KR101341204B1 - Device and method for estimating location of mobile robot using raiser scanner and structure - Google Patents

Abstract

The present invention relates to an apparatus and method for estimating a mobile robot using a laser scanner and a structure. More specifically, the error of the position estimation can be minimized by using a laser scanner having a fixed ceiling structure and a small amount of calculation and a low error. The present invention provides a device and method for estimating a location of a mobile robot using a laser scanner and a structure.
The technical configuration is a method for estimating the position of a mobile robot using a laser scanner and a structure, the first step of scanning the upper surface or the front and the upper surface of the position where the mobile robot travels, patterned and stored as a pattern map ; A second step of scanning the mobile robot 180 degrees based on coordinates orthogonal to the driving direction and extracting the structures located on the upper surface or the front and upper surfaces; A third step of calculating a position of the mobile robot by calculating an inverse of a transformation matrix transformed from a robot coordinate system to a global coordinate system to matched structure coordinates when comparing the extracted structure with the pattern map; And a control unit.

Description

Device and method for location estimation of mobile robot using laser scanner and structure {DEVICE AND METHOD FOR ESTIMATING LOCATION OF MOBILE ROBOT USING RAISER SCANNER AND STRUCTURE}

Embodiments of the present invention relate to an apparatus and method for estimating a position of a mobile robot using a laser scanner and a structure, and more particularly, an error in position estimation using a laser scanner having a fixed ceiling structure and a small amount of calculation and a low error. The present invention relates to a mobile robot positioning apparatus and method using a laser scanner and a structure capable of minimizing the number.

In general, a robot is a machine that automatically handles or operates a given task by its own abilities, and is largely classified into a mobile robot and a fixed robot, and is a space robot, a humanoid robot, an industrial robot, a living organism, and a robot. It is classified into medical robot and network robot.

Industrial robots perform tasks such as welding work or assembly of parts that are performed at a product production site, and typically have a structure of a robot arm. The structure of the robot arm is characterized by having a plurality of joints, the robot arm is fixedly installed in one place, to perform the indicated work. Thus, the working space of the robotic arm can be extremely limited.

Mobile robots, on the other hand, have the ability to operate autonomously as robotic systems with artificial intelligence in their physical machinery, making them perform better than humans perform tasks that traditional mechanical equipment cannot perform in many areas. Can be done.

Vehicles applied to these robotic systems are controlled with their own power for driving and control equipment and sensors to control their operation. This control can be largely divided into a supervisory control robot that observes the movement of the robot and a remotely operated, which is controlled by a human by wire or wirelessly.

As robot control technology advances, supervisory control is becoming more popular than remote operation. To enable this, improved sensor and navigation technology must be combined.

The development of sensors and sensor networks has the following objectives: sensors need to monitor the environment and be interactive, and know their location using relative location.

There are many ways of recognizing the position and direction of a mobile robot, but the method using Odometry is widely used. Odometry is also known as dead-reckoning, and an odometry-based mobile robot obtains velocity information by using an odometer or a wheel sensor on a moving robot, and uses a magnetic sensor. Azimuth information is calculated to recognize the position and direction of the mobile robot itself.

Such an odometry uses only information generated by itself, without inputting additional information from the outside. Odometry acquires the location information at a very high sampling rate, so it is not only fast to update the location information, but also has an advantage that it can be easily implemented because the accuracy is very high and the cost is low at a relatively short distance.

However, since odometry calculates the position and direction through integration, measurement errors accumulate as the driving distance increases. In particular, when the moving distance is determined by the number of wheel rotations of the mobile robot, the mobile robot may not be able to determine the state of the road surface, but may also slip according to the road surface state of the work area, and the error caused by this may be completely compensated. There is a problem that can not accumulate when it is not moved far away.

On the other hand, there is a method that uses a landmark (Landmark) to recognize the position and direction of the mobile robot, which is a principle for calculating the position of the mobile robot from the landmark image obtained from the camera, the camera on the ceiling or wall of the room In contrast to the fixed installation form, there is a form of fixing the landmark and installing the camera on the mobile robot.

