KR20150087585A - pattern recognition system using 2D laser scanner shaking, there of method - Google Patents

Abstract

The present invention relates to a system for recognizing a vehicle front object pattern using up/down swing of a 2D laser scanner and a method thereof. The system of the present invention includes: at least one 2D laser scanner which is installed in the outside of a vehicle; a gear structure which changes up/down glancing angle of the 2D laser scanner; a step motor which is geared to the gear structure, and is mounted in the 2D laser scanner, and drives variation of the up/down glancing angle of the 2D laser scanner; and a central processing unit which controls the 2D laser scanner and the step motor by being connected to the 2D laser scanner, and forms pattern data of a front object by clustering the scanned data, and is installed in the vehicle. The present invention provides the more accurate system for recognizing the vehicle front object, and the method thereof. The system and the method of the present invention can improve the problem of front object pattern recognition according to high speed traveling, road condition and surrounding situation, and also have high reliability. Also, the system and the method can improve pattern recognition reliability, by extracting posture information of the vehicle itself and then using the posture information in data processing.

Description

[0001] The present invention relates to a 2D laser scanner, and more particularly,

BACKGROUND OF THE INVENTION Field of the Invention [0001] The present invention relates to a vehicle front object pattern recognizing system and method, and more particularly, to a vehicular front object pattern recognizing system and a method thereof using a vertical sweep angle sweep of a 2D laser scanner .

For example, people with visual impairments or persons with paraplegia suffer from considerable difficulty in moving to a long distance using a car. In addition, buses for transit weaknesses and transit announcement tones are introduced and commercialized for public transportation, but they are still inconvenient and difficult to use. In order to move to a different area than the same area, traffic disadvantages must go through many difficulties. However, if unmanned autonomous vehicles are developed and used by transportation weak, they will be able to move to other areas as well as ordinary people and live more conveniently.

In recent years, development and interest in the unmanned autonomous vehicles have been heated around the world. Unmanned autonomous vehicles have been novel in the past, but are now nearing commercialization. The self-driving car made by Google has been released and under test, spurring the development of unmanned autonomous vehicles in Toyota as well as in the United States. In Korea, the Ministry of Commerce, Industry and Energy and Hyundai Kia Motors are continuing to develop and compete in autonomous autonomous vehicles.

The obstacle recognition system in the unmanned autonomous vehicle is an essential element for securing the running of the vehicle by using various sensors instead of the human eyes. It recognizes various obstacles existing around the vehicle while driving and provides obstacle related information to avoid obstacles.

Also, in order for the autonomous autonomous vehicle to travel safely, a distance measuring sensor capable of obtaining a distance from other objects in the vicinity of the vehicle is required. This distance measuring sensor must be installed outside the vehicle because the vehicle travels in various environments, and it should be strong against impact, less influence on illumination, waterproof and fast. Considering these requirements, we used SICK's LMS511 pro model.

Three times in 2004, 2005 and 2007, an unmanned auto racing competition was held under the auspices of Defense Advanced Research Project Agency (DARPA), the US Department of Defense's highest research institute.

In the 2004 competition, there were no finished vehicles. In 2005, only 5 out of 23 unmanned cars were completed. In 2007, the event was held at the George Air Force Base in Victorville, California, which was closed as an Urban Challenge, and the course consisted of safe driving within 96 km of the city center within six hours of completing all traffic rules and obstacles and joining vehicles . As the Carnegie Mellon college team, the champion was composed of a 'distributed control system' to cope with the rapid data processing and system failure of the unmanned autonomous navigation system.

Google also succeeded in driving a total of 140,000 miles (224,000 kilometers) on seven road test vehicles on actual roads in September 2010, which were made by retrofitting the Toyota Prius. Currently, there are various road driving tests including crowded city, highway, and mountain road. In 2011, we unveiled the first video of the unmanned car, the Prius, carrying a blind person. Key technologies include sensors such as Lidar, video cameras, precision positioning devices, and radar, combined with Google's Street View technology.

GM unveiled the EN-V at CES 2011, the International Electronics Fair in Las Vegas on January 6, 2011, and unveiled a network electric vehicle equipped with a communications network to enable autonomous navigation. The concept of the EN-V is a future-oriented personalized vehicle that allows the vehicle to self-park and automatically travel to its destination. Major technologies include 15 sensors mounted on the outside of the body to recognize the surrounding situation, GPS, distance sensor, and VtoV Communications technology.

