KR102173776B1 - Method and device for generating extended plane reflectivity map - Google Patents

Abstract

It relates to a method of generating an extended plane reflectivity map, wherein the method of generating an extended plane reflectivity map comprises the steps of: (a) receiving reflectivity data of a road plane in a tunnel and reflectivity data of a wall in a tunnel from a reflectance data acquisition sensor, (b) the road plane Generating a reflectivity image of a road plane and a reflectance image of a wall using the reflectivity data of and (c) projecting the reflectivity image of the wall to both sides of the reflectivity image of the road plane And generating an extended plane reflectivity map in which the reflectivity image of and the reflectivity image of the wall are on the same plane.

Description

Method and device for generating extended plane reflectivity map {METHOD AND DEVICE FOR GENERATING EXTENDED PLANE REFLECTIVITY MAP}

The present application relates to a method and apparatus for generating an extended plane reflectivity map.

For autonomous driving, precise location information of the vehicle is required. In general, location information is obtained from a Global Positioning System (GPS), but in an environment with many signal blocks or multipaths, location information is inaccurate, and GPS signals transmitted from tunnels cannot be received.

In addition, position estimation in a short-distance road section can be estimated through dead reckoning (DR). However, as the driving distance increases, the position error increases due to the accumulation of errors, and the accumulated position error of the guessed navigation must be corrected using various sensors used in autonomous vehicles. When using a camera, errors in the lateral position of the tunnel can be corrected by using lane detection, but there is no detection point such as a lane in the longitudinal direction of the vehicle, and only some shape points can be extracted from the tunnel. Difficult to correct

Moreover, since all lanes are solid lines in a tunnel, when a positioning technique using a road surface reflectivity map is applied, the position in the longitudinal direction cannot be estimated. Due to this limitation, there is a problem in that the position of the autonomous vehicle cannot be determined through a sensor provided in the autonomous vehicle.

The technology behind the present application is disclosed in Korean Patent Publication No. 10-1428239.

The present application is to solve the problems of the prior art described above, by using 3D LIDAR reflectance information to extract the characteristic points of the lane and the wall of the tunnel, the extended plane reflectivity that can grasp the position of the vehicle in the lateral and longitudinal directions inside the tunnel. It is intended to provide a method and apparatus for generating a map.

The present application is to solve the above-described problems of the prior art, and is to provide a method and apparatus for generating an extended plane reflectivity map that provides a map form applicable to 3D LIDAR-based positioning in a tunnel.

The present application is to solve the problems of the prior art described above, by generating a single 2D map in which the tunnel wall information is extended to the existing road surface reflectivity map, an extended plane reflectivity map that can be positioned by performing general 2D correlation matching only once. It is intended to provide a method and apparatus for generating.

However, the technical problems to be achieved by the embodiments of the present application are not limited to the technical problems as described above, and other technical problems may exist.

As a technical means for achieving the above technical problem, the method of generating an extended plane reflectivity map according to an embodiment of the present application includes: (a) reflectivity data of a road plane in a tunnel and reflectivity data of a wall in a tunnel from a reflectivity data acquisition sensor. Receiving, (b) generating a reflectivity image of a road plane and a reflectivity image of a wall using the reflectivity data of the road plane and the reflectivity data of the wall, and (c) the reflectivity image of the wall surface of the road plane. It may include the step of projecting the reflectivity image to both sides to generate an extended plane reflectivity map in which the reflectivity image of the road plane and the reflectivity image of the wall surface exist on the same plane.

According to an exemplary embodiment of the present application, in the step (c), an area exceeding a preset height is removed from among the reflectance images of the wall, and the reflectance image of the wall from which the area exceeding the preset height is removed is converted to the road. The reflectivity of the plane may be projected to both sides of the image.

According to an embodiment of the present disclosure, the reflectance data acquisition sensor may include a 3D LIDAR sensor.

According to an exemplary embodiment of the present disclosure, the reflectivity data of a wall surface in the tunnel may include reflectance data of a wall surface and reflectance data of a structure installed on a wall surface having characteristics different from the reflectivity data of the wall surface. .

According to an embodiment of the present application, the structure installed on the wall in the tunnel may include a fire hydrant or an evacuation guide.

According to an exemplary embodiment of the present disclosure, removing an area exceeding a predetermined height among the reflectivity image of the wall may remove at least some of the area exceeding the height of the reflectivity image of a structure installed on the wall.

According to an exemplary embodiment of the present application, in the step (b), a plurality of points are extracted using reflectivity data of the road plane and reflectance data of the wall surface, and an occupied grid map is generated using the extracted points. An edge measurement value may be extracted through matching of the occupied grid maps accumulated based on nodules.

According to an embodiment of the present application, a vehicle positioning method in a tunnel using an extended plane reflectivity map includes the steps of generating an extended plane reflectivity map in the tunnel using the method of claim 1, and correlation using the extended plane reflectivity map in the tunnel. It may include the step of locating the lateral and longitudinal positions of the vehicle in the tunnel by matching.

