JP2005327156A - Relationship detection system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a relationship detection system capable of detecting personal relationships in a short period of time by detecting the movement loci of persons. <P>SOLUTION: The relationship detection system includes a plurality of readers 20 for detecting radio tags 18 attached to persons. The reader 20 detects the distance to or the direction of each radio tag 18. Based on this distance or direction information, the positions of the radio tags 18, i.e., the persons 14, are detected and the movement loci of the persons 14 are recorded. Based on the movement loci, a determination is made as to, e.g., whether or not one of the persons 14 and another of the persons 14 have moved along a similar locus at about the same time. This allows determining how friendly the persons 14 are to each other. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a relationship detection system, and more particularly to, for example, a relationship detection system that detects the movement of a person in a predetermined area and detects the relationship between a person and another person.

  Conventionally, for example, sociometry devised by Moreno (J.L. Moreno) is well known as a method for grasping human relationships. Sociometry measures the position and relationship of individuals within a group by investigating likes (acceptance) and dislikes (rejection) within the group. For example, it is often applied to grasp the relationships between students in a class in school education. In a so-called sociometric test, questions such as “Who do you like in class? Who do you want to play with?” Are written, and the names of those who like and dislike are written to each person. It is Based on the results of this test, a so-called sociogram or a so-called socio-matrix table is created as necessary, for example, the position within an individual group, such as a popular person, an isolated child, or a rejected child within a group. Friends relations become clear. For example, Non-Patent Document 1 and Non-Patent Document 2 introduce software that outputs a sociogram or the like using information obtained by a sociometric test as input data in a PC.

  However, the prior art using this sociometric test has the following problems. In other words, since each member directly answered the likes and dislikes, there was a risk that the answers included falsehood. In addition, depending on the method of surveying the content and timing of the questions, there was a strong risk of becoming a human rights issue or adversely affecting human relationships thereafter. Therefore, accurate information could not be obtained, and the relationship among members could not be accurately grasped.

Therefore, the present applicant has proposed a relationship detection system capable of accurately grasping the relationship between members in Japanese Patent Application No. 2003-371474 filed on October 31, 2003. In short, in this system, for example, a communication robot is introduced into an organization, and human relationships in the organization are detected based on the time when members stay together in front of the communication robot.
“Sociometrics Frenzu”, [online], Suzuki Education Software Co., Ltd., [searched September 17, 2003], Internet <URL: http://www.suzukisoft.co.jp/products/frends/index. htm> “Sociometric Test for Windows”, [online], CS Development Office, [searched on September 17, 2003], Internet <URL: http://plaza5.mbn.or.jp/~cesdev/socio.html >

  In this prior art, it is possible to accurately grasp the relationship between members in an organization, but it is necessary to collect data for a relatively long period of time for the detection.

  Therefore, a main object of the present invention is to provide a relationship detection system capable of detecting a relationship between one person and another person in a short period of time.

  The invention according to claim 1 is a relationship detection system that detects the movement of a person in a predetermined area and detects the relationship between a person and another person, and includes a plurality of wireless tags attached to each person, A plurality of readers that are arranged on the wireless tag to detect the wireless tag, a position detection unit that detects a position of the wireless tag detected by the plurality of readers, and a movement locus of the wireless tag based on the position detected by the position detection unit Is a relationship detection system comprising a movement locus recording means for recording the relationship and a grasping means for grasping the relationship based on the movement locus recorded by the movement locus recording means.

  According to the first aspect of the present invention, a wireless tag is attached to each person in order to detect the relationship between one person and another person. In the embodiment, an RFID tag or an infrared tag is applied as the wireless tag. The predetermined area in which each person with the wireless tag is present is assumed to be a human movement, for example, and in the embodiment, an exhibition environment such as a museum or a science museum is applied. . In this area, a plurality of readers for detecting the wireless tag are arranged. For example, when the wireless tag is an RFID tag, the reader detects the distance from the RFID tag, and when the wireless tag is an infrared tag, the reader detects the direction of the infrared tag. The position detecting means detects the position of the wireless tag, that is, the human being detected by the plurality of readers. For example, the position of the wireless tag is specified by distance information or direction information of the wireless tag detected by three readers. Based on the detected position, the movement trajectory recording means records a wireless tag, that is, a human movement trajectory. The grasping means grasps a relationship between a certain person and another person based on the movement trajectory. Therefore, according to the first aspect of the present invention, the human relationship can be detected in a short period of time by detecting the movement of the human.