Volkswagen's unmanned vehicle, Shelley, demonstrated with the Stanford University based on the 2009 Audi TTS, unmanned rocky mountains in the United States. It is unique as an unmanned vehicle running on a special mountain road, and the size of the task carried out by industry-academy cooperation is very large. It is designed to carry drift at the off-road corner. Key technologies include inertial sensor technology and drive-by-wire technology for high-precision GPS and body control devices.

In 2007, BMW developed an unmanned driving system that remembers its way with state-of-the-art GPS systems once humans were driving the road. The BMW 5 Series has been prototyped with a test run of 5,000 km. Key technologies include high-precision GPS systems, left and right sensing radar sensors, front and rear sensing video cameras, and vehicle control technology.

In addition, Toyota, Audi and Mercedes-Benz have entered the unmanned vehicle research system, and are eager for a system that starts and stops by itself in response to traffic signals.

On November 11, 1992, Professor Han Minhong of Industrial Engineering Department of Korea University developed the first unmanned automobile in Korea. The KARV-1 unmanned car is equipped with two computers and two cameras, two ultrasonic sensors, one infrared sensor, and a motor in a military jeep. When obstacles appear, the speed is reduced, When it disappears, it starts again.

In November, 2010, an unmanned autonomous vehicle competition was held at Hyundai-Kia Motors' Nanyang Institute. There were 9 missions in total, including avoidance of stationary vehicles, compliance with speed limits, suspension in front of a crosswalk, and passage / tunnel passage. With the completed autonomous vehicle, students actually introduced the idea of making autonomous autonomous vehicles.

In September of last year, the second Hyundai · Kia Motors unmanned autonomous driving car contest was held with 9 missions, including variable lane lights, sudden obstacles / stops, and passing slow cars. First place was Hanyang University, second place was Chungbuk National University, and third place was Keimyung University.

In addition, the international unmanned solar car racing competition was held on both October 20th and 21st, 2012, and the competition was held at the driving test center of the Automobile Performance Research Institute of the Traffic Safety Corporation.

In 2013, the unmanned autonomous driving car contest hosted by the Ministry of Industry and Commerce was held in Yeongam F1 Stadium in October and the Hyundai · Kia Motors Autonomous Autonomous Vehicle Competition will be held in 2014. This year, unmanned autonomous vehicle competition is held every year, and interest and support for autonomous vehicles are spreading in Korea.

Unmanned autonomous vehicle is a vehicle with advanced technology and safety must be the first. It should be possible to cope with various unexpected situations such as a sudden obstacle appearing while the vehicle is running, or a sudden stopping of the preceding vehicle. It is the most important point of research of autonomous autonomous vehicle that it is possible to recognize the surrounding situation quickly and to cope with it safely by using sensors among autonomous autonomous vehicles.

To recognize the situation around you, you need a sensor that will act as a human eye. However, when using a vision sensor such as a camera, it is sensitive to light intensity and weather, and it is difficult to identify obstacles. In contrast, laser scanners use distances and angles to determine the location of obstacles, and there is little difference between day and night.

However, unlike vision sensors, laser scanner data is difficult to see and judge directly by the human eye.

Korean Registered Patent No. 10-0954621 (Registered on Apr. 16, 2010)

An object of the present invention to solve the above problems is to propose a method of recognizing an obstacle when traveling by using a vertical swing or shaking of a laser scanner. In addition, the obstacle recognition is performed by using the scan data of the laser scanner, and the obstacle recognition reliability of the unmanned autonomous vehicle is secured through rough determination of the obstacle.

A first aspect of the present invention for solving the above problems is a system for recognizing a vehicle front object pattern using up and down swings of a 2D laser scanner, comprising: at least one 2D laser scanner installed outside the vehicle; A gear structure for changing the vertical irradiation angle of the 2D racer scanner; A stepping motor which meshes with the gear structure and is mounted on the 2D laser scanner and drives a change in the vertical irradiation angle of the 2D laser scanner; And a central processing unit connected to the 2D laser scanner to control the 2D laser scanner and the stepping motor and to cluster the scanned data to form pattern data of a forward object.

Preferably, the upper and lower irradiation angles are varied between 0 and 5 degrees, and the 2D laser scanner is preferably installed on the front surface of the vehicle and the roof panel of the vehicle.

