JP5567725B2 - Group behavior estimation device - Google Patents

Description

The present invention relates to a group activity estimation equipment, in particular for example, to estimate the group present in the environment is performing what action relates to a novel group activity estimation equipment.

  2. Description of the Related Art Conventionally, a technique has been proposed for estimating what kind of behavior a person is performing based on accumulated personal position information and movement information. For example, in the technique of Patent Document 1, the position information of the target person is acquired using a GPS terminal, and “less interesting search behavior”, “hang around” based on the movement history and the time taken for movement. The behavioral state of the target person such as “searching behavior” and “searching behavior with a target of interest” is estimated.

JP 2008-152655 A [G06Q 50/00]

  However, the technique disclosed in Patent Document 1 is only used to estimate an individual's behavior and cannot estimate what kind of behavior a plurality of persons are performing as a group. Further, since it is necessary to observe the target person for several minutes or more in order to estimate the behavior of the target person, it is not possible to provide a service that quickly responds to the action of the target person.

Another object of the present invention is to provide novel, a group activity estimation equipment.

Another object of the invention can estimate the behavior of the group is to provide a group activity estimation equipment.

Still another object of the present invention can be estimated in a short time of observing behaviors of the group is to provide a group activity estimation equipment.

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

A first invention is a group behavior estimation device for estimating a behavior performed by a group existing in an environment, a group information storage means for storing humans in groups , and a group stored in the group information storage means Position history acquisition means for acquiring a position history for a predetermined time for each person belonging to the position history for each person belonging to the same group acquired by the position history acquisition means, the barycentric coordinates of the people included in the group, First feature quantity calculation means for calculating a first feature quantity indicating a feature quantity between humans , including a distance between the center of gravity coordinates and the position coordinates of each person and a moving speed of the center of gravity coordinates , and first feature quantity calculation means greater than the moving speed of the center of gravity coordinates calculated a predetermined speed by, and the average distance between the center of gravity coordinates and the human position coordinate than the predetermined distance When small, the group comprises the activity estimation means for estimating that the state of being moved together, a group activity estimation device.

In the first invention, the group behavior estimation device (10) includes a computer (30), and is based on the position information of the person (26) detected by the position detection system (12) installed in the environment. It is estimated what kind of behavior a group existing in the group is performing. The group information storage means (38) stores a plurality of persons in groups. The position history acquisition means (32, 36, 38, S7) acquires a position history for a predetermined time, for example, several seconds to several tens of seconds, of each person belonging to the group from the position detection system. Each person's position history data is represented by, for example, n + 1 point sequences on the xy plane including time information. The first feature value calculating means (32, S9) calculates a first feature value that is a feature value related to each other from the position history data of each person. For example, to calculate the barycentric coordinates of the people in the group, the distance between the center of gravity coordinates and the position coordinates between each person, and the moving speed of the gravity center coordinates and the like. Behavior prediction means (32, 40, S11) is that to estimate the behavior the group is performed based on the first feature amount. Specifically, when the movement speed of the center-of-gravity coordinates is higher than a predetermined speed and the average of the distance between the center-of-gravity coordinates and the position coordinates of each person is smaller than the predetermined distance, the action that the group is performing is moved together. Presumed to be in the “in” state.

According to the first invention, by detecting the position history of each person belonging to the group and using the feature quantity related to each other calculated from the position history, the group existing in the environment moves together. an action that is can be estimated in a short period of time. Therefore, it is possible to provide a service that quickly responds to group actions.

A second invention is a group behavior estimation device for estimating a behavior performed by a group existing in an environment, the group information storage means for storing humans in groups, and the group stored in the group information storage means Position history acquisition means for acquiring a position history for a predetermined time for each person belonging to the position history for each person belonging to the same group acquired by the position history acquisition means, the barycentric coordinates of the people included in the group, First feature quantity calculation means for calculating a first feature quantity indicating a feature quantity between humans, including a distance between the center of gravity coordinates and the position coordinates of each person and a moving speed of the center of gravity coordinates, and first feature quantity calculation means The movement amount of the center of gravity coordinates calculated by the above is smaller than the predetermined amount, and the average reduction amount of the distance between the center of gravity coordinates and each human position coordinate is the predetermined amount. When it is large, and a behavior estimation means for estimating that the state gather at a point where there is a group, a group activity estimation device.