According to an embodiment of the present disclosure, the apparatus for generating an extended plane reflectivity map includes: a data receiving unit receiving reflectivity data of a road plane in a tunnel and reflectivity data of a wall surface in a tunnel from a reflectance data acquisition sensor, reflectivity data of the road plane, and the wall surface An image generator that generates a reflectivity image in the tunnel using the reflectivity data of and projects the reflectance image of the wall among the reflectivity images in the tunnel to both sides of the reflectivity image of the road plane to generate an extended plane reflectivity map in the tunnel. It may include a map generator.

According to an embodiment of the present disclosure, the vehicle positioning system in the tunnel includes a reflectivity data acquisition sensor that acquires reflectivity data in a tunnel, a reflectivity data of a road plane in a tunnel from a reflectivity data acquisition sensor, and data that receives reflectivity data of a wall surface in the tunnel. A receiving unit, an image generating unit that generates a reflectivity image in the tunnel using reflectivity data of the road plane and the reflectivity data of the wall, and converts the reflectivity image of the wall among the reflectivity images in the tunnel to both sides of the reflectivity image of the road plane. It may include a map generator for generating an extended plane reflectivity map in the tunnel by projecting, and a positioning part for positioning the lateral and vertical positions of the vehicle in the tunnel by performing correlation matching using the extended plane reflectivity map in the tunnel. .

The above-described problem solving means are merely exemplary and should not be construed as limiting the present application. In addition to the above-described exemplary embodiments, additional embodiments may exist in the drawings and detailed description of the invention.

According to the above-described problem solving means of the present application, by extracting characteristic points of lanes and walls of a tunnel using 3D LIDAR reflectance information, it is possible to grasp the position of the vehicle in the lateral direction and the longitudinal direction inside the tunnel.

According to the above-described problem solving means of the present application, it is possible to provide a map applicable to vehicle positioning based on 3D LIDAR in a tunnel.

According to the above-described problem solving means of the present application, by generating a single 2D map in which the tunnel wall information is extended to the existing road surface reflectivity map, it is possible to generate an extended plane reflectivity map that can be positioned by performing general 2D correlation matching only once. have.

However, the effect obtainable in the present application is not limited to the effects as described above, and other effects may exist.

1 is a schematic diagram of a vehicle positioning system in a tunnel according to an embodiment of the present application.
2 is a schematic block diagram of an apparatus for generating an extended plane reflectivity map according to an exemplary embodiment of the present disclosure.
3 is a map of a camera image in a tunnel and a road reflectivity in the section according to an embodiment of the present application.
FIG. 4 is a view in which reflectivity data for fire hydrants and evacuation guidance, etc. acquired by an extended plane reflectivity map generating apparatus according to an exemplary embodiment of the present disclosure are imaged.
5 is a view for explaining a process of generating an extended planar reflectivity map in the extended planar reflectivity map generating apparatus according to an embodiment of the present application.
6 is an extended planar reflectivity map generated by an extended planar reflectivity map generating apparatus according to an embodiment of the present application.
FIG. 7 is a diagram illustrating an overlap of a map generated using two data of different times around an RTK/INS location in an extended plane reflectivity map generating apparatus according to an exemplary embodiment of the present disclosure.
8 is a result of estimating a location using a road reflectivity map in an extended plane reflectivity map generating apparatus according to an exemplary embodiment of the present application.
9 is a result of estimating a location using a composite reflectivity map in an extended plane reflectivity map generating apparatus according to an exemplary embodiment of the present application.
10 is a flowchart illustrating a method of generating an extended plane reflectivity map according to an embodiment of the present application.
11 is a flowchart illustrating an operation of a vehicle positioning method in a tunnel using an extended plane reflectivity map according to an embodiment of the present application.

Hereinafter, exemplary embodiments of the present application will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art may easily implement the present application. However, the present application may be implemented in various different forms and is not limited to the embodiments described herein. In addition, in the drawings, parts not related to the description are omitted in order to clearly describe the present application, and similar reference numerals are attached to similar parts throughout the specification.

Throughout this specification, when a part is said to be "connected" with another part, this includes not only the case that it is "directly connected", but also the case that it is "electrically connected" with another element interposed therebetween. do.

Throughout this specification, when a member is positioned "on", "upper", "upper", "under", "lower", and "lower" of another member, this means that a member is located on another member. It includes not only the case where they are in contact but also the case where another member exists between the two members.

Throughout the specification of the present application, when a certain part "includes" a certain component, it means that other components may be further included rather than excluding other components unless specifically stated to the contrary.

The present application may provide a method of generating an extended plane reflectivity map capable of positioning a vehicle in a tunnel using a road reflectivity map and a tunnel wall reflectivity map.