A second aspect of the present invention is a method for recognizing a vehicle front object pattern using up and down swings of a 2D laser scanner, the pattern recognition method using the vehicle front object pattern recognition system comprising the steps of: (a) A laser scanning step of changing a vertical scanning angle of the 2D laser scanner; (b) extracting scanned data from the central processing unit and clustering data; (c) determining a width of the pattern by using the clustered data in the central processing unit and calculating a distance to the forward object; (d) forming and recognizing a forward object pattern by arranging patterns through the width and distance data.

In the step (a), it is preferable that the irradiation angle is changed between 0 and 5 degrees, and the step (d) is performed by arranging the array data on the clustered data, step; And rearranging the sorted data patterns so that the data patterns are symmetrically divided by dividing the data patterns by two.

In addition, preferably, the step (a) may include extracting the upper and lower irradiation angle data when the 2D laser is scanned, and may further include a step of detecting the pitch of the vehicle using an acceleration sensor or a tilt sensor, And correcting the scan data according to the attitude of the vehicle by extracting roll, yaw, and yaw values.

As described above, the present invention provides a system and method for recognizing a forward object pattern of a vehicle more precisely, and is capable of improving the problem of forward object pattern recognition according to high-speed driving, road conditions and surrounding conditions, The present invention also provides a system and method for improving pattern recognition reliability by extracting attitude information of a vehicle itself and using the attitude information for data processing.

FIG. 1 is a view showing a configuration of a vehicle front object pattern recognition system using a vertical swing of a 2D laser scanner according to an embodiment of the present invention,
2 is a photograph of a vehicle equipped with a vehicle front object pattern recognition system using a vertical swing of a 2D laser scanner according to an embodiment of the present invention,
3 shows a scanner installation configuration of a vehicle front object pattern recognition system using a vertical swing of a 2D laser scanner according to an embodiment of the present invention,
4 is a flowchart illustrating a vehicle front object pattern recognition method using a vertical swing of a 2D laser scanner according to another embodiment of the present invention,
5 is a graph of LMS data input and output in a vehicle front object pattern recognition method using a vertical swing of a 2D laser scanner according to an embodiment of the present invention,
FIG. 6 is a graph showing clustering of scan data through a method of recognizing a vehicle front object pattern using a vertical swing of a 2D laser scanner according to another embodiment of the present invention,
FIG. 7 is a graph showing a data pattern clustered for 1 ms on a signboard as a forward object in a method of recognizing a vehicle front object pattern using a vertical swing of a 2D laser scanner according to an embodiment of the present invention,
8 is a schematic diagram showing a triangular scanning area of the laser scanner,
FIG. 9 is a pattern graph showing the result of arranging data by Lab VIEW 1D array alignment in FIG. 8,
10 is a photograph showing a result (FIG. 10A) of a vehicle front object pattern recognition method using a vertical swing of a 2D laser scanner according to an embodiment of the present invention and a traffic sign as a front object (FIG. 10 b))

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention, and how to accomplish it, will be described with reference to the embodiments described in detail below with reference to the accompanying drawings. However, the present invention is not limited to the embodiments described herein but may be embodied in other forms. The embodiments are provided so that those skilled in the art can easily carry out the technical idea of the present invention to those skilled in the art.

In the drawings, embodiments of the present invention are not limited to the specific forms shown and are exaggerated for clarity. Also, the same reference numerals denote the same components throughout the specification.

The expression "and / or" is used herein to mean including at least one of the elements listed before and after. Also, singular forms include plural forms unless the context clearly dictates otherwise. Also, components, steps, operations and elements referred to in the specification as " comprises "or" comprising " refer to the presence or addition of one or more other components, steps, operations, elements, and / or devices.

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

FIG. 1 is a view showing a configuration of a vehicle front object pattern recognition system using a vertical swing of a 2D laser scanner according to an embodiment of the present invention. 1, the vehicle 10 front object pattern recognition system according to the embodiment of the present invention includes a 2D laser scanner 100 installed on the roof panel of the vehicle 10; A gear structure (150) for changing the vertical irradiation angle of the 2D racer scanner; A stepping motor 200 that meshes with the gear structure 150 and is mounted on the 2D laser scanner 100 to drive a change in the upward and downward irradiation angles of the 2D laser scanner 100; And the 2D laser scanner 100 to control the 2D laser scanner 100 and the stepper motor 200 and to cluster the scanned data to form pattern data of a forward object And a central processing unit 300 installed therein.