  The invention of claim 2 is dependent on the invention of claim 1, and the grasping means determines that a person and another person are present within a certain distance during a certain past time based on the movement trajectory. And a relationship determination means for determining that the humans are in a friendly relationship.

  According to the second aspect of the present invention, since it is considered that there is a friendly relationship between humans who have moved along the same trajectory almost at the same time, the relationship determining means determines the human friendly relationship based on the moving trajectory. Therefore, according to the invention of claim 2, by detecting the movement, it is possible to detect human friendship in a short period of time.

  The invention of claim 3 is dependent on the invention of claim 1 or 2, and the grasping means determines that a person based on the movement locus does not exist within a certain distance from the average movement locus during the past certain time. When it is determined, it includes behavior determining means for determining that the human behavior is unique.

  According to the invention of claim 3, the action determination means determines whether or not a certain human movement is peculiar based on a certain human movement trajectory and an average movement trajectory. Therefore, according to the invention of claim 3, it is possible to detect a person who takes a peculiar action.

  According to the present invention, the position of the wireless tag can be detected by a plurality of readers, the movement trajectory of a person can be recorded, and the relationship between a person and another person can be grasped based on the movement trajectory. Therefore, by detecting the movement of a human, it is possible to detect a human relationship in a short period of time.

  The above object, other objects, features and advantages of the present invention will become more apparent from the following detailed description of embodiments with reference to the drawings.

  With reference to FIG. 1, the relationship detection system 10 of this embodiment is for detecting the movement of a person 14 in a predetermined area 12 and detecting the relationship between a person 14 and another person 14. . In this embodiment, as an example of the predetermined area 12, a case will be described in which an exhibition environment such as a museum or a science museum is premised on the movement of a person. That is, since a plurality of exhibits 16 are arranged at predetermined positions in the exhibition room 12, the human 14 needs to move in the exhibition room 12 in order to view various exhibits 16. This relationship detection system 10 is intended to grasp a human relationship by detecting a human movement in a predetermined area where a human movement is required or assumed.

  In order to detect movement, a wireless tag 18 is attached to each person 14, and a plurality of readers 20 for detecting the wireless tag 18 are provided in the exhibition room 12. In this embodiment, an RFID tag is used as the wireless tag 18. RFID (Radio Frequency Identification) is an automatic recognition technology using a non-contact IC tag using electromagnetic waves. The RFID tag includes an IC chip, an antenna, and the like including a memory for identification information, a control circuit for communication, and the like. Information that can identify the human 14 is stored in advance in the memory, and the identification information is output from the antenna by electromagnetic waves, radio waves, or the like having a predetermined frequency.

  In this embodiment, since the position of the person 14 in the exhibition room 12 is detected as the transmission method, for example, the microwave method (up to about 5 m) or the electromagnetic induction method (up to about 1 m) having a relatively long communication distance. ) Is preferable. In these cases, when the human 14 enters the detectable area of the reader 20, the person 14 can be easily recognized just by wearing the wireless tag 18. In addition, it is desirable to use an active type with a built-in battery in order to secure an appropriate communication distance region. The form or shape of the tag is arbitrary, and may be, for example, a card shape, a label shape, a coin shape, or a stick shape. The wireless tag 18 is preferably attached to a name tag for each user or the like in order to prevent the human (user) 14 from being nervous or causing unnatural behavior.

  In this embodiment, the reader 20 detects output information from the RFID tag. Specifically, the reader 20 includes an antenna, receives a radio wave superimposed with identification information transmitted from the RFID tag via the antenna, amplifies the radio signal, and separates the identification information from the radio signal. The information is demodulated (decoded). In this embodiment, the reader 20 measures the distance between the RFID tag and the reader 20 based on the measured radio wave intensity.

