JP5223049B2 - Individual recognition device and individual recognition system - Google Patents

Description

The present invention relates to an individual recognition apparatus and an individual recognition system for performing current position and movement prediction of a plurality of various objects.

A laser range finder that measures a two-dimensional environment shape is realized by rotating a sensor that measures the distance according to the time it returns by reflecting a laser beam on the object surface and returning. When this sensor is used, the number and size of moving objects can be measured in real time, but the type and name of the object cannot be recognized.
On the other hand, the IC tag information is associated with data such as the type and name of the object in a database, this is attached to the object, and the IC tag information and the radio wave intensity are read with an IC tag reader, so that the rough distance and recognition of the object A system has been devised, but it was impossible to recognize the exact position.
In order to solve this, a recognition system combined with a laser range finder has been proposed, but there is a problem that a plurality of objects cannot be recognized individually at a high speed.
Patent Publication 2007-148595 (Title of Invention: Mobile)

Individual recognition by image processing allows individual recognition based on the shape and color of the object. However, since the camera is pointed in the direction of the object, focused and photographed, image processing is performed. It is difficult to make them individually recognize.
A laser range finder that measures a two-dimensional environment shape is realized by rotating a sensor that measures the distance according to the time it returns by reflecting a laser beam on the object surface and returning. When this sensor is used, the number and size of moving objects can be measured in real time with the sensor at the center, and this sensor can be used for recognizing people, vehicles, luggage, and the like. However, only the shape on the laser incident direction side of the cross section cut by a certain horizontal plane with respect to the object can be read, and it is impossible to recognize the type and name of the object only by this data.
On the other hand, by associating IC tag information with data such as the type and name of an object in a database, pasting it on the object, and reading the IC tag information and radio wave intensity with an IC tag reader, the system to the object can be roughly classified. Although a system that performs accurate distance and recognition has been devised, it has been impossible to recognize the exact position and movement vector of an object.
In order to solve this problem, a recognition system combined with a laser range finder has been proposed. However, when a large number of objects exist, there is a problem that individual recognition cannot be performed.
The problem to be solved by the present invention is to provide an apparatus and system that solves the above-described problems and recognizes a large number of objects at high speed.

The configuration of the present invention that solves such a problem is as follows: 1) An individual recognition apparatus that automatically identifies each object and predicts individual movements with respect to a plurality of objects attached with IC tags having identification information recorded thereon. A laser range finder that is set at the center of sensing and measures the environmental shape around the sensor, a database that stores information recorded in an IC tag attached to each object, and a Information accumulated in the database from IC tag readers that can read distances, environmental shape changes obtained with the laser range finder, IC tag information obtained with the IC tag readers, and distance changes from the system center to the objects. Based on the above, the object information, position, and movement vector to which each IC tag is attached are calculated, the movement of each object is predicted, and each pair is calculated. It is an individual recognition apparatus characterized by comprising a discriminator for recognizing an object.
2) An individual recognition device that automatically identifies each object and performs individual movement prediction for a plurality of objects to which IC tags with identification information recorded are attached, and is set at the center of sensing. Obtained with a laser range finder that measures the environmental shape around the sensor, a database that stores information recorded in the IC tag of each object, an IC tag reader that can read the distance of each object from the center of the system, and the laser range finder The target to which each IC tag is attached based on the information accumulated in the database from the change in environmental shape, the IC tag information obtained by the IC tag reader, and the distance change from the system center to each object. An individual recognition device consisting of a discriminator that calculates object information, position, and movement vector, performs movement prediction of each object, and recognizes each object. The IC tag information and the IC tag object information attached to the plurality of objects are managed in the database, and the environmental shape change obtained by the laser range finder is obtained by the IC tag reader. An individual recognition system characterized by calculating object information, position, and movement vector to which each IC tag is attached from a change in distance from the IC tag information and the system center, and performing movement prediction for recognizing each object. It is.

According to the present invention, a laser range finder capable of acquiring a partial cross-sectional shape of an object in real time, an IC tag in which object identification information is recorded, and a database for managing the relationship between the identification information of the IC tag and the object information , And calculating the relative movement relationship between the system and the target object, the individual recognition of a large number of objects attached with IC tags and the movement vector without performing image processing with a large amount of calculation processing. Realization of high-speed recognition in real time has become possible.

An IC tag reader antenna was installed on the center axis of the laser range finder so that the measurement center of the laser range finder coincided with the center of the radio wave intensity distribution of the IC tag information. These two sensor data were sent to the computer, and the computer performed individual authentication of multiple objects and movement prediction. Embodiments of the present invention will be specifically described below with reference to the drawings.