1 is a schematic diagram of a vehicle positioning system in a tunnel according to an embodiment of the present application, and FIG. 2 is a schematic block diagram of an apparatus for generating an extended plane reflectivity map according to an exemplary embodiment of the present application. Referring to FIGS. 1 and 2, the vehicle positioning system 100 in the tunnel determines the location of the vehicle in the tunnel by using an extended plane reflectivity map in the tunnel from the reflectivity data acquisition sensor 2 provided in the vehicle 1. As for positioning, referring to FIG. 1, the information (data) obtained from the reflectivity data acquisition sensor 2 is transmitted to the positioning device 20 to generate an extended plane reflectivity map by the positioning device 20, and the generated extension It is possible to locate a vehicle in a tunnel through correlation matching using a planar reflectivity map. The positioning device 20 may receive information obtained from the reflectivity data acquisition sensor 2 through the network 30. For example, the vehicle 1 may be an autonomous vehicle.

Referring to FIG. 1, the vehicle positioning system 100 in a tunnel may include a reflectivity data acquisition sensor 2 and a positioning device 20.

The reflectivity data acquisition sensor 2 may acquire reflectivity data in the tunnel. The reflectivity data acquisition sensor 2 may be provided in a roof of the vehicle 1. The reflectivity data acquisition sensor 2 is provided in the roof of the vehicle 1 to acquire reflectivity data of a road plane in the tunnel and a wall surface in the tunnel. The reflectivity data acquisition sensor 1 may include a 3D LIDAR sensor. 3D Light Detection and Ranging (LIDAR) can accurately measure the distance to structures separated from tunnel walls such as signboards or jet fans, and is more efficient in correcting the longitudinal position error than when using a camera. For example, the reflectivity data acquisition sensor 2 may be a Velodyne HDL-32E 3D LIDAR sensor. The reflectance data acquisition sensor 2 is a 3D laser scanner having 32 channels, and can acquire distance and reflectivity information about the surroundings in a horizontal period of about 0.16 degrees while rotating 360 degrees using 32 channels in the vertical direction. As the reflectance data acquisition sensor 2 provides the LIDAR reflectance data, the positioning device 20 can extract feature points of lanes and tunnel walls.

The positioning device 20 may include an extended plane reflectivity map generating device 10 and a positioning unit 21. The positioning device 20 may perform correlation matching using an extended planar reflectivity map in a tunnel generated using the extended planar reflectivity map generating method to locate the lateral and vertical positions of the vehicle in the tunnel. The horizontal direction may mean the left and right direction of the vehicle, and the vertical direction may mean the traveling direction of the vehicle.

3 is a map of a camera image in a tunnel and a road reflectivity in the section according to an embodiment of the present application.

Referring to FIG. 3, (a) of FIG. 3 is an image of a camera in the tunnel, and (b) of FIG. 3 is a map of road reflectivity in the section. As shown in the camera image of Fig. 3A, there is hardly any special structure protruding from the wall in the tunnel. Referring to (b) of FIG. 3, it can be seen that the features for estimating the longitudinal position also do not appear on the road reflectivity map of the same section.

However, in the tunnel, a fire hydrant as shown in Fig. 3 (a) and an evacuation guide light as shown in No. 2 are periodically installed at regular intervals, respectively. These safety facilities (such as fire hydrants and evacuation guidance) have very different reflectivity characteristics from the surrounding walls. The extended planar reflectivity map generating apparatus 10 may generate an extended planar reflectivity map in the tunnel using the road reflectivity map and the tunnel wall reflectivity map using the reflectivity characteristics.

Referring to FIG. 2, the extended plane reflectivity map generating apparatus 10 may include a data receiving unit 11, an image generating unit 12 and a map generating unit 13. However, the configuration of the extended plane reflectivity map generating apparatus 10 is not limited to those previously disclosed. For example, the extended plane reflectivity map generating apparatus 10 may include a database for storing information (data).

According to an exemplary embodiment of the present disclosure, the data receiver 11 may receive reflectivity data of a road plane in a tunnel and reflectance data of a wall surface in a tunnel from the reflectivity data acquisition sensor 2. The data receiver 11 may receive reflectivity data of a road plane and reflectance data of a wall surface in a tunnel through the network 30.

The reflectivity data of the road plane may include reflectivity data for road surface information such as lanes, direction indications, and crosswalks. The reflectivity data of a wall surface in the tunnel may include reflectivity data of a wall surface and reflectivity data of a structure installed on a wall surface having characteristics different from the reflectivity data of the wall surface. At this time, the structure installed on the wall in the tunnel may include a fire hydrant or an evacuation guide.

The network 30 refers to a wired and wireless connection structure capable of exchanging information between respective nodes such as a terminal and a server, and examples of such networks include a 3rd Generation Partnership Project (3GPP) network and a Long Term (LTE). Evolution) network, 5G network, WIMAX (World Interoperability for Microwave Access) network, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network) ), a Bluetooth network, a satellite broadcasting network, an analog broadcasting network, a Digital Multimedia Broadcasting (DMB) network, and the like, but are not limited thereto.