  Since the reader 20 measures the distance from the wireless tag 18, the wireless tag 18 needs to be detected by at least three readers 20 in order to specify the position of the wireless tag 18 (human 14). Accordingly, a plurality of readers are arranged in such an arrangement that the human 14 wearing the wireless tag 18 can be detected by at least three readers 20 at any position in the exhibition room 12 from the entrance to the exit. 20 is installed.

  FIG. 2 shows an electrical configuration of the relationship detection system 10. Referring to FIG. 2, the relationship detection system 10 includes a CPU 22 in charge of overall control. A memory 24 is connected to the CPU 22 via a bus, and the memory 24 includes RAM, ROM, HDD, and the like. A program for controlling the operation of this system is stored in the ROM or HDD, and the CPU 22 executes processing according to this program. The RAM is used as a work area or buffer area for the CPU 22. A communication device 26 is connected to the CPU 22. The communication device 26 is for communicating with an external computer in a wireless or wired manner via a network such as a wireless LAN or the Internet. Furthermore, a plurality of readers 20 are connected to the CPU 22 via a general-purpose interface such as RS-232C. Data such as identification information and distance of the wireless tag 18 detected by each reader 20 is given to the CPU 22 and stored in a predetermined area of the memory 24.

  2 shows a case where all of the plurality of readers 20 are connected to one computer (CPU 22). However, the relationship detection system 10 includes a plurality of computers in charge of one or more readers 20. It may be configured. In this case, each computer transmits data including information detected by the reader 20 to one main computer via the communication device.

  In the relationship detection system 10, the information of the wireless tag 18 detected by the reader 20 is acquired by the CPU 22 at regular time intervals (for example, 1 second). That is, the CPU 22 detects the wireless tag 18 at intervals of 1 second, for example. In the reader 20, the identification information of the wireless tag 18 is detected, and the distance from the wireless tag 18 is detected based on the radio wave intensity. The reader 20 outputs the detected time information to the CPU 22 together with the detected identification information and distance information of the wireless tag 18. Each reader 20 is synchronized by the CPU 22.

  The CPU 22 stores the acquired information in the memory 24. In the memory 24, user information such as an ID number and a name of the person 14 is registered in advance in association with the identification information of the wireless tag 18. Therefore, the person 14 detected by the reader 20 can be specified based on the acquired identification information.

  In the memory 24, information related to the position coordinates of each reader 20 in the exhibition room 12 is registered in advance. Note that the position data in the exhibition room 12 is represented by an XY two-dimensional plane coordinate system as shown in FIG. 1, and in this embodiment, the vertical direction (Z axis) is not set. The CPU 22 determines the wireless tag 18 (human 14) based on the distance information detected between the three readers 20 that have detected the identification information of the same wireless tag 18 and the position data of each reader 20. The position coordinates in the exhibition room 12 can be specified.

  Further, the CPU 22 records the movement trajectory of the human 14, that is, stores the calculated position coordinate of the human 14 in the memory 24 in association with the detection time tn. Specifically, as shown in FIG. 3, the movement trajectory is stored for each person 14 (identification information or ID number). The movement trajectory data is time-series data. When the position at time tn is Ptn (Xn, Yn), the trajectory from time t0 to tn is represented as a point sequence Pt0, Pt1,..., Ptn. FIG. 3 shows the movement trajectory data (Pat0,..., Patn,...) Of the user whose ID is Ia and the movement trajectory data (Pbt0,..., Pbtn,...) Of the user whose ID is Ib. . In this embodiment, the information on the detection time tn is output from each reader 20, but in other embodiments, the CPU 22 may count the detection time tn.

  Based on this movement trajectory, human relationships (for example, interpersonal relationships, social relationships) are grasped. That is, in this embodiment, after calculating the movement trajectory, the human relationship is determined for all combinations (A, B) of the users. Since it is considered that there is a friendly human relationship between users who have moved through similar trajectories almost simultaneously, the human relationship is determined based on this idea. That is, it is determined whether or not the user A and the user B exist within a certain distance D during the past certain time T. Note that predetermined values are preset for the values of T and D, respectively. Specifically, the average value of the distance between A and B is calculated for each detection time during the past fixed time T, and it is determined whether or not the average value of the distance is within the distance threshold D. When the user A and the user B exist within a certain distance D during the past certain time T, it is determined that there is a friendly relationship between A and B. It is determined that no friendly relationship exists.