FIG. 1 is an explanatory view of a system combining a laser range finder, an IC tag reader antenna, and a discriminator as viewed from directly above. FIG. 2 is an explanatory view of a system including a laser range finder, an IC tag reader antenna, and a discriminator as viewed from the side. FIG. FIG. 3 is an explanatory diagram of the system when viewed from directly above and the IC tag reader reading intensity distribution. FIG. 4 is an explanatory diagram of the shape information of the laser range finder and the IC tag reader reading intensity in the pre-movement state of an example in which a plurality of objects are moving relative to the system. FIG. 5 is an explanatory diagram of the shape information of the laser range finder and the IC tag reader reading intensity in the post-movement state in an example in which a plurality of objects are moving relative to the system. Table 1 is an explanatory table showing a database that associates an object name with an IC tag code of an IC tag attached to each object. Table 2 explains changes in the reading intensity of the IC tag code attached to each object when the five objects move relative to the individual recognition apparatus as shown in FIGS. 4 and 5. Table 3 shows that when five objects move relative to the individual recognition device as shown in FIGS. 4 and 5, each convex shape of the laser range finder shape information before and after the movement exists in any region of the IC tag reader reading intensity distribution. It shows how. Embodiments according to the present invention will be described below with reference to the drawings and tables.

As shown in FIG. 2, the object recognition apparatus of the present embodiment measures the environmental shape information around the system by rotating a detection unit that emits a laser and measures the distance to the object in a 360-degree direction. An IC tag reader antenna 5 having IC tag information inside the IC tag affixed to each of the objects 2a to 2e and a means capable of acquiring the read radio wave intensity is disposed directly above the laser range finder 4 provided with A means for recording and managing a database that associates information such as the name, attribute, shape, and weight of an object with IC tag information, and a means for integrating the information to calculate object recognition and each object movement vector The discriminator 6 is used.

FIG. 3 shows an IC tag information reading intensity distribution when the system is viewed from directly above. From the IC tag reader antenna 5, the IC tag reading intensity 1 is from the area boundary 3a of the IC tag reading intensity 1 to the area boundary 3a of the IC tag reading intensity 1 to the area boundary 3b of the IC tag reading intensity 2 is IC tag reading intensity 2. The IC tag reading intensity 3 is defined from the area boundary 3b of the IC tag reading intensity 2 to the area boundary 3c of the IC tag reading intensity 3.

As an example, assume that five objects are recognized. IC tags were previously attached to the respective objects 2a, 2b, 2c, 2d and 2e. Further, the IC tag information database of each of the objects 2a, 2b, 2c, 2d and 2e shown in Table 1 is registered in the discriminator 6.

First, when each object 2a, 2b, 2c, 2d, 2e is within the system recognition range, each object 2a, 2b, 2c, 2d, 2e exists by the laser range finder 4 as shown in FIG. Since the positions are detected as laser range finder detection shapes 1a, 1b, 1c, 1d, and 1e, and the IC tag reading intensity distribution is known in advance as shown in FIG. 3, the IC tag at each laser range finder detection shape position. Reading intensity could be estimated. At the same time, each IC tag information and the reading intensity of each IC tag can be measured by the IC tag reader antenna 5.
Next, when the object moves relative to the system, as shown in FIG. 5, the post-movement positions of the objects in the laser range finder detection shapes 1a, 1b, 1c, 1d, and 1e can be measured, and the laser shown in FIG. The movement vector of each object 2a, 2b, 2c, 2d, 2e could be measured by comparing with the position before the movement of the range finder detection shape. At the same time, the IC tag reading intensity at each laser range finder detection shape position was estimated, and the IC tag information and the IC tag reading intensity were measured by the IC tag reader antenna 5. By the above measurement, as shown in Tables 2 and 3, the reading strength of each IC tag before and after the movement of the object and the laser range finder detection shapes 1a, 1b, 1c, 1d, and 1e of the object. IC tag reading strength was found.

To identify each object 2a, 2b, 2c, 2d, 2e from the measurement results, as shown in Tables 2 and 3, each IC attached to each object 2a, 2b, 2c, 2d, 2e Correspondence between each laser range finder detection shape and IC tag by selecting the one with the same reading intensity change from the reading intensity change of the tag and the reading intensity changes of the object convex shapes 1a, 1b, 1c, 1d, 1e. Got. Furthermore, each object name was able to be obtained from this result and the IC tag information database shown in Table 1. As a result, the names, positions, and movement vectors of all objects were found.

As an example, in order to specify the position and movement vector of the second object 2b, first, referring to Table 1, when searching for the IC tag code of the IC tag attached to the second object 2b, it is ID5. It turns out.
Next, referring to Table 2, when the IC tag reading intensity before and after the relative movement of the IC tag code ID5 is searched, it is found that the reading intensity before movement is 3 and the reading intensity after movement is 1.
Next, referring to Table 3, when a laser range finder detection shape having a pre-movement reading intensity of 3 and a post-movement reading intensity of 1 was searched, it was found that the second laser range finder detection shape 1d was applicable. As a result, it was found that the second object 2b was at the position of the second laser range finder detection shape 1d, and the position was found. At the same time, the movement vector of the second laser range finder detection shape 1d can be calculated from FIGS. 4 and 5, and the movement vector of the second object 2b is also found.