FIG. 4 is a view in which reflectivity data for fire hydrants and evacuation guidance, etc. acquired by an extended plane reflectivity map generating apparatus according to an exemplary embodiment of the present disclosure are imaged.

For example, (a) of FIG. 4 may be an image of reflectivity data for a fire hydrant. In addition, (b) of FIG. 4 may be an image of reflectance data for an evacuation guide. The image generator 120 may generate a reflectivity image in the tunnel by using a feature in which an area corresponding to a fire hydrant and an evacuation guide is clearly distinguished from the surrounding wall surface.

According to the exemplary embodiment of the present disclosure, the image generator 120 generates a reflectance image in the tunnel using the difference in reflectivity of the tunnel wall, thereby correcting a position error in the longitudinal direction of the vehicle.

The image generator 12 may generate a reflectivity image in the tunnel by using reflectivity data of a road plane and reflectance data of a wall surface. The reflectivity image in the tunnel may include a reflectivity image of a road plane and a reflectivity image of a wall surface.

The image generator 12 may extract a plurality of points by using reflectivity data of a road plane and reflectance data of a wall in order to generate a reflectivity image. In addition, the image generator 12 may generate an occupied grid map using the extracted points. In addition, the image generator 12 may extract edge measurements through matching of the occupied grid maps accumulated based on each node generated in the occupied grid map. The image generator 12 may generate reflectance images 3 and 4 in the tunnel using the extracted edge measurements.

In other words, the image generator 12 extracts the ground from the distance information of the reflectivity data acquisition sensor 2 to generate the reflectivity image 3 of the road plane, and then scans the road surface display using the reflectivity data of the road plane. One point can be extracted. The image generator 12 may generate a 2D occupancy grid map using the extracted points and perform image matching using the 2D occupancy grid map. At this time, since the number of scan pointers of one rotation data is very small, matching is not possible. Therefore, the image generator 12 may accumulate the scan points extracted as road surface marks according to the movement of the vehicle 1. The image generator 12 may extract edge measurements through matching of the occupied grid maps accumulated based on each node.

For example, in order to generate an occupied grid map including only the reflectivity image 3 of the road plane, the image generating unit 12 uses the first algorithm from the reflectivity data of the road plane obtained from the reflectance data acquisition sensor 2 For example, the RANSACN algorithm) can be used to extract points scanned on the ground. The image generator 12 may set the normal vector of the plane corresponding to the ground from the reflectivity data of the road plane and set a threshold value (for example, 5 cm) in the predicted RANSACN model in order to extract a point scanned on the ground. have. The image generator 12 may remove data of a vertical structure having a height such as a vehicle or a curb by a plane predicted by a normal vector and a threshold value set to be small, and extract only a ground point. The image generator 12 sets a reflectivity value that can distinguish road asphalt and road surface information in order to extract a road surface point from the extracted ground point cloud, and extracts only points by scanning the road surface mark based on the set value. I can. In addition, the image generator 12 may extract points obtained by scanning road surface information in consideration of the height (eg, 2.2 m from the ground) of the reflectivity data acquisition sensor 2 provided in the vehicle.

In addition, in order to solve the problem that it is difficult to calculate an accurate edge measurement value due to insufficient information, the image generator 12 extracts a continuous node within a preset distance (eg, 10m) around the reference node The reflectivity image (3) of can be accumulated. Here, the node may refer to the position and attitude of the vehicle acquired from GPS/INS as a node, and the relationship between the nodes may be referred to as an edge. The edge can be calculated as a geometric relationship between two nodes using location information obtained from GPS. The relationship between the two nodes extracted from the reflectance data acquisition sensor 2 may be defined as an edge measurement value. If the location information of all nodes acquired from GPS is accurate without error, you can create a precise map based on the node. However, the node contains errors due to factors such as multipath signals in an environment where GPS signal reception cannot be guaranteed. Therefore, it is necessary to calculate the position error by comparing the precise sensor data measured at the node.

When the image generator 12 accumulates the road surface scan data measured by the extracted node in consideration of the moving distance of the vehicle, it may fill in the road surface information that cannot be scanned by the distance between points for each channel. Accordingly, the image generator 12 may calculate an accurate edge measurement value through a cross-correlation technique between occupied grid maps in which the amount of road surface information is increased.

Also, the image generator 12 may perform a 2D cross correlation technique to calculate an edge measurement value using the generated occupied grid map. The cross-correlation technique is one of the Template Matching techniques and is used in the image processing field to calculate the similarity of one image to another. The edge measurement between two nodes can be calculated by adding the error between the inaccurate nodes calculated from the sensor data at the edge. The image generator 12 can confirm that the exact error is calculated through the cross-correlation technique and the road surface accumulated information measured by each node is exactly the same, and through the occupied grid map generated using the accumulated road surface information Edge measurements can be calculated.