In the second invention, the group behavior estimation device (10) includes a computer (30), and based on the position information of the person (26) detected by the position detection system (12) installed in the environment, It is estimated what kind of behavior a group existing in the group is performing. The group information storage means (38) stores a plurality of persons in groups. The position history acquisition means (32, 36, 38, S7) acquires a position history for a predetermined time, for example, several seconds to several tens of seconds, of each person belonging to the group from the position detection system. Each person's position history data is represented by, for example, n + 1 point sequences on the xy plane including time information. The first feature value calculating means (32, S9) calculates a first feature value that is a feature value related to each other from the position history data of each person. For example, the centroid coordinates of people included in the group, the distance between the centroid coordinates and the position coordinates of each person, the movement speed of the centroid coordinates, and the like are calculated. The behavior estimation means (32, 40, S11) estimates the behavior performed by the group based on the first feature amount. Specifically, when the movement amount of the centroid coordinates is smaller than a predetermined amount and the average reduction amount of the distance between the centroid coordinates and the position coordinates of each person is larger than the predetermined amount, the action performed by the group is “ It is estimated that it is in a state of “gathering at a certain point”.

According to the second invention, by detecting the position history of each person belonging to the group and using the feature quantity related to each other calculated from the position history, the groups existing in the environment gather at a certain point. The behavior of coming can be estimated in a short time. Therefore, it is possible to provide a service that quickly responds to group actions.

A third invention is a group behavior estimation device for estimating a behavior performed by a group existing in an environment, wherein the group information storage means for storing humans in groups and the group stored in the group information storage means Position history acquisition means for acquiring a position history for a predetermined time for each person belonging to the position history for each person belonging to the same group acquired by the position history acquisition means, the barycentric coordinates of the people included in the group, First feature quantity calculation means for calculating a first feature quantity indicating a feature quantity between humans, including a distance between the center of gravity coordinates and the position coordinates of each person and a moving speed of the center of gravity coordinates, and first feature quantity calculation means The movement amount of the center of gravity coordinates calculated by the above is smaller than the predetermined amount, and the average increase amount of the distance between the center of gravity coordinates and each human position coordinate is the predetermined amount. When it is large, and a behavior estimation means for estimating that the state of the group is gradually scattered, a group activity estimation device.

In the third invention, the group behavior estimation device (10) includes a computer (30), and based on the position information of the person (26) detected by the position detection system (12) installed in the environment, It is estimated what kind of behavior a group existing in the group is performing. The group information storage means (38) stores a plurality of persons in groups. The position history acquisition means (32, 36, 38, S7) acquires a position history for a predetermined time, for example, several seconds to several tens of seconds, of each person belonging to the group from the position detection system. Each person's position history data is represented by, for example, n + 1 point sequences on the xy plane including time information. The first feature value calculating means (32, S9) calculates a first feature value that is a feature value related to each other from the position history data of each person. For example, the centroid coordinates of people included in the group, the distance between the centroid coordinates and the position coordinates of each person, the movement speed of the centroid coordinates, and the like are calculated. The behavior estimation means (32, 40, S11) estimates the behavior performed by the group based on the first feature amount. Specifically, when the movement amount of the centroid coordinates is smaller than a predetermined amount and the average increase amount of the distance between the centroid coordinates and the position coordinates of each person is larger than the predetermined amount, the action performed by the group is “ Estimated to be in a “scattering” state.

According to the third invention, the position history of each person belonging to the group is detected, and the feature amount related to the person calculated from the position history is used, so that the groups existing in the environment are scattered. The behavior can be estimated in a short time. Therefore, it is possible to provide a service that quickly responds to group actions.

  According to the present invention, by detecting the position history of each person belonging to the group and using the feature amount related to each other calculated from the position history, the action performed by the group existing in the environment can be performed in a short time. Can be estimated. Therefore, it is possible to quickly provide a service corresponding to the behavior of the group to the group.