As described above, the present invention proposes a method of recognizing an obstacle when driving by using a laser scanner 100 that shakes the upper and lower irradiation angles. A vehicle capable of recognizing an obstacle by using a laser scanner 100 installed on a front surface of the vehicle 10 and / or a roof panel and securing the reliability of obstacle recognition of the unmanned autonomous vehicle through rough determination of the obstacle 10) is proposed.

1, the laser scanner 100 used in the forward object pattern recognition system of the vehicle 10 according to the embodiment of the present invention uses the LMS511 Pro manufactured by SICK, Obstacles between 5 ° and 185 ° can be measured and have resolutions of 0.167 °, 0.25 °, 0.333 °, 0.5 °, 0.667 ° and 1 ° depending on the setting. A lower resolution allows for faster scanning. Here, the resolution is 0.167 °, and the distance is measured up to 80 m. You can also set the frequency. In this study, the resolution is 0.167 °, the frequency is 25Hz, and the scan time is 40ms.

Used as a 2D scanner, the LMS511 Pro is a distance measuring sensor. In the sensor distance measurement method, a TOF (Time of Flight) method is used to calculate the distance to a target object by measuring the time of reflection of a laser beam scanned in a specific direction.

In order to recognize the pattern of an object with the laser scanner 100, the shape of the object must be known. The 2D laser scanner 100 reads the 2D plane and therefore can not accurately determine the shape of the object. Therefore, the 2D laser scanner 100 used in the embodiment of the present invention proposes a system for recognizing the shape of an object by using a method of shaking up and down. That is, the apparatus is configured to scan the laser scanner 100 while changing the irradiation angle of the laser scanner 100 upwards and downwards, and is installed in the vehicle 10.

2 is a photograph of a vehicle 10 provided with a vehicle front object pattern recognition system using a vertical swing of a 2D laser scanner 100 according to an embodiment of the present invention. As shown in FIG. 2, the laser scanner 100 is attached to the upper end of the vehicle 10. The attached height is 1.52m. The laser scanner 100 attached to the top is used as an example for pattern recognition of a sign board. In order to read the pattern of the sign, the laser scanner 100 scans while changing the irradiation angle while shaking while setting the minimum angle to 0 deg. And the maximum angle to 3 deg.

3 shows a scanner installation configuration of a vehicle front object pattern recognition system using a vertical swing of a 2D laser scanner 100 according to an embodiment of the present invention. As shown in FIG. 3, a step motor 200 and a gear were installed on the rear side of the laser scanner 100, and the step motor 200 was controlled using ATmega 128 of ATMEL. The pulse of the step motor 200 is given at 1200 Hz. When the step motor 200 is rotated by 180 °, the laser scanner 100 has a maximum angle of 3 °. When the motor moves 360 °, the laser scanner 100 has a minimum angle of 0 °.

Although the upper and lower irradiation angles of the 2D laser scanner 100 set in the embodiment of the present invention are set at 0 to 3 degrees as in the examples of Figs. 2 and 3, It can be expanded or reduced.

1, the front object pattern recognition system of the vehicle 10 according to the embodiment of the present invention includes a central processing unit 300 for processing laser scanned data, ). The central processing unit 300 is a computer installed inside the vehicle 10 and is constituted by an MCU. The central processing unit 300 is capable of integrally controlling the scanner and the step motor 200 and performs an operation or a process of recognizing a pattern through the scanned data Lt; / RTI >

FIG. 4 is a flowchart illustrating a method of recognizing a front object pattern of a vehicle 10 using a vertical swing of a 2D laser scanner 100 according to another embodiment of the present invention. 4, a method of recognizing a front object pattern of a vehicle 10 according to an embodiment of the present invention is a pattern recognition method using a front object pattern recognition system of the vehicle 10, comprising the steps of: (a) 200) to change a vertical scanning angle of the 2D laser scanner (100) and perform laser scanning (S100); (b) extracting the data scanned by the central processing unit 300 and clustering the data (S200); (c) determining a width of the pattern by using the clustered data in the central processing unit (300) and calculating a distance to the forward object (S300); and (d) And forming and recognizing a forward object pattern (S400).

