KR20070072305A - Localization system for moving object using state group of moving object and method thereof - Google Patents

Abstract

A moving object localization system and method are provided to reduce the number of RFID(Radio Frequency Identification) tags for localizing a moving object by improving accuracy in the localization. A moving object localization system perceives movement of a moving object by obtaining position information from a plurality of RFID tags(200). The moving object includes an operation sensing unit(110) for acquiring odometry information while moving the moving object; an operation control unit(120) for driving the operation sensing unit by controlling the speed and direction of the operation sensing unit; an RFID tag reader unit(130) for receiving information stored in the RFID tags through communication with the RFID tags; and a computer unit(140) for controlling the operation sensing unit, the operation control unit and the RFID tag reader unit. The computer unit localizes the moving object by using a plurality of state points having state information of the moving object selected by the periodically received odometry information and tag information from the RFID tag reader unit.

Description

Localization System for Moving Object Using State Group of Moving Object and Method Thereof}

1 is a block diagram showing a moving object position estimation system to which the moving object state group according to the present invention;

2 is a view illustrating a coordinate system of an RFID tag according to the present invention;

3 is a diagram illustrating a relationship between an odometry coordinate system and an RFID coordinate system in a position estimation system according to the present invention;

4 is a flowchart for explaining a moving object position estimation method according to the present invention;

5A is a view showing an initial position estimation of a moving object according to the present invention;

5b is a view showing the movement of the moving object after the initial position estimation according to the present invention,

6 is a diagram illustrating a position estimation of a moving object when a tag is recognized according to the present invention;

7 is a view illustrating a position estimation of a moving object when tag is not recognized according to the present invention;

8 is a view showing a state point selected according to the movement of the moving object according to the present invention;

9 is a view showing selection of state points to be used for position estimation of a moving object according to the present invention.

Explanation of symbols on the main parts of the drawings

100: moving object 110: drive detection unit

120: drive control unit 130: RFID reader unit

140: computer unit 200: RFID tag

The present invention relates to a moving object position estimation system and a method for estimating the moving object state group, in particular, when the position of the regions that can recognize the movement of the moving object on the coordinate system can be known, The present invention relates to a moving object position estimation system and a method for estimating a moving object state group for estimating the position of a moving object using location information of an area and odometry information of a moving object.

In general, a technique using radio frequency identification (RFID) and odometry is used for position estimation.

First of all, RFID is a type of chip that attaches an electronic tag to an object so that the object can recognize the surrounding situation and exchange and process information with an existing IT system in real time. RFID can identify information about a tagged object with only an antenna and a tag, and can insert necessary information. The RFID system consists of three elements: a tag that electrically stores unique information in the RFID, a reader that acts as a transceiver, and a host server and an application program. The basic operation principle is that the antenna of the RFID and the antenna of the reader communicate with each other through radio waves to exchange data. First, the antenna embedded in the RFID tag receives radio waves from the reader. The IC chip embedded in the RFID receiving the radio wave starts to signal the information in the chip and transmits the signal through the antenna of the tag. The reader receives the signal transmitted from the RFID tag through the antenna in the reader to receive information, and the transmitted information is transmitted to the server by wired or wireless communication. RFID can achieve precise location measurement and consumption reduction by adjusting tag spacing according to user's needs.

As such, an example of a position measuring technology using RFID is disclosed in Korean Patent Laid-Open Publication No. 2003-0080436 (published on October 17, 2003, apparatus and method for measuring position of a mobile robot).

The technology disclosed in Korean Patent Laid-Open Publication No. 2003-0080436 is an RFID tag having unique location information, and obtains the location information from at least one sensor cell and the sensor cell installed in a work area of a mobile robot, thereby obtaining the mobile robot. It is composed of a sensor mounted on the mobile robot to measure the current position of the mobile robot, one sensor cell is installed in the working area of the mobile robot, and then the position information of the sensor cell is acquired through the sensor mounted on the mobile robot. It is described that it is possible to accurately measure the current position, the moving distance, the moving direction and the like of the mobile robot.

Next, odometry is a technology mainly used to recognize the position and direction of the robot, and recognizes information such as kinematic state, position, speed, acceleration, direction of the moving object. That is, the speed information is obtained by using an odometer or a wheel sensor of the moving object, the azimuth information is obtained by using a magnetic sensor, etc., and the information about the distance and the direction of the moving object is calculated. Recognize location and orientation. Odometry uses only its own information without receiving additional information from the outside. Therefore, the position information is acquired at a high sampling rate, and the update of the position information is also fast. In addition, high accuracy and low cost at short distances.