Conversely, as an example of identifying what the object is from the laser range finder shape, the object name of the fourth laser range finder shape 1b is identified. First, referring to Table 3, when the IC tag reading intensity before and after movement is searched, it is found that the reading intensity before movement is 3 and the reading intensity after movement is 2.
Next, referring to Table 2, when an IC tag code having a pre-movement IC tag reading intensity of 3 and a post-movement reading intensity of 2 is searched, it is found that the IC tag code is ID1.
Next, referring to Table 1, when searching for the name of the object to which the tag having the IC tag code ID1 is attached, it is found that the object is the second object 2d. As a result, the fourth laser range finder detection shape 1b was found to be the second object 2d.

The reading intensity of the IC tag attached to each of the objects 2a, 2b, 2c, 2d, and 2e may vary depending on the surrounding environmental conditions even at the same reading distance. In order to improve the identification accuracy of each object 2a, 2b, 2c, 2d, and 2e, the correlation calculation of Table 2 and Table 3 can be expanded not only before and after movement but also three times or more. In addition, when a movement vector that is not physically considered is detected, it is regarded as an error and can be excluded from the correlation calculation. In addition, the reading intensity is expressed by a numerical model that includes not only absolute values but also before and after.

The detection information of the laser range finder 4 may instantaneously output a shape different from the actual depending on conditions such as the surface reflectance of the object. In order to improve the identification accuracy of each of the objects 2a, 2b, 2c, 2d, and 2e, the correlation calculation in Tables 2 and 3 was expanded not only before and after the movement but also three times or more. In addition, if a motion vector that is not physically considered is detected, it is regarded as an error and excluded from the correlation calculation.

When there is a stationary object around which the IC tag is not attached around the recognition device, the shape of the stationary object is included in the detection information of the laser range finder 4, so that it may not be discriminated from the object to which the IC tag is attached. The shape of a stationary object detected by the laser range finder 4 is recorded in advance in the discriminator 6, and at the time of recognition of an object attached with an IC tag, the recorded stationary object is determined from the shape information detected by the laser range finder 4. Thus, only the shape of the object to which the IC tag is attached can be extracted.

The recognition apparatus according to the present invention includes indoor authentication with a high security level, a guidance robot mounted on a mobile robot, a cell production system that recognizes a large number of parts by mounting on a robot hand, vehicle, aircraft, ship This is useful in fields where a large number of objects must be recognized, such as recognition.

It is the top view which looked at the individual recognition apparatus from the top. It is a side view of an individual recognition apparatus. It is explanatory drawing which looked at the sensor part of the individual recognition apparatus of an Example, and IC tag reading intensity distribution from the top. The relationship between the laser range finder detection shape and the IC tag reading intensity distribution in the pre-movement state when the five target objects are moving relative to the sensor unit of the individual recognition device of the embodiment from above FIG. The relationship between the laser range finder detection shape and the IC tag reading intensity distribution in the post-movement state when the five target objects are moving relative to the sensor unit of the individual recognition device of the embodiment from above FIG.

Explanation of symbols

1a Laser range finder detection shape 1b of the fifth object Laser range finder detection shape 1c of the fourth object Laser range finder detection shape 1d of the first object Laser range finder detection shape 1e of the second object Laser range finder detection shape 2a of the object 3a 1st object 2b 2nd object 2c 3rd object 2d 4th object 2e 5th object 3a Area boundary 3b of IC tag reading intensity 1 IC tag reading intensity 2 area boundary 3c IC tag reading intensity 3 area boundary 4 Laser range finder 5 IC tag reader (antenna)
6 Discriminator

Claims (2)

An individual recognition device for automatically identifying each object and predicting individual movements with respect to a plurality of objects to which an IC tag having identification information recorded is attached. A laser range finder that measures the environmental shape of the object, a database that stores information recorded in an IC tag attached to each object, an IC tag reader that can read the distance of each object from the sensing center, and the laser range finder Each IC tag is attached based on the information accumulated in the database from the change in the environmental shape obtained in step 1, the IC tag information obtained from the IC tag reader, and the distance change from the sensing center to each object. It consists of a discriminator that calculates the target object information, position, and movement vector, predicts the movement of each object and recognizes each object. Individual recognition apparatus according to claim Rukoto. An individual recognition device for automatically identifying each object and predicting individual movements with respect to a plurality of objects to which an IC tag having identification information recorded is attached. Obtained by the laser range finder for measuring the environmental shape of the above, a database storing information recorded on the IC tag of each object, an IC tag reader capable of reading the distance of each object from the sensing center, and the laser range finder The object to which each IC tag is attached based on the information accumulated in the database from the change in environmental shape, the IC tag information obtained by the IC tag reader, and the distance change from the sensing center to each object. Individual recognition consisting of a discriminator that calculates information, position, and movement vector, predicts movement of each object, and recognizes each object In the apparatus, the information of the IC tag and the object information of the IC tag attached to the plurality of objects are managed in the database, and the environmental shape change obtained by the laser range finder is obtained by the IC tag reader. Individual information characterized by calculating object information, position, and movement vector to which each IC tag is attached from IC tag information and a change in distance from the sensing center, and performing movement prediction for recognizing each object. Recognition system.

Images