However, this method is limited to the movement range of the robot indoor space, sensitive to the influence of the light in the image acquisition method, and in order to improve the difficulty in use in outdoor environment, Korean Patent Publication No. 10-2009 In -0034007, a technique relating to robot position recognition using a laser is disclosed.

Referring to FIG. 1, the robot position recognition laser sensor system includes four sensor modules 1 and one processor module. The light emitting unit shoots light on the bottom surface, and the image sensor 11 views the bottom surface. Shoot in real time. In this case, when there is motion, the bottom image captured by the image sensor 11 is different from the previous image, and thus the motion of the X and Y axes is calculated by performing motion estimation by performing image processing.

However, when shooting forward or shooting the floor, if there are obstacles, moving objects, or the position of the obstacle constantly changing in front or bottom surface, the feature points cannot be extracted due to the obstacle or constantly changing objects and obstacles, When shooting with devices such as image sensors or cameras, the amount of computation required to extract features at exponential resolutions increases exponentially, maximizing cost and load by requiring high specifications to locate the mobile robot as it moves. There was a problem such as being.

Korean Patent Publication No. 10-2009-0034007

An embodiment of the present invention was devised to solve the above-mentioned problems, and by scanning the ceiling or wall of a mobile robot with a laser scanner, even if there are obstacles or moving objects on the front or the bottom, a feature point can be easily extracted. An object of the present invention is to provide an apparatus and method for positioning a mobile robot using a laser scanner and a structure.

An embodiment of the present invention, by using a laser scanner to shoot the floor surface and the wall surface and the upper surface, by taking a laser scanner and structure that can be reduced exponentially by calculating the required amount of feature extraction from 180 degrees to 180 degrees An object of the present invention is to provide a location estimation apparatus and method for a mobile robot.

In the embodiment of the present invention, since the mobile robot does not require high specifications to determine its position as it moves, it is possible to extract its location within a short time with a low load with a low specification and move to a desired location in less time. An object of the present invention is to provide a mobile robot positioning apparatus and method using a laser scanner and a structure.

In order to achieve the above object, the present invention provides a device for estimating the position of a mobile robot using a laser scanner and a structure, among the embodiments, the upper surface or the front and the upper surface of the position at which the mobile robot travels. A scanning unit to scan; A pattern map unit which patterns the structure and stores the pattern map in advance; A line fitting unit fitting the data scanned by the scan unit in a straight line when the mobile robot scans the upper surface or the front and upper surfaces while driving; A structure extracting unit extracting a structure when the distance or angle with the fitted line exceeds a first threshold value; A structure matching unit which matches the extracted structure with the pattern map and extracts a structure whose matching rate is equal to or less than a second threshold value; A position calculator for calculating a position of the mobile robot using coordinates and transformation matrices of the structure extracted by the structure matching unit; It is made to solve the problem by using a position estimation device of the mobile robot using a laser scanner and a structure comprising a.

In one embodiment, the scan unit includes a position estimation device of the mobile robot, characterized in that scanning from 0 degrees to 180 degrees of the coordinates orthogonal to the movement direction or the same as the movement direction.

In one embodiment, the line fitting unit includes a position estimation device of the mobile robot, characterized in that using any one of the Least Squre fitting method, Ransac algorithm, Haugh Transform when fitting the scanned data.

In one embodiment, the pattern map unit includes a position estimation device of the mobile robot, characterized in that for storing the patterned structure located on the upper surface symmetrical to the bottom surface on which the mobile robot is located.

In one embodiment, the structure extractor, when fitting in a straight line in the line fitting, when the distance is greater than the first threshold relative to the fitted line, or the angle exceeds the third threshold, the upper portion Positioning device of the mobile robot, characterized in that the extraction to the surface or the front and upper surface features.