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

It is an illustration figure which shows one Example of the service provision system of this invention. It is an illustration figure which shows the mode of the environment where the position detection system of FIG. 1 was applied roughly. It is an illustration figure which shows the electrical structure of the position detection system of FIG. It is an illustration figure which shows an example of the position history data which the position detection system of FIG. 1 detects. It is an illustration figure which shows an example of the group action "move together." It is an illustration figure which shows an example of the group action "gathering at a certain point." It is an illustration figure which shows an example of a group action "scattering". It is an illustration figure which shows an example of the group action which is "stray". It is an illustration figure which shows an example of the group action of "chasing". It is a flowchart which shows an example of operation | movement of the group action estimation apparatus of FIG.

Referring to FIG. 1, a service providing system (hereinafter referred to as “providing system”) 100 according to an embodiment of the present invention is a group behavior estimating device (hereinafter referred to as “estimating device”) 10.
And a position detection system 12. The providing system 100 is applied to an environment (event venue, shopping center, art museum, amusement park, etc.) in which a plurality of people visit as a group and act together or act individually, and exists in the environment. Estimate what the group is doing as a group. Based on the estimated behavior of the group, the robot 14 and the like are controlled to provide a service suitable for the behavior to the group. A group is a group of people who are connected by some kind of mutual relationship, and includes family members, lovers and friends, and tour customers who participate in the same tour.

  First, the position detection system 12 will be described. FIG. 2 is an illustrative view schematically showing an environment where the position detection system 12 is applied, and FIG. 3 is an illustrative view showing an electrical configuration of the position detection system 12. As shown in FIGS. 2 and 3, the position detection system 12 includes a computer 20 that performs a position detection process, a plurality of laser range finders (LRF) 22 and a wireless ID tag reader 24 installed in the environment, and each person 26. The wireless ID tag 28 attached to the is included.

The computer 20 of the position detection system 12 is a general-purpose computer such as a personal computer or a workstation. This computer 20 includes, for example, RS-
A plurality of LRFs 22 and wireless ID tag readers 24 are connected via a general-purpose interface such as 232C. Data including information detected by these devices 22 and 24 is transmitted to the computer 20 and is appropriately stored in a predetermined area of the memory of the computer 20 or an external database together with the detection time. The memory of the computer 20 stores in advance position data of the devices 22 and 24 represented by an XY two-dimensional plane coordinate system and map data of the environment.

The LRF 22 is an image sensor using a laser beam, and irradiates a pulse laser, and instantaneously measures the distance from the object from the time when the laser light is reflected back to the object. As LRF22, LRF (model LMS 200) manufactured by SICK or the like can be used. The LRF 22 is installed, for example, on a pedestal having a height of about 90 cm, and is used to detect the position coordinates of the human 26 existing in the environment.

In this embodiment, the computer 20 acquires sensor information from the LRF 22 at a frequency of, for example, 37.5 times / second, and estimates the position coordinates of the waist of the human 26, the body direction, and the like using a particle filter and a human shape model. . The details of this estimation method are described in Japanese Patent Application No. 2008-6105 filed earlier by the present applicant.

  The wireless ID tag 28 is attached to each person 26 present in the environment. As the wireless ID tag 28, for example, an RFID (Radio Frequency Identification) tag can be used. RFID is an automatic recognition technology using a non-contact IC tag using electromagnetic waves. Specifically, the RFID tag includes an IC chip, an antenna, and the like provided with a memory for identification information, a control circuit for communication, and the like. Information that can identify the person 26 is stored in advance in the memory of the RFID tag, and the identification information is output from the antenna by electromagnetic waves or radio waves having a predetermined frequency.

  The wireless ID tag reader 24 is installed on, for example, a rail laid on the ceiling in the environment, and detects output information from the wireless ID tag 28 described above. Specifically, the wireless ID tag reader 24 receives a radio wave on which the identification information transmitted from the wireless ID tag 28 is superimposed via an antenna, amplifies the radio signal, and separates the identification information from the radio signal. The information is demodulated (decoded). The wireless ID tag reader 24 detects the distance from the wireless ID tag 28 based on the detected radio wave intensity.

  In this embodiment, the computer 20 acquires sensor information (identification information and distance information) from the wireless ID tag 28 detected by the wireless ID tag reader 24 at a frequency of, for example, 5 times / second. And based on the acquired identification information, the person 26 who possesses the wireless ID tag 28 is specified. Further, a person who has the wireless ID tag 28 based on the distance information acquired from the three wireless ID tag readers 24 for the same wireless ID tag 28 and the position data of the wireless ID tag reader 24 stored in advance. 26 position coordinates are specified.