An example of a position recognition technology for solving the disadvantages of such an odometry is disclosed in Korean Patent Laid-Open Publication No. 2005-0091549 (published on September 15, 2005, a method for recognizing a position of a mobile robot using an odometry).

The technique disclosed in the Republic of Korea Patent Publication No. 2005-0091549 is a first step of moving any mobile robot from the start node to the end node along the sensor-based driving path, by detecting the number of wheel revolutions according to the movement of the mobile robot A second step of acquiring first odometry path information, a third step of moving the mobile robot located at an end node in the opposite direction along a sensor-based driving path, and wheel rotation according to movement in the opposite direction of the mobile robot A fourth step of obtaining a number of second geometry path information by detecting a number; recalculating the first and second geometry diagram information using an error model and an error variable of the mobile robot, and calculating the first and second geometry path information. A mobile robot according to a fifth step of acquiring 2 odometry correction path information, and the first and second odometry correction path information obtained in the fifth step The sixth step is to determine the location information, and it is possible to minimize the kinematic position inaccuracy of the mobile robot by accurately predicting the 'error variable' which minimizes the error after reciprocating the GVG (Generalized Voronoi Graph) two (2) times. It is described that the inaccuracy of the odometry due to the unpredictable misdimensional person after the accuracy improvement using the 'error variable' can be expressed mathematically.

However, in the technology disclosed in Korean Patent Laid-Open Publication No. 2003-0080436, since the accuracy of position and direction recognition of a moving object is determined according to the distribution density of the RFID tag, if the distribution density of the RFID tag is low, the position and direction of the moving object If the accuracy of recognition is low and the distribution density of the RFID tag is high, there is a problem that an error occurrence rate is increased due to mutual interference between RF signals.

In addition, in the technique disclosed in Korean Patent Laid-Open Publication No. 2003-0080436, there is a problem that the update on the new position and direction information is slow.

In addition, the technique disclosed in Korean Patent Laid-Open Publication No. 2005-0091549 has a problem in that measurement errors accumulate as the driving distance increases because the position and the direction are calculated using the integral.

In addition, in the related arts related to RFID and odometry and the techniques disclosed in the above publications, only the position information is corrected, but the attitude error of the moving object is not corrected.

An object of the present invention is to solve the problems described above, the moving object state that divides the recognizable area and improves the accuracy of the position and attitude estimation of the moving object by using the position information of the area and the odometry information of the moving object. It is to provide a moving object location system and method applying the group. That is, even after the moving object belongs to any one of the recognizable areas, the moving object has the coordinate history of the moving object and the position and posture of the moving object on the coordinate system even if the moving object moves only to the errored odometry. The state group is used to estimate the current position of the moving object.

Another object of the present invention is to provide a moving object position estimation system and method applying a moving object state group in which the advantages of both technologies are integrated by obtaining information from RFID and odometry. In other words, by using the absolute position information of the RFID tag with the odometry, the error of the position estimation of the moving object is accumulated.

Moving object position estimation system applying the moving object state group according to the present invention to achieve the above object

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. In addition, in describing this invention, the same code | symbol is attached | subjected and the repeated description is abbreviate | omitted.

A moving object position estimation system applying the moving object state group according to the present invention will be described with reference to FIGS. 1 to 3.

1 is a block diagram illustrating a moving object position estimation system to which the moving object state group according to the present invention is applied.

As shown in FIG. 1, the position estimation system according to the present invention includes a moving object 100 and a plurality of RFID tags 200 in which unique IDs are stored.

The moving object 100 shown in FIG. 1 includes an encoder for moving the moving object 100 and measuring an amount of rotation, thereby acquiring odometry information, and a driving detecting unit 110 and a driving detecting unit 110. RFID reader for controlling the speed and the direction of the drive driving unit 120 to drive the drive detection unit 110, each of the plurality of RFID tags 200 to receive the information stored in the RFID tag (200) And a computer unit 140 for controlling the entire moving object 100 such as the driving unit 130, the driving detection unit, the driving control unit, and the RFID reader unit.

The drive control unit 120 operates the drive detection unit 110 in response to the speed and the direction transmitted from the computer unit 140, the computer unit 140 is misleading from the drive detection unit 110 at 0.1 second intervals. Receive metadata information.