The image generator 12 may calculate by adding an error between inaccurate nodes calculated through sensor data at the edge and adding edge measurements between the two nodes. For example, the image generator 12 may calculate an edge error existing between initial nodes by performing a two-dimensional cross-correlation technique using an occupied grid map generated from two nodes extracted by a loop closure condition.

A two-dimensional cross-correlation technique using an occupied grid map generated based on two nodes can be expressed as [Equation 1] below.

[Equation 1]

Here, (x _i , y _i ) denotes the position of the grid in the occupied grid map, and Corr denotes the cross-correlation value in each grid. The occupied grid map consisting of 0 and 1 is multiplied by the value of each grid to estimate the grid position with the largest sum of products (Equation 2), and the grid position with the maximum value is multiplied by the set grid size to determine the actual two. The position error between nodes is calculated. (Equation 3)

[Equation 2]

[Equation 3]

In this way, the image generator 12 may calculate the edge measurement value by adding the error calculated through [Equation 1] to [Equation 3] to the edge, which is a positional relationship between initial nodes. (Equation 4)

[Equation 4]

Accordingly, the image generator 12 may calculate an accurate edge measurement value through an occupied grid map generated using the road surface accumulated information. When calculating the relationship between two images using a cross-correlation technique, if the physical area is processed in a full search method, the number of calculations may vary depending on the grid size. As an example, the image generator 12 may perform a cross-correlation technique using a 2D FFT technique in order to solve a problem that takes a long time to calculate due to the complete search method. If the 2D FFT technique is used, frequency analysis is applied by considering the value of the grid as a signal while following the occupied grid map in the x- and y-axis directions. Since the FFT divides the real data of the occupied grid map into imaginary elements to appear in the frequency domain, the frequency constellations are equal to the number of pixels in the coordinates of the actual map.

[Equation 5] below expresses the calculation method when calculating the cross-correlation technique using 2D FFT.

[Equation 5]

는 2차원 FFT함수이고,

는 역변환 FFT 함수이고,

는 FFT를 수행할 때 0Hz 요소가 중앙에 오도록 수행하는 함수이고,

는 FFT를 이용한 상호상관 결과이다. here,

Is a two-dimensional FFT function,

Is the inverse transform FFT function,

Is a function that performs the FFT so that the 0Hz element is in the center,

Is the cross-correlation result using FFT.

The image generator 12 may generate the reflectance image 4 for the tunnel wall in the same manner as the method for generating road plane (road surface) information described above.

In other words, the image generator 12 uses the first algorithm (for example, from the reflectivity data in the tunnel obtained from the reflectivity data acquisition sensor 2) to generate an occupied grid map including only the reflectivity image 4 of the wall in the tunnel. For example, RANSACN algorithm) can be used to extract a point scanned by the tunnel wall. The image generator 12 sets the normal vector of the plane corresponding to the structure from the reflectivity data of the wall surface in the tunnel to extract the point scanned by the tunnel wall, and calculates a threshold value (for example, 5 cm) from the predicted RANSACN model. Can be set. The image generator 12 sets a reflectivity value that can distinguish the wall surface from the structure installed on the wall surface in order to extract the wall surface point from the extracted wall point cloud, and extracts only the points by which the wall display is scanned based on the set value. I can. The image generation unit 12 sets a reflectivity value that can distinguish between the wall structure and the wall information in order to extract the wall point from the extracted wall point cloud, and extracts only the point scanned for the wall structure display of the tunnel based on the set value. can do.

In addition, in order to solve the problem that it is difficult to calculate an accurate edge measurement value due to insufficient information, the image generator 12 extracts a continuous node within a preset distance (eg, 10m) around the reference node and The reflectivity image 4 of the wall can be accumulated. When the image generator 12 accumulates the scan data of the wall measured by the extracted node in consideration of the moving distance of the vehicle, it may fill in the wall information in the tunnel that could not be scanned by the distance between points for each channel. Accordingly, the image generator 12 may calculate an accurate edge measurement value through a cross-correlation technique between occupied grid maps in which the amount of road surface information is increased.

According to the exemplary embodiment of the present disclosure, the map generator 13 may generate an extended plane reflectivity map in the tunnel by projecting the reflectance image of the wall among the reflectance images in the tunnel to both sides of the reflectivity image of the road plane. The map generator 13 uses the reflectivity data for the structure installed on the wall surface, which has different characteristics from the reflectivity data for the wall surface as shown in FIGS. 4A and 4B. You can create a map in the same shape as a map.

5 is a view for explaining a process of generating an extended planar reflectivity map in the extended planar reflectivity map generating apparatus according to an embodiment of the present application.