  In this embodiment, the friendship between users can be grasped by detecting the position and movement locus of the found wireless tag 18 and comparing and determining the movement locus of the user. Therefore, by detecting the movement trajectory of the user, it is possible to detect the human relationship almost in real time, that is, in a short period. Of course, it is also possible to detect human relationships after detecting the movement behavior of all users.

  In addition, the relationship between each user's moving behavior and the standard moving behavior is grasped. That is, it is determined whether or not a certain user A exists within a certain distance D from the average trajectory α during the past certain time T. The standard movement trajectory α is obtained by collecting the movement trajectories of a large number of users. In other words, the standard trajectory α is calculated from, for example, the average of movement trajectories regarding all persons who have used the relationship detection system 10 in the past, and is stored in advance in a predetermined area of the memory 24. Alternatively, in another embodiment, a plurality of average movement trajectories may be provided and stored as a standard trajectory set α = {α0, α1,..., Αn}. α is compared with the movement trajectory of the user.

  In the comparison determination of the movement trajectory, specifically, the distance between the user A and the standard trajectory is calculated for each detection time during the predetermined time T in the same manner as the determination of the friendly relationship described above, and the average value of the distances is calculated. Is within the distance threshold D. Note that the value of the fixed time T and the value of the fixed distance D do not have to be set to be the same as the values of T and D in the above-described friendly relationship determination. Then, when the user A exists within a certain distance D from the average trajectory α during the past certain time T, it is determined that A's moving behavior is average. Is determined to be unique.

  In this way, it is possible to detect the relationship between each user's movement behavior and standard behavior, and to grasp whether each user is a standard behavior or a specific behavior. . In addition, as a result of the action determination, it is possible to find a user who is particularly interested or not interested in each exhibit 16. Therefore, it is possible to use the detected relationship for display guidance, and it is also possible to detect abnormalities in the display environment.

  FIG. 4 and FIG. 5 show an example of the operation of this relationship detection system 10. In the first step S1 of FIG. 4, the CPU 22 of the relationship detection system 10 acquires data including information on IDs (identification information), distances, and detection times of all the wireless tags 18 found from each reader 20. And stored in the memory 24. Next, in step S3, a current position and a movement locus are calculated for each acquired ID. The current position of the ID (wireless tag 18 or human 14) is specified based on the distance from the three readers 20. Further, as shown in FIG. 3, the movement trajectory data is stored as time-series data of position coordinates at each detection time tn.

  Subsequently, in step S5, an ID combination (A, B) for determining the relationship is selected. That is, two IDs (A and B) for determining the relationship are selected from the IDs of all users. Then, in step S7, based on the movement trajectory data of A and B, it is determined whether A and B exist within a certain distance D during a past certain time T. Specifically, the average value of the distance between A and B is calculated for each detection time within the past fixed time T, and it is determined whether or not this distance average value is within the distance threshold value D.

  If “YES” in the step S7, it is determined that a friendly relationship exists between A and B in a step S9. On the other hand, if “NO” in the step S7, it is determined in a step S11 that there is no friendly relationship between A and B. As the determination result, for example, the IDs of other users who have a friendly relationship as friendly relationship data may be recorded in each user information. Or you may make it record the combination of the users who are in a friendly relationship as friendly relationship data.

  In step S13, it is determined whether or not the relationship determination has been completed for all ID combinations. If “NO”, the process returns to step S5 to repeat the process. On the other hand, if “YES” in the step S13, the process proceeds to the next step S15 in FIG.

  Although omitted in FIG. 4, the determination result may be displayed on, for example, a display (not shown) of the relationship detection system 10 or may be output by voice. In these cases, the operator of the relationship detection system 10 can know the user's human relationship in almost real time.