  Then, the computer 20 of the position detection system 12 integrates the sensor information (positioning data) detected by the devices 22 and 24 or the information estimated from the sensor information, and determines the position coordinates (x, y) of each person 26. The data is stored in a memory or database in association with the detection time t. That is, for each person 26 (identification information or ID number), the position history is stored as time series data. For example, 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. That is, as shown in FIG. 4, the position history data (movement trajectory data) of each person 26 is represented by n + 1 point sequences on the xy plane, and each point (indicated by a circle in FIG. 4) is , Including information on the time at which the point was observed.

  The position detection system 12 is not limited to the above-described configuration as long as it can identify each person 26 and detect position coordinates or position history. For example, an infrared sensor, a video camera, a floor sensor, GPS or the like may be used as appropriate.

  Based on the position history data of each person 26 transmitted from the position detection system 12 as described above, the estimation apparatus 10 estimates an action performed by a group existing in the environment.

  Returning to FIG. 1, the estimation apparatus 10 includes a computer 30 that calculates a feature amount from position history data and estimates a group behavior. The computer 30 is, for example, a general-purpose personal computer or workstation including a CPU 32, a memory 34, a communication device 36, an input device, a display device, and the like. The memory 32 of the computer 30 includes an HDD, a ROM, and a RAM. A program for controlling the operation of the estimation apparatus 10 is stored in the HDD or ROM, and the CPU 32 executes processing according to this program. The RAM is used as a work area or buffer area for the CPU 32. The communication device 36 is for communicating with an external computer (such as the position detection system 12 or the robot 14) wirelessly or via a wire such as a wireless LAN or the Internet.

  Further, a group information database (DB) 38 and a discriminant DB 40 are connected to the computer 30 of the estimation apparatus 10. These DBs 38 and 40 may be provided outside the computer 30 as shown in FIG. 1, or may be provided in an HDD or the like in the computer 30. The DBs 38 and 40 may be provided in other computers on a network that can communicate with the computer 30.

The group information DB 38 stores a plurality of humans 26 in groups. Specifically, in the group information DB 38, the identification information of the wireless ID tag 28 attached to the person 26 and the group ID number of the group to which the person 26 belongs are associated with the ID number of the person 26 in advance. be registered. Also, in the group information DB 38, user information such as the type of the group (family, friends, etc.) and the age of the person 26 is registered as appropriate according to the environment to which the providing system 100 is applied. Therefore, the computer 30 refers to the group information DB 38 to determine, for example, which group 26 the data transmitted from the position detection system 12 belongs to, and whether the group exists in the environment. Can be recognized.

  The discriminant DB 40 stores a discriminant for estimating an action performed by a group existing in the environment based on the feature amount calculated from the position history data of each person 26. When estimating the action performed by the group, a feature amount described later is appropriately calculated according to the discriminant stored in the discriminant DB 40. A specific example of this discriminant will be described later.

  In the providing system 100 having such a configuration, as described above, the position detection system 12 detects data relating to the position coordinates or position history of the person 26 existing in the environment. Then, a feature amount is calculated from the position history data by the estimation device 10, and a group action is estimated based on the feature amount.

  The feature amount calculated from the position history data by the estimation device 10 includes a feature amount (first feature amount) between persons belonging to the same group and a feature amount (second feature amount) related to each individual.

  The features between people include the centroid coordinates of the people in the group, the average and variance of the distance between the centroid coordinates and the position coordinates of each person, the average movement speed of the centroid coordinates, and the There are distance average and variance.

In addition, the feature amount related to each individual includes, for example, a feature amount related to distance, speed, angle, and variance. Referring to FIG. 4, the feature quantity related to the distance is the Euclidean distance, the x-direction distance, the y-direction distance, and the like between the 0th point and the kth point. The feature quantity related to speed is the speed between the 0th point and the nth point, for example, the speed between the kth point and the k + 1st point is obtained from k = 0 to k = n. The mean and variance of the food. The feature quantity related to the angle is an angle formed by the 0th point, the kth point, and the nth point. The feature amount related to the variance (value variation) includes the variance of the x coordinate and the variance of the y coordinate from the 0th point to the kth point.