Referring to FIG. 5A, the map generator 13 may generate a reflectivity image in the tunnel by projecting the reflectance image of the road plane generated by the image generator 12 onto the reflectance image of the wall. In other words, FIG. 5A may be a reflectivity image in a tunnel generated by combining a reflectivity image of a road plane and a reflectivity image of a wall surface.

Referring to FIG. 5B, the map generator 13 may remove an area exceeding a preset height from among the reflectance images of the wall surface. In other words, the map generation unit 13 may remove an area of a predetermined height or more with respect to the upper end of the fire hydrant and the evacuation guide from the vertical section of the tunnel. The area exceeding the preset height may be an area in which a structure installed on the wall does not exist. When the map generator 13 removes the area exceeding the predetermined height among the reflectivity images of the wall, at least some of the areas exceeding the height of the reflectivity image of the structure installed on the wall may be removed.

According to an exemplary embodiment of the present application, when the map generator 13 removes an area exceeding a predetermined height among the reflectivity images of the wall, at least some of the areas exceeding the height of the reflectivity image of a structure installed on the wall are removed. Can be. For example, structures installed on the wall (eg, fire hydrants and evacuation guidance) may be installed at a certain height and period. The map generator 1 may remove an area that exceeds a predetermined height from the top of the structure installed on the wall.

Referring to FIG. 5C, the map generator 13 may project the reflectance image of the wall surface from which the area exceeding the predetermined height is removed to both sides of the reflectivity image of the road plane. In other words, the map generator 13 may project a reflectance image on a wall surface like a developed diagram of a three-dimensional figure on both sides of the reflectivity image on the road plane.

Referring to (d) of FIG. 5, the map generator 13 may generate an extended plane reflectivity map existing on one 2D plane by projecting a reflectivity map for a vertical wall onto a road plane. For example, referring to FIG. 6, it can be seen that the map generation unit 13 generates a map by clearly dividing areas corresponding to 1 (fire fighting) and 2 (evacuation guidance, etc.) shown in FIG. 6 from the surrounding wall surface. .

6 is an extended planar reflectivity map generated by an extended planar reflectivity map generating apparatus according to an embodiment of the present application.

Referring to FIG. 6, the map generator 13 may generate an extended plane reflectivity map 5 using a reflectance image of a wall surface and a reflectance image of a road plane among reflectance images in a tunnel. As shown in FIG. 6, it can be seen that the areas where 1 (fire hydrant) and 2 (evacuation guidance, etc.) are located are clearly distinguished from the surrounding walls in the map. The extended plane reflectivity map 5 may be generated based on the reflectivity data 3 of the road plane and the reflectivity data 4 of the wall surface.

According to an exemplary embodiment of the present disclosure, the positioning unit 21 may perform correlation matching using an expanded plane reflectivity map in the tunnel to locate the lateral and longitudinal positions of the vehicle in the tunnel. Correlation matching is one of template matching techniques, and may be a cross-correlation technique used when calculating the similarity between one image and another image in the image processing field. In other words, the positioning unit 21 may position the vehicle even in a tunnel through a single correlation matching using the extended plane reflectivity map generated through the extended plane reflectivity map generating method. The positioning unit 21 generates a single 2D map in which the tunnel wall information (reflectivity image of the wall surface) is expanded to the existing road surface reflectivity map (reflection image of the road plane), thereby performing general 2D correlation matching only once and Position can be positioned.

According to an exemplary embodiment of the present disclosure, the vehicle positioning system 100 may transmit the position of the vehicle to be measured by the positioning unit 21 to a control module (not shown) of the vehicle 2. The vehicle lane positioning system 100 provides the current vehicle location information to the control module of the vehicle 2, thereby helping to drive more safely even in a tunnel where GPS signals cannot be received.

According to an embodiment of the present disclosure, the vehicle positioning system 100 may measure a result of correlation matching with an extended plane reflectivity map in an extended Kalman Filter (EKF). The designed EKF is time updated and measurement updated. It can be classified as

First, the time update uses the output value of GPS/DR. For example, for a GPS/DR sensor, Microinfinity's CruizCore DS6200 can be used, and the azimuth accuracy in an open space is within 5°. The filter state variable can be expressed as [Equation 6] below.

[Equation 6]

Here, δx and δy represent the 2D horizontal position error of the vehicle in the ENU coordinate system.

In addition, the equation of state can be expressed as [Equation 7].

[Equation 7]

In [Equation 7], the error of GPS/DR did not change for a short time.

Next, the measurement equation can be expressed as [Equation 8].

[Equation 8]

및

는 상관관계 일치의 결과를 나타낸다. here,

And

Represents the result of correlation agreement.

Measurement update can be performed as shown in [Equation 9].

[Equation 9]

Here, K is the Kalman gain.

FIG. 7 is a diagram illustrating an overlap of a map generated using two data of different times around an RTK/INS location in an extended plane reflectivity map generating apparatus according to an exemplary embodiment of the present disclosure.

7(a) shows the outer cross section of the tunnel, and FIGS. 7(b) and 7(c) show the inner cross section of the tunnel.