In one embodiment, the structure includes a positioning device of the mobile robot, characterized in that any one of a ceiling structure, the upper surface and the front edge, concave or convex structure, reflective structure.

In one embodiment, when the structure is a reflective structure, the position of the mobile robot using any one of the distance between the reflective structure, the angle of the line segment connecting the reflective structure, the number of the reflective structure extracted at the same time And an estimating device.

In one embodiment, the structure extractor, when fitting in a straight line in the line fitting, when the distance is greater than the first threshold relative to the fitted line, or the angle exceeds the third threshold, the upper portion Positioning device of the mobile robot, characterized in that the extraction to the surface or the front and upper surface features.

Among the embodiments, in the method for estimating the position of the mobile robot using a laser scanner and the structure, the first surface or the front and the upper surface of the position at which the mobile robot travels is scanned and patterned and stored as a pattern map step; A second step of scanning the mobile robot 180 degrees based on coordinates orthogonal to the driving direction and extracting the structures located on the upper surface or the front and upper surfaces; A third step of calculating a position of the mobile robot by calculating an inverse of a transformation matrix transformed from a robot coordinate system to a global coordinate system to matched structure coordinates when comparing the extracted structure with the pattern map; It is made to solve the problem by using a position estimation method of the mobile robot using a laser scanner and a structure comprising a.

In one embodiment, after performing the scan, performing a segmentation for separating the upper surface and the front by using the angle and the distance of the scanned data, characterized in that the mobile robot position estimation method It includes.

In one embodiment, the third step includes distributing at least one equal weight particle in the pattern map; When the structure is extracted by scanning from the mobile robot, calculating a distance to the structure at the position of the at least one particle and giving a weight proportionally as the distance is minimum; Deleting particles whose weight is less than or equal to a predetermined weight, and replicating and deleting the deleted particles near particles that are greater than or equal to a predetermined weight; And when the particle converges within a certain error range, estimating the position of the particle having the lowest error range among the particles within the predetermined error range as the position of the mobile robot. Include estimation methods.

In one embodiment, the step of estimating the position of the robot includes a method for estimating the position of the mobile robot, characterized in that it uses at least one of the maximum weighted particle (particle), the weighted mean (Weighted Mean). .

As described above, the disclosed technology of the present invention having the above-described configuration, by scanning the ceiling or wall of the mobile robot with a laser scanner, even if there are obstacles or moving objects on the front or bottom surface, it is easy to extract feature points By using the laser scanner to shoot the bottom and wall and the top surface, from 0 to 180 degrees, the amount of computation required for feature extraction can be reduced exponentially. Since high specification is not required to determine the location, it is possible to extract its own location in a short time with low load with low specification and move it to the desired location in less time, so that the error rate is low and the reliability is high but the unit price is low. To produce a high quality, preemptive price and customer satisfaction Can achieve such effects.

1 is a schematic diagram of a robot position recognition laser sensor system according to the prior art.
2 is a block diagram illustrating an apparatus for estimating a position of a mobile robot using a laser scanner and a structure of the present invention.
3 is a flowchart illustrating a method for estimating a position of a mobile robot using a laser scanner and a structure of the present invention.
4 is a diagram illustrating the difference between the apparatus and method of the present invention and the apparatus and method of the prior art.
5 is a view illustrating feature extraction using a structure in the position estimation device of the mobile robot using the laser scanner and the structure of the present invention.
6 is a view for explaining another embodiment of the feature extraction method of FIG.
7 is a view for explaining different embodiments of matching the pattern map and the extracted pattern in the position estimation device of the mobile robot using the laser scanner and the structure of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, a detailed description of known techniques well known to those skilled in the art may be omitted.

In describing the constituent elements of the present invention, the same reference numerals may be given to constituent elements having the same name, and the same reference numerals may be given thereto even though they are different from each other. However, even in such a case, it does not mean that the corresponding component has different functions according to the embodiment, or does not mean that the different components have the same function. It should be judged based on the description of each component in the example.