As shown in FIG. 4, the 2D laser scanner 100 installed outside the vehicle 10 in the step (a) does not simply scan the data within a specific angle to the left and right, but also changes the scanning angle of the scanner ), And scan data is obtained by scanning left and right. As described above, scanning while changing the upper and lower irradiation angles is a method for more accurately and stereoscopically recognizing a pattern of an object located in front of the vehicle 10, and clustering each pattern according to the variation of the upper and lower irradiation angles, It is possible to recognize a precise pattern of the front object by arranging or arranging them.

In the embodiment of the present invention, since the data read by the laser scanner 100 is Raw data of Hexa ASCII, it is interpreted to be usable by using National Lab Instruments LabVIEW. 5 is a graph of LMS data input and output in a method of recognizing a forward object pattern of a vehicle 10 using up and down swings of a 2D laser scanner 100 according to an embodiment of the present invention. Figure 5 (a) shows the LMS data input and Figure 5 (b) shows the LMS data output.

As shown in Fig. 5 (a), the output of Fig. 5 (b) can be obtained by using the data of the laser scanner 100 converted into the hexadecimal number. Since the laser scanner 100 scans in a plane and finds the distance, it is indicated by a dot in the graph. In order to efficiently process the data, an area to be examined with the laser scanner 100 is specified. This sets the left and right 7 m to the detection range based on the vehicle 10 on which the laser scanner 100 is installed.

In the step (b), the scan data extracted by the 2D laser scanner 100 is data clustering. To recognize an object, it is necessary to recognize the points output on the graph in LabVIEW as a single chunk. When the data of the laser scanner 100 is outputted to LabVIEW, it is output to each point rather than one chunk, so it must be recognized as a chunk and designated as the same object. Clustering each pointer into a single chunk is called clustering.

Data clustering is a method of receiving laser scanner (100) data in an array, comparing the first data with the next data, and then lengthening it to another lump if the distance is more than 1.5m and lumping it to less than 1.5m. And the first and last part of the chunk was the starting point and the ending point. We also extracted the rightmost point, leftmost point, closest point, and farthest point of the chunk, and clustered them together with the starting and ending points to create a vertex of the chunk. 6 is a graph showing clustering of scan data through a method of recognizing a forward object pattern of a vehicle 10 using up and down swings of a 2D laser scanner 100 according to another embodiment of the present invention.

In addition, in the embodiment of the present invention, since the portion of the same object, which is not easily measured by an angle, may be measured with another object, clustering is added once more to increase the reliability.

In the embodiment of the present invention, the laser scanner 100 reads data and performs a clustering operation by shaking the upward and downward irradiation angles. Find the width of the x-axis left and right end points of clustering data. At this time, since the laser scanner 100 changes the upward and downward irradiation angles, the width of each portion of the sign can be obtained. FIG. 7 is a graph showing a data pattern clustered for 1 ms on a signboard as a forward object in a method of recognizing a forward object pattern of a vehicle 10 using up and down swings of a 2D laser scanner 100 according to an embodiment of the present invention. When the width thus obtained is output to the graph, it appears as shown in the left side of FIG. 7, and the right side of FIG. 7 shows the distance information.

FIG. 8 is a schematic diagram showing a triangular scanning area of the laser scanner 100, and FIG. 9 is a pattern graph showing the result of arranging data by Lab VIEW 1D array alignment in FIG.

에 의해 구할 수 있다. 스윙하는 각도가 3°이기 때문에 r 이 x보다 큰 것을 알 수 있다. 그리고 도 9에 나타낸 바와 같이, 좌측에 있는 데이터는 폭을 나타내고 이를 역정렬 후 2로 나누어 대칭을 이루게 하고, 거리 순으로 배열을 시킨다.
As shown in Fig. 8, when the minimum angle is 0 DEG, the distance is x, and when the maximum angle is 3 DEG,

. ≪ / RTI > It can be seen that r is greater than x because the swing angle is 3 °. As shown in FIG. 9, the data on the left side represents the width, is rearranged, divided by 2 to make it symmetrical, and arranged in order of distance.

10 is a photograph showing a result (FIG. 10 (a)) of the method of recognizing the front object pattern of the vehicle 10 using the up and down swings of the 2D laser scanner 100 according to the embodiment of the present invention and a traffic sign as a forward object (Fig. 10 (b)).

As shown in FIG. 10A, after the alignment of the distance data and the width data is completed, x is the width and y is the output using the distance information, which shows the shape of the triangle.