Referring to (a) of FIG. 7, two maps acquired at different time periods may be accurately overlapped when road lines coincide. This agreement indicates that the RTK/INS post-processing results are very accurate. However, as shown in (b) of FIG. 7, some road lines may not be accurately matched. This error may be an error caused by the grid size of the map. In addition, it can be seen that the area of the emergency exit is almost invisible compared to the error in the longitudinal direction as shown in FIGS. 7(b) and 7(c). Since the tunnel in which this experiment was performed was almost straight, it can be expected that the longitudinal position error will be less than the transverse error after correction in consideration of the exact position of the starting and ending points of the tunnel. These errors are marked as errors in the generation of the extended plane reflectivity map and can ultimately be reflected in localization errors.

FIG. 8 is a result of estimating a location using a road reflectivity map (a reflectivity image of a road plane) in an extended plane reflectivity map generating apparatus according to an exemplary embodiment of the present disclosure.

The red line in FIG. 8 indicates the result of location destruction through correlation matching, and the blue line indicates the result of GPS/DR. The vertical pink lines represent the starting and ending points of the tunnel. As can be seen from FIG. 8, since there is a road inside the tunnel, the lateral position error is minimal. However, since correlation matching was hardly performed, it can be confirmed that the longitudinal position error is very large. For example, the RMS position error of the tunnel cross section is 0.19m and 1.39m, respectively. As described above, the location of the vehicle in the tunnel section cannot be determined using only the road reflectivity map (the reflectivity image of the road plane).

9 is a result of estimating a location using a composite reflectivity map in an extended plane reflectivity map generating apparatus according to an exemplary embodiment of the present application.

Referring to FIG. 9, the lateral position error is minimal as when a road reflectivity map (a reflectivity image of a road plane) is used. In addition, the longitudinal position error was also limited to a certain level of error. It can be seen that this is because the correlation matching performance in the longitudinal direction is improved due to the influence of fire hydrants and evacuation guidance. In addition, referring to FIG. 9, it can be seen that the end position error of the GPS/DR at the end point of the tunnel changes abruptly, but this is due to the sudden correction of the accumulated DR error due to re-reception of the GPS signal when the vehicle leaves the tunnel. to be. As a result, the GPS/DR location input was greatly changed during the time update, affecting the correlation matching. In conclusion, the position error estimated by the filter changed greatly, and the changed error was directly expressed as the position error.

Hereinafter, based on the details described above, the operation flow of the present application will be briefly described.

The extended planar reflectivity map generation method illustrated in FIG. 10 may be performed by the extended planar reflectivity map generating apparatus 10 described above. Therefore, even if omitted below, the description of the extended planar reflectivity map generating apparatus 10 may be equally applied to the description of the extended planar reflectivity map generating method.

10 is a flowchart illustrating a method of generating an extended plane reflectivity map according to an embodiment of the present application.

In step S101, the extended plane reflectivity map generating apparatus 10 may receive reflectivity data of a road plane in the tunnel and reflectance data of a wall surface in the tunnel from a reflectivity data acquisition sensor.

In step S102, the extended plane reflectivity map generating apparatus 10 may generate a reflectance image of a road plane and a reflectance image of a wall by using the reflectivity data of the road plane and the reflectance data of the wall surface.

In step S102, the extended plane reflectivity map generating apparatus 10 projects the reflectance image of the wall to both sides of the reflectivity image of the road plane to generate an extended plane reflectivity map in which the reflectivity image of the road plane and the reflectance image of the wall exist on the same plane. can do.

The vehicle positioning method in the tunnel using the extended plane reflectivity map shown in FIG. 11 may be performed by the vehicle positioning device 20 described above. Therefore, even if omitted below, the description of the vehicle positioning apparatus 20 may be equally applied to the description of the vehicle positioning method in the tunnel using the extended plane reflectivity map.

11 is a flowchart illustrating an operation of a vehicle positioning method in a tunnel using an extended plane reflectivity map according to an embodiment of the present application.

In step S201, the vehicle positioning apparatus 20 may generate an expanded plane reflectivity map in the tunnel using the expanded plane reflectivity map generation method.

In step S202, the vehicle positioning apparatus 20 may perform correlation matching using an expanded plane reflectivity map in the tunnel to locate the lateral and longitudinal positions of the vehicle in the tunnel.

In the above description, steps S101 to S103 and S201 to S202 may be further divided into additional steps or may be combined into fewer steps, depending on the embodiment of the present application. In addition, some steps may be omitted as necessary, and the order between steps may be changed.

The method for generating an extended plane reflectivity map and a method for positioning a vehicle in a tunnel using the extended plane reflectivity map according to an exemplary embodiment of the present disclosure may be implemented in the form of program commands that can be executed through various computer means and recorded in a computer-readable medium. . The computer-readable medium may include program instructions, data files, data structures, and the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -A hardware device specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of the program instructions include not only machine language codes such as those produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The above-described hardware device may be configured to operate as one or more software modules to perform the operation of the present invention, and vice versa.

