described above, the robot action output is selected depending on the result of the elderly person's action estimation, the action level of the robot is changed based on the degree of interest of the close relative, and the action expression in the matrix is determined. That is, the content of action output of the robot 200 and its operation level are determined. This is output from the action expression determination unit 102 as action expression data. Note that the content in the matrix is an example for explanation, and is set as appropriate according to the expression capability of the robot.

  In addition, in the situation where you enjoy watching TV in the living room while watching the baseball game on the elderly, the living room's infrared sensor etc. or a sensor attached to the TV power supply, This can be achieved by using a sensation sensor for detecting a laughing voice of the subject and referring to an electronic program guide (EPG). As a result of such behavior estimation, a general “behavior” of “watching a baseball game on TV” and a “state level” of the elderly are transmitted from the home of the elderly. The state level of an elderly person is, for example, “the state of watching quietly”, “it is frustrating (in the losing game of a niece ’s team)” or “(with the audience on the television). The actions are further classified in the “behavior” such as “supporting together”. The table is subdivided into five stages as described above from the viewpoint of the exaggerated behavior and aggressiveness of the elderly, and is shown as examples of five different expressions. Note that the correspondence between the “behavior” and “state level” described above and the action output content of the robot is determined in advance and stored in the server 100. Such a behavioral expression of a robot may be an output of a gesture imitating the movement of an elderly person in the case of a humanoid robot or the like. In the present embodiment, the latter case is described.

  In the figure, for example, when the robot action output is determined to be “2” and the degree of interest in the robot is evaluated as “2”, the robot 200 moves both hands, and the range of the swing is a swing. 50% of the entire movable range. Also, if the robot action output is determined to be “2” and the degree of interest in the robot is evaluated as “1”, the degree of interest is lower than when the degree of interest is “2”, so that the close relatives are interested. In addition, the movement of shaking both hands is the same, but the range of the swing is 75% of the whole. If the degree of interest in the robot is evaluated as “0”, the robot 200 moves both hands to attract more interest. The operation which shakes frequently is performed.

  On the other hand, when the degree of interest in the robot is evaluated as “3”, it can be said that the degree of interest in the close relative of the robot 200 is higher than in the case where the degree of interest in the robot 200 is evaluated as “2”. However, the swing is 25% of the whole, and when the degree of interest in the robot is evaluated as “4”, it can be determined that it is not necessary to attract the close relative's interest, so the robot 200 shakes one hand. It becomes operation degree.

  As described above, the behavior of the elderly person at the remote location estimated by the robot 200 is expressed by the motion, the motion of the close relative who sees it is detected by the motion detection sensor 105, and the relative detection result of the relative 200 with respect to the robot 200 of the relative The interest level is estimated by the interest level estimation unit 106, and the estimation result is fed back to the behavior expression determination unit 102. In other words, the degree of interest of the close relative to the behavioral expression of the robot 200 that is the output of the server 100 is fed back to the server 100 and reflected in the operation of the robot 200. As a result, the robot 200 can attract the close relative's attention, and has the effect of being able to reduce the “familiarity” and “satisfaction” of the close relative.

  Next, a method for determining the robot action output and the action level of the robot action output with reference to past data of the interest degree of the close relatives in the robot 200 in the interest degree DB 107 read in step S206 will be described. For example, when the average value of the degree of interest of the close relative is low, the degree of interest is changed to a value obtained by subtracting, for example, 1 from the degree of interest evaluated in step S203. It can be set so that the robot movement is one rank higher than the case. Instead of changing the degree of interest evaluated in step S203, the operation level in the matrix of FIG. 4 may be changed to a setting that is significant for one rank. Also, by preparing a table that can handle interest levels less than 0 and shifting the matrix data to the right by one, the robot's behavior is set to be one rank higher with the interest level unchanged. You can also.

  As a result, it is possible to cope with an observer who tends to get bored.

  Further, when a history of high blood pressure or the like is registered in the interest level DB 107, for example, the interest level is added to a value obtained by adding, for example, 1 to the interest level evaluated in step S203 in order to avoid a long excitement state. Or the action output of the robot 200 in the matrix of FIG. 4 is changed by one rank. In addition, when the interest level DB 107 is registered in the profile as to which team the fan is, it is also possible to determine the robot action output and the action level of the robot action output in consideration thereof.

  As described above, it is possible to operate the system in a state closer to the personality and preferences of the close relatives by referring to the past data and the past history of the interest level of the close relatives using the interest level DB 107. There is an effect that can be done.

  Although FIG. 4 shows an example of “enjoying watching a baseball game”, a plurality of similar tables may be prepared, and they correspond to each test subject's action. For example, “enjoy watching boxing” or “enjoy TV drama” may be used. These table data are registered in the memory in the server 100, the interest level DB 107, and the like, and are sequentially referred to as necessary. In the present embodiment, the example is described in which the action expression of the robot is determined using the table. However, the present invention is not limited to the table, and a calculation formula or a function for determining the action expression may be used.

  In this embodiment, when the robot 200 is equipped with a self-propelled function as in this embodiment in step S203, the feature is utilized and, for example, the face is moved to the front of the close relative to express the facial expression. In addition to acquiring images as an image, it is possible to perform a more detailed detection by drawing a reaction to a close relative's robot 200 by talking with a voice or contacting a close relative with a limb or the like. Can be measured more accurately.