In the following description of the embodiments of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

In addition, in describing the component of this invention, terms, such as 1st, 2nd, A, B, (a), (b), can be used. These terms are intended to distinguish the constituent elements from other constituent elements, and the terms do not limit the nature, order or order of the constituent elements. When a component is described as being "connected", "coupled", or "connected" to another component, the component may be directly connected or connected to the other component, Quot; may be "connected," "coupled," or "connected. &Quot;

It should be understood that the singular " include "or" have "are to be construed as including a stated feature, number, step, operation, component, It is to be understood that the combination is intended to specify that it does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

In each step, the identification code (e.g., a, b, c, etc.) is used for convenience of explanation, the identification code does not describe the order of each step, It may occur differently from the order specified. That is, each step may occur in the same order as described, may be performed substantially concurrently, or may be performed in reverse order.

Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

2 is a block diagram illustrating an apparatus for estimating a position of a mobile robot using a laser scanner and a structure of the present invention. Referring to FIG. 2, the position estimation device 1 of the mobile robot using the laser scanner and the structure of the present invention includes a driving unit 10, a scanning unit 20, a control unit 30, a pattern map storage unit 40, The line fitting unit 50, the structure extracting unit 60, the structure matching unit 70, and the position calculating unit 80 are included.

The driving unit 10 is provided to move the mobile robot, and may be one means for moving the mobile robot, and may be moved by a rotational drive such as a motor.

The scan unit 20 may be provided to identify a structure provided or installed indoors or outdoors by using a laser scanner (Raiser Sanner), to determine the position of the mobile robot, or to move in a predetermined movement direction or artificial intelligence. have. Also, the scan unit 20 may scan from 0 degrees to 180 degrees of coordinates orthogonal to the movement direction or the same as the movement direction.

The control unit 30 controls each component, and is connected to each component and a network to detect movement or data of each component and transmit a command.

The pattern map storage 40 stores the upper surface of the place where the mobile robot moves, or the upper surface and the front structures in a patterned form. For example, the pattern map unit 40 may pattern and store the structure located on an upper surface symmetrical with the bottom surface on which the mobile robot is located.

The line fitting unit 50 scans the upper surface of the place where the mobile robot moves through the scanning unit 20 or the upper surface and the front structures while the mobile robot moves, and then fits the data obtained by the scanning unit 20 in a straight line. Then, the feature is extracted by the point of deviation from the fitted line or the point where the angle change is sharp. Here, when the scan unit 20 fits the scanned data, any one of a Least Squre fitting method, a Ransac algorithm, and a Haugh Transform may be used.

When the structure extracting unit 60 extracts a feature that points away from a line fitted in the line fitting unit 50 or a point where an angle change is sharp, the structure extracting unit 60 recognizes and extracts such a point as a structure. For example, when the structure extractor fits in a straight line in the line fitting part 50, when the distance exceeds the first threshold value or the angle exceeds the third threshold value based on the fitted line, the upper surface Or it can be extracted by the features of the front and top surfaces. Here, the structure may be any one of a ceiling structure, an upper surface and a front edge, a concave or convex structure, and a reflective structure. When the structure is a reflective structure, any one of the distance between the reflective structures, the angle of the line segment connecting the reflective structure, and the number of the reflective structures extracted at the same time may be used.

The structure matching unit 70 compares the pattern of the structure extracted by the structure extracting unit 60 with the pattern map stored in the pattern map storage unit 40 to find a matching structure.

When the position calculation unit 80 finds a matching structure in the structure matching unit 70, the position calculation unit 80 calculates the position of the mobile robot located on the bottom surface symmetric to the position of the upper surface by using a transformation matrix.

Hereinafter, the driving method will be described in detail with the above configuration.

According to an embodiment of the present invention, first, the global coordinates and patterns of the upper surface, that is, the ceiling structure, are stored in the map form in the pattern map storage 40 of the mobile robot.