The RFID tag 200 is placed on the floor on which the moving object 100 moves and is recognized by the RFID reader 130 while the moving object 100 moves on the RFID tag. In the RFID tag 200, a group ID indicating a tag type and an individual ID for distinguishing each tag are stored, and the computer unit 140 recognizes coordinate information and recognition of a tag in a map corresponding to the ID. Possible recognition range information is stored in a table. In this way, storing the recognition range data according to the tag type is effective when several types of tags are mixed. The RFID reader unit 130 checks whether the RFID tag 200 is in the recognition area at an interval of 0.02 seconds to detect the RFID tag 200.

The computer unit 140 periodically receives the odometry information and the tag information from the driving detecting unit 110 and the RFID reader 130 to estimate the position of the moving object 100. The location is estimated using a plurality of status points with status information. The state point is represented by two-dimensional coordinates (x, y) and a posture (θ) component representing a position.

2 is a diagram illustrating a coordinate system of an RFID tag according to the present invention.

As shown in FIG. 2, among the RFID tags 200 laid on the floor, the coordinates (X _A , Y _A ) of the RFID tag 200 recognized by the moving object 100 at this point in time are the current moving object 100. ) Position. Each RFID tag 200 stores a unique ID, and the computer unit 140 of the moving object 100 stores an RFID coordinate value corresponding to the unique number. The moving object 100 recognizes the RFID tag 200 through the RFID reader 130 to obtain an ID, and finds an RFID coordinate corresponding to the ID to estimate the position.

3 is a diagram illustrating a relationship between an odometry coordinate system and an RFID coordinate system in a position estimation system according to the present invention.

As shown in FIG. 3, when the moving object 100 finds the RFID tag 200 at the point (X _A , Y _A ) of the RFID coordinate system at time t and proceeds by the distance d at time t + 1, the origin point If the angle between the line segment formed by O _R and the center of the moving object 100 and the X axis of the odometry coordinate system is θ _AC , the position in the RFID coordinate system is (Y _A + d · sin θ _AC , X _A + d · cosθ _AC ), and the attitude is θ _{A , t + 1} plus θ _{A, t} and θ _{R, t + 1} .

Next, a moving object position estimation method to which the moving object state group according to an embodiment of the present invention is applied will be described with reference to FIGS. 4 to 9.

4 is a flowchart illustrating a method for estimating a moving object according to the present invention, FIG. 5A is a view illustrating an initial position estimation of a moving object according to the present invention, and FIG. 5B is a moving object after an initial position estimation according to the present invention. 6 is a diagram illustrating a moving shape of FIG. 6 is a view illustrating a position estimation of a moving object when a tag is recognized according to the present invention, and FIG. 7 is a diagram illustrating a position estimation of a moving object when a tag is not recognized according to the present invention. 8 is a view illustrating a state point selected according to a movement of a moving object according to the present invention, and FIG. 9 is a view illustrating selection of state points to be used for estimating a position of a moving object according to the present invention. to be.

The process of estimating the actual position of the moving object 100 by randomly selecting a plurality of state points, which are points having the state of the moving object 100 at a place where the moving object 100 is located, and analyzing the state points may be performed in one step. In step 4, step 2-1 or step 2-2 is performed according to whether the RFID tag 200 is recognized. Position estimation process according to the movement of the moving object 100 will be described as follows.

As shown in FIG. 4, first, even before the moving object 100 obtains information, state points are evenly disposed in all regions to be located on the map stored in the computer unit 140. When the moving object 100 is driven by rotating the wheel which is the drive detecting unit 110, the driving detection unit 110 acquires the odometry information along with the driving and transmits the information to the computer unit 140 at an interval of 0.1 second (ST). 4010). The odometry information is obtained from the rotation speed of the wheel 110 of the moving object 100. When the RFID reader unit 130 recognizes the first RFID tag 200 (ST 4020), the computer unit 140 stores a plurality of state points having state information of the moving object 100 in the computer unit 140. Initialize the map.

The first stage, which is the initialization stage, is as follows.

Step 1: reset

The computer unit 140 generates the map (ST 4030), and generates status points at positions (x, y) on the map corresponding to the position of the RFID tag 200 recognized by the driving detection unit 110. (ST 4031). At this time, the position of the state points is the position of the moving object 100 to be expected. That is, each of the state points is one of the expected candidate positions where the moving object 100 is estimated. The state points are also given an arbitrary direction θ (ST 4032) because when one RFID tag 200 is found, the direction is unknown.