  Of course, feature quantities other than these can be calculated and used for estimation of group behavior.

  In this embodiment, the above-described feature amounts are used as appropriate, and “moving together”, “collecting at a certain point”, “scattering”, “stray”, and “chasing”. ”Or the like (or the situation or state of the group). Hereinafter, the group behavior estimation method will be described with reference to FIGS. 5 to 9 exemplify the case where the number of people constituting the group is three (A, B, C), but the number of people constituting the group may be two or four or more. Good.

As shown in FIG. 5, the group is “moving together” in a certain period of time (for example, several seconds to several tens of seconds from time t0 to tn). It is considered that the people involved are in a state where they move within a predetermined range without being separated from each other. Therefore, the centroids of the people included in the group are moving at a speed greater than a predetermined speed (SpeedTH), and the average of the distance between the centroid coordinates of the people included in the group and the position coordinates of each person is the predetermined distance (DistTH ), It can be determined that the group is moving together. In this case, the discriminant DB 40 describes, for example, a discriminant representing “the moving speed of the center of gravity> SpeedTH and the distance average between each person and the center of gravity coordinates <DistTH”.

As shown in FIG. 6, in the state of “gathering at a certain point”, the center-of-gravity coordinates of the people included in the group do not change so much during a certain period of time while the action is being performed. It is considered that the position coordinates and the barycentric coordinates are close to each other. Therefore, when the movement amount of the centroid coordinates of the people included in the group is smaller than a predetermined amount (MoveTH) and the average decrease amount of the distance between each human position coordinate and the centroid coordinates is larger than a predetermined amount (DecreseTH) It can be determined that the group gathers at a certain point. In this case, the discriminant DB 40 describes, for example, a discriminant representing “the amount of movement of the center of gravity coordinates <MoveTH and the amount of average distance reduction between each person and the center of gravity coordinates> DecreseTH”.

  As shown in FIG. 7, in the “scattering” state, the coordinates of the center of gravity of the people included in the group do not change so much during a certain time during the action, It is considered that the center of gravity coordinates are in a state of separating. That is, when the movement amount of the centroid coordinates of the people included in the group is smaller than the predetermined amount (MoveTH) and the average increase amount of the distance between each human position coordinate and the centroid coordinates is larger than the predetermined amount (IncreseTH) It can be determined that the group is scattered. In this case, the discriminant DB 40 describes, for example, a discriminant representing “the amount of movement of the center of gravity coordinates <MoveTH, and the amount of increase in distance average between each person and the center of gravity coordinates> IncreseTH”.

As shown in FIG. 8, the “stray” state is a state in which a person constituting the group is divided into two small groups (A in FIG. 8). And C and B are separated). That is, when two small groups exist in a group and the distance between these small groups is larger than a predetermined distance (DistGroupTH), it can be determined that the groups are separated. In this case, the discriminant DB 40 describes, for example, a discriminant representing “the number of small groups in the group = 2 and the distance between the small groups> DistGroupTH”.

As shown in FIG. 9, in the state of “chasing”, the humans constituting the group are divided into two small groups at a certain time during the action. It is considered that the movement direction is the same and the distance between the small groups is getting closer. That is, there are two small groups in the group, the angle between the movement directions of the two small groups is smaller than the predetermined angle (AngleTH), and the amount of decrease in the distance between the small groups is the predetermined amount (DecreseGroupTH) When it is larger than that, it can be determined that the player is chasing. In this case, the discriminant DB
In 40, for example, a discriminant representing “the number of small groups in a group = 2 and the angle between the movement directions of the small groups <AngleTH and the distance between the small groups> DistGroupTH” is described.

  Note that the above-described group behavior and its discriminant are merely examples, and the estimation device 10 (providing system 100) may estimate other group behavior, or use other discriminant to determine the group behavior. You may make it estimate. Further, the present invention is not limited to estimating the group behavior by the discriminant defined using the threshold values as described above. For example, a well-known learning algorithm such as a support vector machine (SVM) or a neural network (NN) and learning data are used. And can be used to estimate group behavior. Also in this case, the above-described feature amount calculated from the position history data of each person 26 is appropriately used.