As shown in FIG. 10, it can be seen that a pattern similar to a triangle is obtained although the result of scanning an intersection sign by applying an embodiment of the present invention is not a complete triangle shape.

If the speed of the vehicle 10 is high and the speed of the vehicle is high, the distance of the lowest angle before 1 second becomes longer than the distance of the maximum angle after 1 second. Therefore, 10) travels in a zigzag or uphill or downhill, the shape of the sign may not completely appear in the pattern. In the case of the uneven road, the shape of the pattern may be very distorted. Therefore, in the embodiment of the present invention, the angle at which the laser scanner 100 scans is also calculated and applied, The more reliable the pattern can be obtained.

As described above, the present invention has the following conclusions.

1) A pattern of an object was detected by swinging the laser scanner 100 at a height of 1.52 m from 0 ° to 3 °. It was confirmed that the change of the pattern according to the speed of the vehicle 10 was not large in the straight line section.

2) It was confirmed that the shape of the pattern and the object were almost similar when the linear section was run at a low speed.

3) When zigzag traveling at a high speed, a pattern and a shape of a target are different from each other. As an alternative, a method of adding a laser scanner 100, measuring a swing angle of the laser scanner 100, Do.

4) If the sensing direction of the 2D laser scanner 100 is fixed, there is a disadvantage in that detection of an object is delayed depending on the road environment such as uphill and downhill during autonomous driving. However, by swinging the laser scanner 100, Can be improved.

5) In order to increase the reliability of the pattern, reference information that can determine the angle information and order when the 2D laser scanner 100 reads data is needed. By sorting the data based on this reference information, a pattern with high reliability can be obtained.

6) There is a high probability that the shape of the pattern will be distorted according to the situation of the road. In order to solve this problem, it is possible to improve reliability by using the Pitch, Roll, and Yaw values of the vehicle 10 for data processing.

While the invention has been shown and described with respect to the specific embodiments thereof, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined by the appended claims. Anyone with it will know easily.

10: vehicle, 100: 2D laser scanner, 150: gear structure,
200: step motor, 300: central processing unit

Claims (8)

At least one 2D laser scanner installed outside the vehicle;
A gear structure for changing the vertical irradiation angle of the 2D racer scanner;
A stepping motor which meshes with the gear structure and is mounted on the 2D laser scanner and drives a change in the vertical irradiation angle of the 2D laser scanner;
And a central processing unit connected to the 2D laser scanner to control the 2D laser scanner and the stepping motor and to cluster the scanned data to form pattern data of a forward object. A Vehicle Front Object Pattern Recognition System Using Vertical Swing of Scanner.
The method according to claim 1,
Wherein the upper and lower irradiation angles are varied between 0 DEG and 5 DEG. The system for recognizing a vehicle front object using up and down swings of a 2D laser scanner.
3. The method according to claim 1 or 2,
Wherein the 2D laser scanner is mounted on a front surface of a vehicle and a roof panel of a vehicle.
A pattern recognition method using the vehicle front object pattern recognition system of claim 1,
(a) driving the stepping motor to change a vertical scanning angle of the 2D laser scanner and performing laser scanning;
(b) extracting scanned data from the central processing unit and clustering data;
(c) determining a width of the pattern by using the clustered data in the central processing unit and calculating a distance to the forward object;
(d) arranging a pattern through the left and right width and distance data to form and recognize a forward object pattern, and recognizing the forward object pattern using the up and down swing of the 2D laser scanner.
5. The method of claim 4,
Wherein, in the step (a), the irradiation angle is changed between 0 and 5 degrees. The method for recognizing a vehicle front object using up and down swing of a 2D laser scanner.
5. The method of claim 4,
The step (d)
Arranging the data by the array alignment through the clustered data, the left-right width and the distance data;
And rearranging the arranged data patterns so as to be symmetrical with respect to each other by dividing the rearranged data patterns by 2, and arranging the rearranged data in descending order.
7. The method according to any one of claims 4 to 6,
The step (a)
And extracting the upper and lower irradiation angle data when the 2D laser is scanned.
7. The method according to any one of claims 4 to 6,
Further comprising the step of extracting a pitch, a roll and a yaw value of the vehicle using an acceleration sensor or a tilt sensor to correct the scan data according to the posture of the vehicle A Method of Pattern Recognition of Vehicle Front Object Using Up / Down Swing of 2D Laser Scanner.

Images