As shown in FIG. 5, large circles arranged in a square shape are areas that can be recognized by the moving object 100 as the RFID tag 200, and red circles of which are currently recognized by the moving object 200. RFID tag 200. In addition, the small hollow circle indicates the position of the actual moving object 200, and the indigo blue dots forming a circle shape by the dots gathered at the center of the area first detected by the moving object 100 are state points. When the moving object 100 recognizes the first RFID tag 200, the state point clusters are concentrated around the position of the RFID tag 200. At this time, the position of the moving object 100 can be known but not the posture.

That is, the state points are densely and evenly disposed only in the region within the recognition range from the center of the recognized RFID tag 200, and each state point has an arbitrary posture in the range of 0 ^° to 360 ^° . This is because when the reader unit 130 detects the RFID tag 200 due to the characteristics of the RFID sensor, an area in which the moving object 100 may be located is certainly limited, but the direction is unknown. The data of each status point is defined as follows.

P _i : position (x, y, θ)

P _i is the coordinate (x, y) where the i-th state point is located and the attitude (θ) component of the moving object in the area represented by the X, Y coordinates of the floor space where the moving object is active to estimate the position of the moving object 100. Indicates. At this time, one state point may have the same position as the other state point.

After the first step, the moving object 100 is driven again by rotating the wheel 100, obtaining the odometry information, and transmitting the information to the computer unit 140 at an interval of 0.1 second (ST 4050). The moving object 100 moves while checking whether the RFID tag 200 is recognized (4060). Therefore, when the moving object 100 moves, the state points on the map are also moved in consideration of the moving direction and distance. At this time, the direction in which the state points move is changed by the value obtained from the odometry information based on the direction arbitrarily selected in the first step. The method of estimating the new position of the moving object 100 according to the movement depends on whether the RFID tag 200 is recognized. As in step 1, when the moving object 100 discovers the first RFID tag 200 and proceeds, only the position information is known and the posture is not known. As shown in FIG. Spread out while keeping the circle shape as far as you have traveled.

If the moving object 100 moves after the step 1 and recognizes the RFID tag 200, step 2-1 is performed.

Step 2-1: Tag Recognition

When the moving object 100 recognizes a certain area by recognizing the RFID tag 200, 2 of the recognition range of the area, that is, the distance from the center to the edge point farthest from the center of the recognized area. Status points existing within the ship are left, and state points located more than twice are removed from the map (ST 4062).

The circular state points shown in FIG. 5B are selected only by the state points valid for the position estimation as shown in FIG. 6 by the step (ST 4062). In FIG. 6, the direction from the center of the small circle to the end of the protrusion represents the attitude of the moving object 100.

If the moving object 100 does not recognize the RFID tag 200 after moving to step 1, the following step 2-2 is performed.

Step 2-2: Progress of the moving object

If the moving object 100 does not recognize the tag, the computer unit 140 selects the N state points in the map as valid state points, and obtains the N state points from the driving detection unit 110. The distance, direction, and random noise obtained from the obtained odometry information are taken into consideration and moved to a new position determined by Equation 1 (ST 4061).

Pi, t: position and attitude of the i-th status point at time t

Δl: Moving distance of the moving object 100

V _t : Variable proportional to the degree of change in the moving distance and the attitude of the moving object 100 at time t

rand (): a function that returns any one decimal place from 0 to 2

Here, the random noise is an error generated when reading the odometry information and a measurement error caused by the surrounding environment. That is, V _t and rand () are variables of the expression for expressing the characteristics of the noise, and the variable V _t represents the distance and the attitude as one, and each time the function rand () is called, 0.0, 0.1, 0.2,. , 0.9,… , 1.0, 1.1,… Returns a number between 0 and 2, such as 1.9, 2.0, and so on.

As shown in (a) of FIG. 7, if the state point located at Pi, t at time t moves exactly by the distance and direction obtained from the odometry, the position of this state point at time t + 1 is P'i, will be t ₊₁ . However, the odometry contains noise. Therefore, it is assumed in the present invention that random noise occurs in a range proportional to the distance. Therefore, the position of the state point at time t + 1 considering this becomes Pi, t _{+ 1} . Such state points move as shown in FIG.