  Further, as shown in FIG. 1, the above embodiment is a case where the elderly person's behavior estimation unit 402 is present in the subunit 400 of the elderly person's house, but the server 100 as shown in FIG. It may exist inside (behavior estimating part 108). In this case, the following changes are made to the above embodiment.

  That is, in the subunit 400, the motion detection sensor 401 and the transmission unit 403 are the same as those in FIG. 1, and the behavior estimation unit 402 is not mounted. In this case, the subunit 400 in FIG. 5 has only a function of transmitting the detection result of the motion detection sensor 401.

  On the other hand, in the server 100 of FIG. 5, a behavior estimation unit 108 that performs the same processing as the behavior estimation unit 402 in FIG. 1 is inserted between the reception unit 101 and the behavior representation determination unit 102. That is, the reception unit 101 receives data from the transmission unit 403 via the Internet 300 and transmits the data to the behavior estimation unit 108, and the behavior estimation unit 108 transmits the estimation result to the behavior expression determination unit 102. In this case, the server 100 performs the behavior estimation of the elderly and the degree of interest of the close relative.

  As described above, according to the present invention, it is possible to indirectly monitor an elderly person by detecting the action of the elderly person who is far away by the motion detection sensor 401 and expressing the action by the robot 200 in the close relative's house. In addition, the action detection sensor 105 and the robot-side sensor 201 provide feedback to the behavioral expression of the robot 200 by feeding back the interest level of the close relative to the robot 200, thereby changing the behavioral expression. It is possible to reduce the “familiarity” and “satisfaction”, and it is possible to achieve continuity of system operation.

  Further, by using the degree of interest DB 107, the behavior expression system and the behavior expression processing device according to the present invention can be made to correspond to the personality and preference of a close relative.

  It should be noted that the behavior expression system in the present embodiment can be realized in hardware by a CPU, memory, or other LSI of an arbitrary computer. In terms of software, it is realized by a program having an action expression function loaded in a memory. FIG. 1 shows functional blocks of an action expression system realized by hardware and software. However, it goes without saying that these functional blocks can be realized in various forms such as hardware only, software only, or a combination thereof.

  The embodiments of the present invention can be appropriately modified in various ways within the scope of the technical idea shown in the claims.

It is a block diagram of the action expression system which concerns on embodiment. It is a figure which shows the robot which concerns on embodiment. It is an operation | movement flowchart concerning the action expression system which concerns on embodiment. It is a table figure which determines the action output and action level of the robot which concern on embodiment. It is a block diagram of the action expression system which concerns on another embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Action expression system 100 Server 101 Reception part 102 Action expression determination part 103 Action expression controller 105 Motion detection sensor 106 Interest level estimation part 107 Interest level DB
DESCRIPTION OF SYMBOLS 108 Action estimation part 200 Robot 201 Robot side mounting sensor 400 Sub unit 401 Motion detection sensor 402 Action estimation part 403 Transmission part

Claims (9)

First motion detection means for detecting the motion of the object to be observed;
An action expression unit that performs an operation corresponding to the operation of the object to be observed on an observer;
Second motion detection means for detecting the motion of the observer;
With
Control information for controlling the behavior expression unit is obtained from the first detection result obtained by the first motion detection means and the second detection result obtained by the second motion detection means. Having control information generating means for obtaining,
An action expression system characterized by that. The behavior expression system according to claim 1,
The control information generating means
Obtaining the control information by adding a change based on the second detection result to the original control information obtained from the first detection result;
An action expression system characterized by that. The behavior expression system according to claim 1,
Motion estimation means for estimating the motion of the object to be observed based on the first detection result;
Interest level estimation means for estimating the level of interest of the observer who has observed the action of the behavior expression unit based on the second detection result;
With
The control information generating means
Obtaining the control information from the motion of the observed object estimated by the motion estimation means and the interest level of the observer estimated by the interest level estimation means;
An action expression system characterized by that. The behavior expression system according to claim 3,
Transmitting means for transmitting information on the motion of the observed object estimated by the motion estimating means;
Receiving means for receiving information relating to the operation of the observed object transmitted by the transmitting means;
Have
The control information generating means
Obtaining the control information from the information on the motion of the observed object received by the receiving means and the interest level of the observer estimated by the interest level estimation means;
An action expression system characterized by that. The behavior expression system according to claim 2,
Storage means for storing the control information for controlling the behavior expression unit and the second detection result relating to the observer detected by the second motion detection means in association with each other;
The control information generating means
Obtaining the original control information from the first detection result and the information stored in the storage means;
An action expression system characterized by that. The behavior expression system according to claim 3 or 4,
Storage means for storing the control information for controlling the action expression unit and the interest level of the observer controlled by the interest level information estimated by the interest level estimation unit in association with each other;
The control information is
Obtaining the control information from the motion of the observed object estimated by the motion estimation means, the interest level of the observer estimated by the interest level estimation means, and information stored in the storage means;
An action expression system characterized by that. The behavior expression system according to any one of claims 1 to 6,
The behavior expression unit includes the second motion detection unit.
An action expression system characterized by that. Control information for controlling an action expression unit that performs an action corresponding to the action of the object to be observed on the observer,
A first detection result obtained by a first motion detection means for detecting the motion of the object to be observed;
A second detection result obtained by a second motion detection means for detecting the motion of the observer;
Control information generating means for obtaining from
An action expression processing device characterized by that. The behavior expression processing device according to claim 8,
Obtaining the control information by adding a change based on the second detection result to the original control information obtained from the first detection result;
An action expression processing device characterized by that.

Images