Then, while driving indoors, the ceiling is scanned using the scan unit 20 such as an upward laser scanner.

Here, referring to FIG. 4A, the conventional method using the forward scan has a high possibility of failing feature point extraction due to an obstacle or moving object in front such as (a) or (b). On the other hand, when scanning upwards, that is, the ceiling, the influence of obstacles or moving objects can be minimized. For this reason, in the embodiment of the present invention, it is possible to use the upper surface, the upper surface or the front surface, that is, the wall surface which is the tetrahedron of the ceiling or the lower portion thereof. For example, a relatively small or fixed object such as a light bulb on the upper surface or a clock or calendar on the wall may be used as a structure.

Referring to FIG. 4B, feature extraction methods using cameras, which have been studied a lot, are affected by environment such as ambient light and shadow, and have a disadvantage in that the amount of computation required to extract features is proportional to the resolution of an image. However, in the case of a laser scanner, if there is no instant obstacle, a consistent result is guaranteed, and the calculation amount required for feature extraction is significantly lower than that of the camera method. For example, in the case of image acquisition using a camera, the general resolutions of 320 * 240 and 640 * 480 require 76,800 and 307,200 calculation amounts, respectively, but in the case of a laser scanner, the angle is based on 180 degrees depending on the angle. With 180 degrees and 360 degrees of 0.5 degrees each, the amount of computation can be significantly reduced compared to the camera method.

That is, assuming that the scan unit 20 scans while moving forward, the right wall is taken as a laser dot based on the bottom surface of the right wall, that is, the zero degree point, based on the location where the mobile robot is located. If it is taken over the edge of the right wall surface to the top, that is, the position corresponding to its own position on the ceiling, it is 90 degrees. If you continue to scan from 90 degrees to the left, you can scan up to 180 degrees down the left wall. Such a scan can be deformed into any shape, such as a point or a line, and the angle can be further reduced according to this shape deformation. For the sake of accuracy, even if scanning is performed with dots or lines at intervals of 0.5 degrees, it is 360 (360 * 0.5 degrees = 180 degrees), so it can be seen that the calculation amount decreases exponentially.

The line fitting unit 50 performs segmentation that separates data on both wall and ceiling surfaces using the angle value between the scanned data and the distance. The reason for this is to distinguish the ceiling from the wall within the mobile robot itself. In the embodiment of the present invention, the pattern map can be stored based on the ceiling. In order to remove the data, the data on the walls and ceilings is separated.

The line fitting unit 50 performs linear fitting on the separated wall and ceiling surfaces using the Least-Square fitting method, respectively, and the structure extracting unit 60 determines the error between the fitted line and the scan data. The ceiling structure defines and extracts a point whose reflectance is above or above a threshold or above or above a specific third threshold in the scan data.

That is, there may be two methods for extracting the feature of the ceiling using the scan unit 20, which may be a laser scanner. Referring to FIG. 5, the first method may be a method of extracting an edge existing between the ceiling and a wall or a fluorescent or incandescent lamp existing in the middle of the ceiling as a feature of the ceiling. This method is a method of extracting a feature characterized in that the data obtained by the laser is fitted in a straight line, the point of departure from the fitted line or the point where the angle change is sharp. The linear fitting algorithm of the line may be a method such as a Least Square fitting method, a Ransac algorithm, a Haugh Transform, and the like, and FIG. 6A illustrates the first method, the Least Square method.

That is, the straight line is a fitted line, the data located on the straight line is scan data, and the data not located on the straight line is scan data having high reflectance. The equation for fitting the straight line is represented by Equation 1 below.

In Equation 1, if the linear equation is y = ax + b, and the scan data has coordinates of Y and x values, the error distance of the i-th point is the error distance of the scan data is the i-th straight line from the i-th scan data. It is equal to subtracting the equation, so the total error S is equal to the sum of i plus 1 to n.