When the moving object 100 moves along the path shown in FIG. 8A, the shape of the state points changes from the point A of FIG. 8A as shown in FIG. 8B. That is, the moving object 100 does not recognize the exact position at point A, and the state points are widely distributed. Then, when the RFID tag 200 is found, the state points in front of the moving object close to the RFID tag 200 remain and the state points behind the moving object 100 far from the RFID tag 200 are removed little by little. On the contrary, only the front status points are removed when leaving the RFID tag 200, and the status points are generated in a circular shape around the position of the mobile object 100 immediately before the moving object 100 recognizes the RFID tag 200. . As a result, immediately after leaving the RFID tag 200, the moving object 100 estimates the position more accurately than before the RFID tag 200 is discovered. As described above, the moving object 100 continuously repeats the process of recognizing and not recognizing the RFID tag 200 while moving.

After the step 2-1 or step 2-2, the position (x, y) and the posture (θ) of the moving object 100 are estimated by using the position estimation method of the selected valid state points in three steps.

Step 3: estimate location

The computer unit 140 generates a histogram (number graph of state points) of the x-axis, the y-axis, and the θ-axis by projecting valid state points of the map on the x-axis, y-axis, and θ-axis. (ST 4070). In this way, the position of the largest cluster of the valid state points in the map is found (ST 4071). Next, if x, y, θ of each axis having the maximum value of the histogram is P _max (x _max, y _{max ,} θ _max ), x _max and The average position of the state points whose distance between the point formed by y _max is L (500mm) and θ _max is K (15˚-20˚) or less is the position of the moving object 100, and the average posture of the state points Determine the posture of the moving object (100).

In this embodiment, L can be arbitrarily designated, but the reason for specifying 500 mm is as follows. It is determined that the moving object 100 may be located near a point formed by x _max and y _max . Therefore, it is determined that the position information of the state points located too far from the point formed by x _max and y _max has a great difference from the position information of the actual moving object 100, Exclude. Therefore, only state points within a certain distance (inside the circle centered on x _max and y _max ) are selected, and the distance (radius of the circle) to be selected is L.

In addition, although K can be arbitrarily designated, the reason for designating it as 15 degrees-20 degrees is as follows. In the present embodiment, it is determined that the direction of θmax is likely to be the direction of the moving object 100, and only the points within the range of 15 ° are selected except the state points having the direction component that is too far from the direction, and the direction of the robot is determined. Estimate

As shown in FIG. 9, the point located at the center of the circle is the point having the maximum value in the histogram. This point (x _max , y _max ) center points within a certain distance (circle) are selected (ST 4072). Among the selected state points, only state points having a posture between θ _max -K to θ _max + K are selected again (ST 4073). Therefore, in FIG. 9, the state points in the B, C, and D directions are selected as valid state points, and A, E, F, and G are discarded.

Finally, the position of the moving object 100 is estimated by Equation 2 from the selected valid state points.

P _{moving object} : estimated position and attitude of the moving object 100

P _i : position and attitude of the i th state point among the valid state points

N: total number of valid state points

When the position of the moving object 100 is estimated in the third step, the following four steps are performed.

Step 4: Status point  Add

(N_dim: 제거된 상태점의 수) 이하의 원안에 임의의 위치에 추정된 이동물체(100)의 자세와 같은 자세를 갖는 상태점을 N_dim개 추가 생성한다.The three-step position estimation process reduces the number of status points and compensates for the reduced number (ST 4080). Distance based on the last estimated position of the moving object 100

(N _dim : number of removed state points) An additional N _dim state points having the same posture as the posture of the moving object 100 estimated at an arbitrary position are generated in the following circle.

As the moving object 100 continues to move, steps 2 to 4 are repeated to estimate the position and attitude of the moving object 100.

Next, a moving object position estimation method to which a moving object state group is applied according to a more specific embodiment according to the present invention will be described.

In this embodiment, it is assumed that the RFID tag 200 is placed on a flat floor and the robot, which is the moving object 100, recognizes a position while moving on the robot tag 200.

First, the moving object position estimation system according to the present invention will be described.

In the location estimation system according to the present invention, the RFID tag 200 is laid on the floor in a grid of square grids. At this time, each RFID tag 200 is given a unique ID.