In addition, the above-described extended planar reflectivity map generation method and the vehicle positioning method in a tunnel using the extended planar reflectivity map may be implemented in the form of a computer program or application executed by a computer stored in a recording medium.

The foregoing description of the present application is for illustrative purposes only, and those of ordinary skill in the art to which the present application pertains will be able to understand that it is possible to easily transform it into other specific forms without changing the technical spirit or essential features of the present application. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not limiting. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope of the present application is indicated by the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present application.

1: car
2: Reflectivity data acquisition sensor
10: extended plane reflectivity map generating device
11: data receiver
12: image generator
13: Map generator
20: vehicle positioning device
30: network

Claims (11)

In the extended plane reflectivity map generation method,
(a) receiving reflectivity data of a road plane in the tunnel and reflectivity data of a wall surface in the tunnel from a reflectivity data acquisition sensor;
(b) generating a reflectivity image of a road plane and a reflectance image of a wall by using the reflectivity data of the road plane and the reflectivity data of the wall surface; And
(c) projecting the reflectivity image of the wall surface to both sides of the reflectivity image of the road plane to generate an extended plane reflectivity map in which the reflectivity image of the road plane and the reflectivity image of the wall surface exist on the same plane,
Including,
The step (c),
An extended plane reflectivity of removing an area exceeding a preset height among the reflectivity images of the wall surface, and projecting the reflectivity image of the wall surface from which the area exceeding the preset height has been removed to both sides of the reflectivity image of the road plane. How to create a map. delete The method of claim 1,
The reflectivity data acquisition sensor includes a 3D LIDAR sensor. The method of claim 1,
The reflectivity data of the wall surface in the tunnel includes reflectivity data for the surface of the wall and reflectance data for structures installed on the wall having different characteristics from the reflectance data for the surface of the wall. The method of claim 4,
The structure installed on the wall in the tunnel includes a fire hydrant or an evacuation guide. The method of claim 4,
The method of generating an extended planar reflectivity map, wherein the removing of the area exceeding the predetermined height among the reflectivity image of the wall surface is to remove at least a portion of the area exceeding the height of the reflectivity image of the structure installed on the wall surface. The method of claim 1,
The step (b),
By extracting a plurality of points using the reflectivity data of the road plane and the reflectivity data of the wall surface, generating an occupied grid map using the extracted points, and matching the occupied grid maps accumulated based on each road The method of generating an extended planar reflectivity map to extract the measurements. In the vehicle positioning method in the tunnel using an extended plane reflectivity map,
Generating an extended plane reflectivity map in the tunnel using the method of claim 1; And
Locating the lateral and longitudinal positions of the vehicle in the tunnel by performing correlation matching using the expanded plane reflectivity map in the tunnel,
Vehicle positioning method in the tunnel comprising a. In the extended plane reflectivity map generating device,
A data receiver configured to receive reflectivity data of a road plane in the tunnel and reflectance data of a wall surface in the tunnel from the reflectivity data acquisition sensor;
An image generator that generates a reflectivity image in the tunnel using reflectivity data of the road plane and the reflectivity data of the wall; And
A map generation unit for generating an extended plane reflectivity map in the tunnel by projecting the reflectivity image of the wall among the reflectivity images in the tunnel to both sides of the reflectivity image of the road plane,
Including,
The map generator,
An extended plane reflectivity of removing an area exceeding a preset height among the reflectivity images of the wall surface, and projecting the reflectivity image of the wall surface from which the area exceeding the preset height has been removed to both sides of the reflectivity image of the road plane. Map generating device. In the vehicle positioning system in the tunnel,
A reflectivity data acquisition sensor that acquires reflectivity data in the tunnel;
A data receiver configured to receive reflectivity data of a road plane in the tunnel and reflectance data of a wall surface in the tunnel from the reflectivity data acquisition sensor;
An image generator that generates a reflectivity image in the tunnel using reflectivity data of the road plane and the reflectivity data of the wall;
A map generator configured to generate an extended plane reflectivity map in the tunnel by projecting the reflectivity image of the wall among the reflectivity images in the tunnel to both sides of the reflectivity image of the road plane; And
A positioning unit for positioning the vehicle's lateral and longitudinal positions in the tunnel by performing correlation matching using the expanded plane reflectivity map in the tunnel,
Including,
The map generator,
A vehicle in a tunnel in which an area exceeding a predetermined height is removed from among the reflectivity images of the wall surface, and the reflectivity image of the wall surface from which the area exceeding the predetermined height is removed is projected to both sides of the reflectivity image of the road plane. Positioning system. A computer-readable recording medium on which a program for executing the method of any one of claims 1 and 3 to 7 is recorded on a computer.

Images