  In this way, by detecting the position history of each person 26 belonging to the group, calculating the feature quantity related to each other and the person related to each person belonging to the same group, and appropriately giving these feature quantities to the discriminant , It is possible to accurately estimate the actions performed by groups present in the environment. The group behavior can be estimated using only the feature values related to each other. However, by using the feature values related to each person together with the feature values related to each other, more accurate and more You can estimate the type of group behavior.

In addition, since such group behavior estimation can be performed by simply detecting the position history of each person 26 for a short time of several seconds to several tens of seconds, a service corresponding to the behavior of the group can be provided quickly. .

  For example, an interaction-oriented robot 14 having a function of executing communication using voice and body motion is placed in the environment, and an instruction is given to the robot 14 in accordance with the estimated group behavior. A suitable service may be provided to the robot 14. As the robot 14, Roboby (registered trademark) developed by the present applicant can be used.

  Specifically, when the group moves together or gathers in a specific place, it is preferable to control the conversation content of the robot 14 to talk to the group, not to the individual. . In addition, when a group gathers at a certain point, it is preferable to move the robot 14 to the waiting point in accordance with the timing when the group gathers so as to provide a service to the group. Thereby, useless movement of the robot 14 can be eliminated. In addition, if someone is missing in the group, for example, if a child is missing, you can talk to the child and guide him to his / her location, or tell the parent who is looking for the child where their child is. Service can be provided.

  Below, an example of operation | movement of the above estimation apparatuses 10 is demonstrated using a flowchart. Specifically, the CPU 32 of the computer 30 of the estimation device 10 executes behavior estimation processing according to the flowchart shown in FIG. As shown in FIG. 10, the CPU 32 of the computer 30 determines whether or not a group exists in the environment in step S1. For example, the identification information of the wireless ID tag 28 existing in the environment is acquired from the position detection system 12 via the communication device 36, and the person 26 belonging to the same group exists in the environment with reference to the group information DB 38. Determine whether or not. If “NO” in the step S1, that is, if there is no group in the environment, the action estimating process is ended as it is. On the other hand, if “YES” in the step S1, that is, if a group exists in the environment, the process proceeds to a step S3.

  In step S3, a variable X indicating a group ID is initialized (X = 0). For example, when the group ID of the group confirmed to exist in step S1 is set to 0 to L, X = 0 is set. In a succeeding step S5, it is determined whether X = L. That is, it is determined whether or not the behavior has been estimated for all groups existing in the environment. If “YES” in the step S5, that is, if behavior estimation is performed for all groups existing in the environment, the behavior estimation processing is ended. On the other hand, if “NO” in the step S5, that is, if the behavior estimation is not performed for all the groups existing in the environment, the process proceeds to the step S7.

  In step S7, a position history for a predetermined time of the person 26 belonging to the group having the group ID X (group X) is acquired. Specifically, the position history of each person 26 belonging to the group X is acquired from the position detection system 12 via the communication device 36 for several seconds to several tens of seconds from time t0 to tn, for example. At this time, for example, the position coordinates of the person 26 whose ID number is i and whose group ID number is k when the time is j are expressed by the expression Traj_ijk (i = 0... M, j = t0... Tn, k = The position history data of each person 26 is represented by n + 1 point sequences on the xy plane including time information, as shown in FIG.

In the subsequent step S9, the feature quantity (second feature quantity) related to each individual and the feature quantity (first feature quantity) related to each other are calculated from the position history data acquired in step S7. For example, a feature amount related to distance, speed, angle, and variance is calculated as a feature amount related to each individual, and a distance between the barycentric coordinates of people included in the group and the position coordinates of each person 26 is calculated as a feature amount related to each other. The average and the movement speed of the barycentric coordinates are calculated. The feature amount calculated in step S9 is appropriately determined according to the discriminant stored in the discriminant DB 40.

  In the subsequent step S11, the feature amount calculated in step S9 is given to the discriminant read from the discriminant DB 40, and the action performed by the group is estimated. For example, when the movement speed of the centroid of the people included in the group> SpeedTH and the distance average between each person and the centroid coordinates <DistTH, it is determined that the group is moving together.

  In step S13, the behavior pattern of the group X is set to the group behavior calculated in step S11, for example, “move together”, and in step S15, the variable X is incremented and the process returns to step S5. The computer 30 provides a service suitable for the group behavior by giving an instruction to the robot 14 according to the behavior pattern set in step S13.