The robot 100 has two wheels, each of which can be driven separately, as the driving detection unit 110, and the central axes of the two wheels 110 are in a straight line. And the RFID reader 130 is located on the central axis of the two wheels (110). In this case, when the direction of the robot 100 is to be changed, the directions of the two wheels 110 may be changed by changing the speed (rotation per hour) of the two wheels 110.

In addition, the wheels 110 of the robot 100 have encoders, respectively, so that the amount of rotation can be measured. Two wheels 110 have a motor (Motor), each of the motor is controlled by the rotation control speed 120 of the robot 100.

In addition, the robot 100 has a computer unit 140, and the position information of the RFID tag 200 found, the speed information of the robot wheel 110 from the encoder, the attitude and position information of the existing robot 100 Based on the position of the robot 100 is recognized.

In addition, the initial state and background of the position estimation system according to the present invention are as follows.

The robot 100 knows the information about the position coordinates and the recognition range for the RFID ID. Therefore, when the RFID tag 200 is found, the ID of the found tag 200 can be read to know the center position of the tag 200 and the information that the robot 100 is located within the recognition range. However, the first start does not know the position and direction of the robot 100 itself.

The position estimation process of the robot 100 itself is as follows.

1. Until the robot 100 finds the first RFID tag 200, the robot 100 operates only to find the tag 200 (ST 4010).

2. When the robot 100 finds the first RFID tag 200, it performs the initialization process of step 1. That is, a map is created on the computer unit 140 (ST 4030), and state points are generated at the location on the map corresponding to the position of the recognized RFID tag 200 (ST 4031). At this time, the position of the points becomes the position of the robot 100 to be expected. That is, each point is one of the expected candidate positions where the robot 100 is estimated. These points also include direction information of the robot 100, initially giving an arbitrary direction for each state point (ST 4032). This is because the direction is not known when only one RFID tag 200 is found.

3. When the robot proceeds in this state (ST 4050), the state points are moved in consideration of the direction and distance to proceed. At this time, the direction in which the state points move is changed by the value obtained by the encoder based on the direction arbitrarily selected in step 1. The state points also move in proportion to the distance traveled. At this time, the state point moves to a value obtained by the encoder in consideration of some arbitrary component, and the method is the same as step 2-2 above (ST 4061).

Since the direction of the robot 100 is arbitrarily determined in step 1, each state point is scattered while forming a donut shape. This is because the direction of the robot 100 is not known yet.

4. When the second RFID tag 200 is found, only the state points at a distance of up to twice the recognition range from the recognized tag are left, and the remaining state points are removed (ST 4062). The remaining state points are those that express the position and direction of the robot 100 now because the first randomly selected directions are similar to those of the actual robot 100.

The positions of the robots 100 are estimated from the selected state points by the three-step position estimation method (ST 4070 to ST 4074), and compensated by the number of reduced state points (ST 4080).

5. After finding the second tag, if the robot 100 proceeds again, it proceeds as in step 2-2 again, and estimates the position of the robot 100 every hour by the position estimation method of step 3 again at intervals of 0.1 second. (ST 4070 to ST 4074), compensates for the reduced number of status points (ST 4080).

6. After the other RFID tag 200 is found, the above 4 and 5 are repeated.

By repeating the above process, the method for estimating the position of the moving object according to the present invention is completed.

As mentioned above, although the invention made by this inventor was demonstrated concretely according to the said Example, this invention is not limited to the said Example and can be variously changed in the range which does not deviate from the summary.

As described above, according to the moving object position estimation system and method applying the moving object state group according to the present invention, the effect that the accuracy of the position and attitude estimation of the moving object can be improved.

Further, according to the moving object position estimation system and method applying the moving object state group according to the present invention, the position estimation accuracy is higher than other conventional methods, such as an RFID tag required per unit width in order to estimate the position of the moving object. The number of recognition devices can be reduced, so that the effect of cost is economical.

In addition, according to the moving object position estimation system and method applying the moving object state group according to the present invention, it is also possible to obtain the information from the RFID and odometry to compensate for the disadvantages of both technologies and to take advantage.