  Subsequently, in step S15 of FIG. 5, selection of ID (A) for performing behavior determination is performed. That is, one ID (A) is selected from the IDs of all users. Subsequently, in step S17, based on the movement trajectory data of A and the standard movement trajectory data α, whether or not A exists within a certain distance D from the average trajectory α during the past certain time T. Is judged. Specifically, the average value of the distance between A and α is calculated for each detection time within the past fixed time T, and it is determined whether or not this distance average value is within the distance threshold D.

  If “YES” in this step S17, it is determined that the moving action of A is average in a step S19. On the other hand, if “NO” in the step S17, it is determined that the moving action of A is unique in a step S21. As the determination result, for example, each user information may be recorded with data indicating that the person is performing a specific action, or the ID of a user who performs a specific action may be recorded as the action data. May be.

  Subsequently, in step S23, it is determined whether or not the action determination has been completed for all the IDs. If “NO”, the process returns to step S15 to repeat the process. Note that the behavior determination result may be notified to the operator by display or voice output, similar to the above-described relationship determination result.

  On the other hand, if “YES” in the step S23, it is determined whether or not the relationship detection process is ended in a succeeding step S25. If “NO” in the step S25, the process returns to the step S1 in FIG. 4 to repeat the process. In addition, the process from this step S1 is performed for every fixed time (for example, 1 second). Therefore, the detection of the wireless tag 18, the detection of the position, the detection of the movement trajectory, the grasp of the relationship, and the like are repeatedly executed. On the other hand, if “YES” in the step S25, that is, for example, when there is an operation input instructing the end from an operation unit (not shown) by the operator, or when a predetermined time has elapsed, this relationship detection is performed. The process ends.

  In the above-described embodiment, the relationship determination process (step S5 to step S13) is executed at regular intervals such as every detection time, but this relationship determination process is performed more than the detection time interval. You may make it perform every long predetermined time. The same applies to the action determination process (from step S15 to step S23).

  Further, in each of the above-described embodiments, the movement in one exhibition room 12 is detected and the relationship is grasped, but the movement over a plurality of exhibition rooms 12 may be detected.

  In each of the above-described embodiments, the case where an RFID tag is applied as the wireless tag 18 has been described. However, an infrared LED tag may be applied as the wireless tag 18. In this case, an infrared sensor such as an infrared camera is applied as the reader 20. The reader 20 detects the identification information of the infrared LED tag at a predetermined detection time tn, and detects the direction in which the infrared LED tag exists with a resolution of several degrees or less. Therefore, the CPU 22 can specify the position of the wireless tag 18, that is, the human 14 based on the direction information given from the three readers 20 as in the case of the RFID tag.

It is an illustration figure which shows the outline | summary of the relationship detection system of one Example of this invention. It is a block diagram which shows the electric constitution of a relationship detection system. It is an illustration figure which shows an example of the content of the movement locus | trajectory memorize | stored in memory. It is a flowchart which shows a part of example of operation | movement of a relationship detection system. FIG. 5 is a flowchart showing a continuation of FIG. 4.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Relationship detection system 12 ... Exhibition room 14 ... Human 18 ... Wireless tag 20 ... Reader 22 ... CPU
24 ... Memory

Claims (3)

A relationship detection system that detects the movement of a person in a predetermined area and detects the relationship between one person and another person,
Multiple wireless tags attached to each person,
A plurality of readers arranged in the region for detecting the wireless tag;
Position detecting means for detecting the position of the wireless tag detected by the plurality of readers;
A movement locus recording means for recording the movement locus of the wireless tag based on the position detected by the position detection means; and a grasping means for grasping the relationship based on the movement locus recorded by the movement locus recording means. , Relationship detection system.   When the grasping means determines that a person and another person are present within a certain distance during a certain past time based on the movement trajectory, the relationship is determined that the persons are in a friendly relationship with each other. The relationship detection system according to claim 1, further comprising a determination unit.   The grasping means determines that the human action is peculiar when it is determined that a human being does not exist within a predetermined distance from the average moving trajectory during the past fixed time based on the moving trajectory. The relationship detection system according to claim 1, comprising behavior determination means.

Images