When minimizing to make the total error close to zero, it is possible to identify a point or angle that deviates from the fitted line so as to minimize the error rate, and the first threshold which is a preset distance threshold or the third which is an angular threshold. It is possible to find the points that exceed the threshold.

The second method, which will be described with reference to FIG. 6B, is a method of using an artificial mark such as a reflector or a mirror ball. In other words, the use of a reflector, etc., may be a method of using an object having a high reflectance with respect to the laser on the wall and the ceiling to use an angle or a distance between the artificial marks. At the same time, by combining features extracted from a short distance and recognizing them as a pattern, matching efficiency and location estimation performance with a pattern map can be improved. As the pattern information, as shown in FIG. 6B, a distance between features, an angle of a line segment connecting each feature, and the number of features simultaneously extracted may be used.

Here, the ceiling structure is a convex or concave object that can be distinguished from the ceiling surface or the wall surface when the scan unit 20 of the mobile robot scans, at least one structure may be installed in a pattern as a set, It may not be the same as other ceiling structure set patterns in the pattern map.

The structure extracting unit 60 may recognize and extract such a structure as a set, and the structure matching unit 70 matches the extracted pattern with the pattern of each structure set on the map, so that the error is less than or equal to the second threshold. When the matched structure is found, the position of the robot is calculated by multiplying the matched structure coordinates by the inverse of the transformation matrix transformed from the robot coordinate system to the global coordinate system.

In the embodiment of the present invention, there are two methods for matching the pattern map and the extracted pattern in the structure matching unit 70. First, when the size of the pattern map is not large, the position of the mobile robot is estimated by comparing the patterns extracted by the mobile robot with all the patterns of the stored map.

Second, when the size of the map is large, the number of patterns stored in the map is large, and the pattern is complicated, an optimal prediction technique such as a particle filter may be applied. The position estimation method using the particle filter is as follows.

Like 6-1, each particle has the location information of the mobile robot and is evenly scattered over the map with initialization. The initial particles all have the same weight, and the patterns on the pattern map all have different patterns.

Using the linear velocity and angular velocity information measured from the sensor of the mobile robot, all particles are moved equally as in 6-2. In other words, assuming that the mobile robot has moved southwest, for example, all particles are moved on the pattern map stored in the mobile robot by the length moved southwest. However, patterns drawn on the map are fixed values and do not move. In this case, noise data may be added and sampled by the angular velocity and linear velocity measurement sensor error.

When a pattern is observed in a mobile robot, the pattern is searched for at each particle position, and the weight of each particle is set as 6-3 based on the information.

Particles whose weight falls below a certain level are deleted, and particles that are as high as the weight are deleted as shown in 6-5. For example, suppose that 10 particles were evenly scattered on a pattern map of a mobile robot, and as the mobile robot moved, it found a pattern that matches the closest pattern that actually exists. Of course, the mobile robot moves until it finds the pattern with the lowest matching error rate, that is, the highest matching rate, so that the particles all move in the same direction and distance. Assuming that the position of the actual pattern having the lowest error rate, that is, the highest matching rate, is found, the position on the pattern map may be estimated as its own position. However, in large and wide maps, the margin of error can be large to estimate the position of the pattern as its own position.

Therefore, by weighting only the particles closest to the matched pattern and repeating the process, if the position of the particle closest to the matched pattern is estimated as its own position, the error is much reduced.

That is, the patterns are arranged every 5m, and when the particles are duplicated and concentrated, the particles may be arranged in cm units, the pattern closest to the particles is found, and the position of the particle closest to the pattern is determined by the position of the mobile robot itself. If we assume that the distance between the particle and the pattern is reduced in cm, the error unit can be reduced to 100/1.

Therefore, if the weight falls below a certain level, delete the particles, duplicate the weights as high as they are deleted, and repeat the process continuously to converge within a certain range, so that the maximum weighted particle or the weighted mean ) Can be used to estimate the position of the robot.