Claims (11)

In a location estimation system for recognizing the movement of a moving object by recognizing the location information from a plurality of RFID tags (Radio Frequency Identification Tag) stored unique ID (ID) and location information, The moving object A driving detecting unit which moves the moving object and obtains Odometry information; A drive controller for driving the drive detector by controlling the speed and direction of the drive detector; An RFID reader unit communicating with each of the plurality of RFID tags to receive information stored in the tag; It includes a computer unit for controlling the drive detection unit, the drive control unit and the RFID reader, The computer unit estimates the position of the moving object using a plurality of state points having the state information of the moving object selected by the odometry information periodically received and the tag information received from the RFID reader unit. Moving object position estimation system applying a moving object state group. The method of claim 1, And the computer unit stores the coordinate information and the recognizable recognition range information corresponding to the ID of the RFID tag in the form of a table. The method of claim 1, The state point is a moving object position estimation system applying a moving object state group, characterized in that it comprises a two-dimensional coordinates and attitude components. A drive detection unit for moving the moving object and obtaining odometry information, a drive control unit for driving the drive detection unit by controlling the speed and direction of the drive detection unit, and communicating with a plurality of RFID tags in which unique IDs are stored. An RFID reader unit for receiving the information, the computer and the computer unit for controlling the entire moving object to store the coordinate information and the recognizable recognition range information of the map corresponding to the ID of the RFID tag in the form of a table and the position information of the moving object In the location estimation method to recognize from the RFID tag, (a) periodically transmitting the odometry information to the computer unit while driving the drive detection unit; (b) the RFID reader unit recognizing the first RFID tag; (c) the computer unit initializing a plurality of state points having state information of the moving object in the map; (d) checking, by the computer unit, whether the RFID tag is recognized according to the driving of the driving detection unit, and selecting status points valid for position estimation from the map; (e) the computer unit estimating the position (x, y) and attitude (θ) of the moving object from the valid state points, (f) adding a status point in the map, The method of estimating a moving object state group, wherein the position and attitude of the moving object are estimated while repeating the steps (d) to (f) until the driving of the moving object stops. The method of claim 4, wherein Step (c) is Generating, by the computer unit, the map; Generating the status points at a location on the map corresponding to a location of the RFID tag; Moving object position estimation method using a moving object state group comprising the step of giving a direction (θ) to the state points. The method of claim 4, wherein If the tag is not recognized in the step (d) The computer unit selects the state points of the map as valid state points, and moves the valid state points in consideration of distance, direction, and noise obtained from the odometry information obtained from the driving detection unit. Moving object location estimation method using object state group. The method of claim 6, The position and posture to which the valid state points are to be moved are

Is computed by the execution of P _{i, t} is the position and attitude of the i-th state point at time t, Δl is the moving distance of the moving object, V _t is the degree of change of the moving distance and attitude of the moving object at time t Rand () is a function that takes one of the numbers from 0 to 2 and returns the first decimal place, and V _t and rand () are variables for expressing the characteristics of the noise. Moving object position estimation method applying the moving object state group characterized in that. The method of claim 4, wherein If the tag is recognized in the step (d) The moving object position estimation method according to claim 1, wherein the computer unit selects, as a valid state point, state points existing within two times the recognition range of the area from the center of the recognized area on the map. The method of claim 4, wherein Step (e) is (e1) generating valid histograms of the map on the x-axis, y-axis, and θ-axis to generate a histogram of the x-axis, the y-axis, and the θ-axis, (e2) finding a location of the largest cluster of the valid state points in the map, (e3) selecting state points within a predetermined distance (L) from the center point in the map using a point formed between the maximum value of the x-axis maximum value (x _max ) and the y-axis maximum value (y _max ) of the histogram as the center of the circle; step, (e4) Among the state points selected in step (e3), between the maximum value of the histogram (θ _max ) -the predetermined angle (K) to the maximum value of the (θ-axis) (θ _max ) + the predetermined angle (K) Selecting status points having a posture as valid state points, (e5) a moving object position to which the moving object state group is applied, comprising estimating a position (x, y) and a posture (θ) of the moving object from the valid state points selected in the step (e4). Estimation method. The method of claim 9, Step (e5) is the formula

Is computed by the execution of _Wherein the P _{moving object} is an estimated position and attitude of the moving object, P _i is a position and attitude of an i-th state point among the valid state points, and N is a total number of valid state points. Moving object position estimation method applying the moving object state group. The method of claim 4, wherein Step (f) is a distance based on the estimated position of the moving object

In the following rounds to generate a state that it has the same position and posture of said moving object by adding N _dim more, and the N _dim is characterized in that the number of the state point removed in the step (d) and (e) Moving object position estimation method applying the moving object state group.

Images