  According to this embodiment, groups existing in the environment perform by detecting the position history of each person and giving the feature quantity related to each individual and the feature quantity related to each other belonging to the same group to the discriminant. The behavior can be estimated in a short time. Therefore, a service corresponding to the behavior of the group can be provided quickly.

  In the above-described embodiment, the position detection system 12 and the estimation device 10 share the respective processes (position detection process, action estimation process, etc.) using different computers 20 and 30. One computer can perform these processes collectively, or three or more computers can further share the processes.

  In the above-described embodiment, the group information DB 38 stores a plurality of humans 26 in advance, but the present invention is not limited to this. For example, based on the detected position history of each person 26, it is automatically determined which persons 26 belong to the same group, and the group information obtained as a result is used for estimation of subsequent group behavior. May be. In this case, for example, if a plurality of persons exist within a predetermined distance for a predetermined time or more, it is determined that the plurality of persons belong to the same group, and are registered as a group in the group information DB 38. It is good to.

  Furthermore, when the number of persons 26 belonging to the same group is large, for example, in the case of a group composed of six or more persons, the group is further divided into small groups, and each person 26 is assigned a small group ID separately from the group ID number. A number may be assigned and stored in the group information DB 38. And it is good to estimate a group action about each of the whole group and a small group. In this case, similarly to the feature amount between humans, the feature amount between small groups can be calculated, and the behavior of the entire group can be estimated using the feature amount.

DESCRIPTION OF SYMBOLS 10 ... Group action estimation apparatus 12 ... Position detection system 14 ... Robot 20 ... Position detection system computer 22 ... Laser range finder 24 ... Wireless ID tag reader 28 ... Wireless ID tag 30 ... Group action estimation apparatus computer 38 ... Group information DB
40 ... Discriminant DB
100 ... service providing system

Claims (3)

A group behavior estimation device that estimates a behavior performed by a group existing in an environment,
Group information storage means for storing humans in groups,
Position history acquisition means for acquiring a position history for a predetermined time for each of the persons belonging to the group stored in the group information storage means;
From the position history of each person belonging to the same group acquired by the position history acquisition means, the centroid coordinates of people included in the group, the distance between the centroid coordinates and the position coordinates of each person, and the movement speed of the centroid coordinates A first feature amount calculating means for calculating a first feature amount indicating a feature amount related to humans, and a moving speed of the barycentric coordinates calculated by the first feature amount calculating means is greater than a predetermined speed, and A group behavior estimation apparatus comprising behavior estimation means for estimating that the group is moving together when the average distance between the barycentric coordinates and the position coordinates of each person is smaller than a predetermined distance . A group behavior estimation device that estimates a behavior performed by a group existing in an environment,
Group information storage means for storing humans in groups,
Position history acquisition means for acquiring a position history for a predetermined time for each of the persons belonging to the group stored in the group information storage means;
From the position history of each person belonging to the same group acquired by the position history acquisition means, the centroid coordinates of people included in the group, the distance between the centroid coordinates and the position coordinates of each person, and the movement speed of the centroid coordinates A first feature amount calculating means for calculating a first feature amount indicating a feature amount related to humans, and
When the movement amount of the barycentric coordinates calculated by the first feature amount calculating means is smaller than a predetermined amount and the average reduction amount of the distance between the barycentric coordinates and each human position coordinate is larger than a predetermined amount, A group behavior estimation device comprising behavior estimation means for estimating that a group is gathering at a certain point. A group behavior estimation device that estimates a behavior performed by a group existing in an environment,
Group information storage means for storing humans in groups,
Position history acquisition means for acquiring a position history for a predetermined time for each of the persons belonging to the group stored in the group information storage means;
From the position history of each person belonging to the same group acquired by the position history acquisition means, the centroid coordinates of people included in the group, the distance between the centroid coordinates and the position coordinates of each person, and the movement speed of the centroid coordinates A first feature amount calculating means for calculating a first feature amount indicating a feature amount related to humans, and
When the movement amount of the barycentric coordinates calculated by the first feature amount calculating means is smaller than a predetermined amount, and the average increase amount of the distance between the barycentric coordinates and each human position coordinate is larger than a predetermined amount, A group behavior estimation apparatus comprising behavior estimation means for estimating that a group is scattered.

Images