Although the process of FIG. 7 has described the process of finding the particle closest to the pattern as the robot moves, it is possible to estimate the nearest particle to its position by calculating the weight in a fixed state without moving, or rooting the particle while moving. The number may be in the form, the numbers 1 to 5 shown in the drawings may not occur in order, it may be possible to estimate the location only by one process, and various changes and modifications are possible.

3 is a flowchart illustrating a method for estimating a position of a mobile robot using a laser scanner and a structure of the present invention. As shown in FIG. 3, when the structure is stored in a pattern form in the mobile robot (S10), while the mobile robot travels (S20), when the upper surface such as the ceiling of the room is scanned in the scan unit (S30), the line The fitting unit performs data segmentation to distinguish the ceiling surface from the wall (S30).

If the ceiling fitting and the wall are divided by the line fitting part, the segment of the ceiling and the segment of the wall are divided, and line fitting is performed for each segment (S43), and if the line error is greater than or greater than the first threshold, which is a distance error, or It is checked whether the angle error is equal to or greater than the third threshold value (not shown) (S45).

The structure extracting unit recognizes a portion where a distance error or an angular error has occurred as a structure, extracts and patterns the structure (S50), and matches the pattern map stored in the pattern map storage unit with the pattern extracted by the structure extracting unit in the structure matching unit. (S61).

If the matching error is less than or equal to the second threshold value (S63), since the matching rate is high, the position calculating unit calculates the position of the robot, which is a relative position of the position of the corresponding structure, using the transformation matrix (S70).

The disclosed technology of the present invention scans the ceiling or the wall of a mobile robot with a laser scanner, so that even if there are obstacles or moving objects on the front or the bottom, feature points can be easily extracted, and the bottom surface can be extracted using a laser scanner. By shooting from 0 to 180 degrees to capture the wall and upper surface, the calculation required for feature extraction can be reduced exponentially, and high specifications are not required for mobile robots to locate themselves as they move. With its low load, it can extract its location in a short time and move it to the desired location in a short time, so it is possible to produce mobile robots with high reliability and low unit cost with low error rate. Customer satisfaction can be realized.

The present invention is not necessarily limited to these embodiments, as all the constituent elements constituting the embodiment of the present invention are described as being combined or operated in one operation. That is, within the scope of the present invention, all of the components may be selectively coupled to one or more of them. In addition, although all of the components may be implemented as one independent hardware, some or all of the components may be selectively combined to perform a part or all of the functions in one or a plurality of hardware. As shown in FIG. In addition, such a computer program may be stored in a computer-readable medium such as a USB memory, a CD disk, a flash memory, etc., and read and executed by a computer, thereby implementing embodiments of the present invention. As the storage medium of the computer program, a magnetic recording medium, an optical recording medium, a carrier wave medium, or the like may be included.

Furthermore, all terms including technical or scientific terms have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, unless otherwise defined in the Detailed Description. Terms used generally, such as terms defined in a dictionary, should be interpreted to coincide with the contextual meaning of the related art, and shall not be interpreted in an ideal or excessively formal sense unless explicitly defined in the present invention.

The foregoing description is merely illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. In addition, the embodiments disclosed in the present invention are not intended to limit the scope of the present invention but to limit the scope of the technical idea of the present invention. Accordingly, the scope of protection of the present invention should be construed according to the claims, and all technical ideas within the scope of equivalents should be interpreted as being included in the scope of the present invention.

1: Positioning device of mobile robot using laser scanner and structure
10: driving unit 20: scanning unit
30: control unit 40: pattern map storage unit
50: line fitting portion 60: structure extraction portion
70: structure matching unit 80: position calculation unit

Claims (11)

An apparatus for estimating the position of a mobile robot using a laser scanner and a structure,
A scan unit scanning an upper surface or a front and an upper surface of a position where the mobile robot travels;
A pattern map unit which patterns the structure and stores the pattern map in advance;
A line fitting unit fitting the data scanned by the scan unit in a straight line when the mobile robot scans the upper surface or the front and upper surfaces while driving;
A structure extracting unit extracting a structure when the distance or angle with the fitted line exceeds a first threshold value;
A structure matching unit which matches the extracted structure with the pattern map and extracts a structure having a matching ratio less than or equal to a second threshold value;
It includes a position calculation unit for calculating the position of the mobile robot using the coordinates and the transformation matrix of the structure extracted by the structure matching unit,
The pattern map unit is a laser scanner and the position estimation device of the mobile robot using the structure, characterized in that for storing the patterned structure located on the upper surface symmetrical to the bottom surface on which the mobile robot is located.
The method of claim 1,
The scanning unit is a laser scanner, the position estimation device of the mobile robot using a laser scanner and the structure, characterized in that scanning from 0 to 180 degrees of the coordinates orthogonal to the movement direction or the same as the movement direction.
delete The method of claim 1,
The line fitting unit is a laser scanner and the position estimation device of the mobile robot using a structure, characterized in that any one of the method of fitting the scanned data, the Least Square fitting method, Ransac algorithm, Haugh Transform.
The method of claim 1,
When the structure extractor fits in a straight line in the line fitting part, when the distance exceeds the first threshold value or the angle exceeds the third threshold value based on the fitted line, the upper surface or the front and the upper part Positioning device for a mobile robot using a laser scanner and the structure, characterized in that the extraction as a feature of the surface.
The method of claim 1,
Wherein the structure is a position of the mobile robot using a laser scanner and the structure, characterized in that any one of a ceiling structure, the upper surface and the front edge, concave or convex structure, reflective structure.
The method according to claim 6,
When the structure is a reflective structure, any one of a distance between the reflective structures, an angle of a line segment connecting the reflective structure, and the number of the reflective structures extracted simultaneously is used, and the reflectance of the reflective structure is set to a preset threshold value. When exceeding, the position extraction device of the mobile robot using a laser scanner and the structure, characterized in that for extracting the structure from the structure extraction unit.
In the method for estimating the position of the mobile robot using a laser scanner and the structure,
A first step of scanning and patterning the upper surface or the front and the upper surface of the position where the mobile robot travels and storing the pattern map;
A second step of scanning the mobile robot 180 degrees based on coordinates orthogonal to the driving direction and extracting the structures located on the upper surface or the front and upper surfaces;
Comparing the extracted structure and the pattern map to find a matched structure, a third step of calculating the position of the mobile robot by calculating the inverse of the transformation matrix transformed from the robot coordinate system to the global coordinate system to the matched structure coordinates Include,
The second step
After performing the scan, performing segmentation to separate the upper surface and the front by using the angle and the distance of the scanned data;
Extracting into the structure when an error between the line fitted by the straight fitting method using the least square fitting method and the scan data is equal to or greater than a first threshold value that is a distance error and a third threshold value that is an angle error;
If the structure is a reflective structure, if the reflectance of the scan data is more than a predetermined threshold value, extracting to the structure comprising a laser scanner and a method for estimating the position of the mobile robot using the structure.
delete The method of claim 8,
In the third step,
Distributing at least one particle having the same weight on the pattern map;
When the structure is extracted by scanning from the mobile robot, calculating a distance to the structure at the position of the at least one particle and giving a weight proportionally as the distance is minimum;
Deleting particles whose weight is less than or equal to a predetermined weight, and replicating and deleting the deleted particles near particles that are greater than or equal to a predetermined weight;
If the particle converges within a certain error range, estimating the position of the particle having the lowest error range among the particles within a certain error range as the position of the mobile robot,
And estimating the position of the robot using at least one of a maximum weighted particle and a weighted mean.
11. The method of claim 10,
Moving the at least one particle in a direction and a distance in which the mobile robot travels using the linear speed and the angular velocity of the mobile robot running when the mobile robot travels;
Positioning method of the mobile robot using a laser scanner and the structure, characterized in that it further comprises.

Images