KR102087046B1 - Method and apparatus for providing information of a blind spot based on a lane using local dynamic map in autonomous vehicle - Google Patents

Abstract

Disclosed are a method and an apparatus for providing information on a blind spot based on vehicle lanes using a local dynamic map (LDM). According to the present invention, the method of providing information on a blind spot based on vehicle lanes using the LDM comprises: a step of acquiring the central line in the vehicle lanes located adjacent to the current vehicle by using the LDM; a step of acquiring a three-dimensional coordinate value on a point where the blind spot created by a first vehicle overlaps with the central line in the vehicle lanes; a step of converting the acquired three-dimensional coordinate value into a reference coordinate system on the current vehicle and creating a blind spot display section; and a step of displaying the created blind spot display section to a user. Accordingly, the present invention is able to precisely identify a blind spot created when a vehicle drives autonomously.

Description

METHOD AND APPARATUS FOR PROVIDING INFORMATION OF A BLIND SPOT BASED ON A LANE USING LOCAL DYNAMIC MAP IN AUTONOMOUS VEHICLE}

The present invention relates to a method and apparatus for providing information on a blind spot on a lane basis using LDM in an autonomous vehicle. More particularly, the present invention relates to a user based on a blind spot that cannot be directly observed in an autonomous vehicle. The present invention relates to a method and apparatus for displaying and detecting a vehicle located in a blind spot and displaying the same to a user.

Autonomous vehicles are vehicles that can sense the surrounding environment and movement with little or no human manipulation or input. For these autonomous vehicles to be realized, various sensors capable of detecting surrounding conditions, such as radar, computer vision, GPS, odometry and inertial measurement units, must be combined. .

At this time, the autonomous vehicle observes the surrounding road conditions using various sensors such as a radar and a camera. In this case, when there is a specific vehicle or an obstacle in the sensor area detected by the autonomous vehicle, a blind spot may not be detected due to the vehicle or the obstacle. In this case, the existing autonomous vehicles display only the detected vehicles or objects to the driver, and do not display the undetected blind spots, so it is difficult for the driver to pay attention to the blind spots. In addition, in determining and sensing a forward position to be detected, a curvature and a slope of a lane are currently used, and thus a sensing point may not be accurately specified.

In addition, a technique of providing vehicle driving assistance information to the autonomous vehicle is used to easily identify a surrounding traffic situation that is difficult for the autonomous vehicle to detect. For example, LDM (Local Dynamic Map) is a technology or three-dimensional spatial information that links, stores, and manages standardized vehicle driving support information for autonomous driving, and is based on a car-level precision electronic map (or static information). By combining road traffic and surrounding vehicle status (dynamic information), it provides information to autonomous vehicles in real time.

By the way, the conventional autonomous vehicles are mostly merely collecting and using the LDM information, and there is a lack of a method for improving the operation reliability of the autonomous driving by processing the LDM information.

An object of the present invention for solving the above problems is to provide a method for providing information on blind spots on a lane basis using LDM.

Another object of the present invention for solving the above problems is to provide an apparatus for providing information on blind spots on a lane basis using LDM.

One aspect of the present invention for achieving the above object, provides a method for providing information on the blind spot on a lane-based by using the LDM.

A method of providing information on a blind spot based on a lane using an LDM may include obtaining a lane center line for a lane located near a current vehicle using a local dynamic map (LDM), which is formed by the first vehicle. Acquiring a three-dimensional coordinate value for a point where a blind spot and the lane center line overlap each other, generating a blind spot display section by converting the obtained three-dimensional coordinate value into a reference coordinate system for the current vehicle, and generating the blind spot And displaying the zone display section to the user.

The method for providing information on the blind spot based on the lane may include detecting lane center points of two lanes forming the lane before obtaining the lane center line, detected in each of the two lanes. Detecting a lane center point by averaging lane center points from each other, combining the detected lane center points to detect the lane center line, and updating the LDM using the detected lane center line; can do.

After acquiring the center line of the lane, acquiring a three-dimensional coordinate value of a point to be sensed by using the three-dimensional coordinate value of the obtained lane center line, and referring the acquired three-dimensional coordinate value to the current vehicle. The method may include converting to a coordinate system, acquiring sensing data by detecting a point corresponding to the converted 3D coordinate value using a sensor, and verifying the acquired sensing data.

The detecting of the center points of the lane may include obtaining point cloud data measured on a road surface on which the two lanes are painted by using a LIDAR device, and centering an input point on the obtained point cloud data. Extracting points within a constant radius, obtaining a lane direction vector through principal component analysis of the extracted points, and separating the extracted points into points located on a plane perpendicular to the direction vector. And detecting the lane center point by analyzing the reflection intensity with respect to the separated points.

In the method for providing information on the blind spot based on the lane, after generating the blind spot display section, acquiring a three-dimensional coordinate value for the blind spot display section and obtaining the obtained three-dimensional coordinate value; The method may further include transmitting to the first vehicle.

The method for providing information on the blind spot based on the lane may include: after transmitting the acquired 3D coordinate value to the first vehicle, of the second vehicle located in the blind spot display section from the first vehicle; Acquiring a position coordinate value, converting the acquired position coordinate value into a reference coordinate system for the current vehicle, and displaying the second vehicle to the user using the converted position coordinate value. have.

Another aspect of the present invention for achieving the above object provides an apparatus for providing information on the blind spot on a lane-based basis using the LDM.

An apparatus for providing information on a blind spot based on a lane using LDM may include a memory configured to store at least one processor and instructions instructing the at least one processor to perform at least one step. (memory) may be included.

The at least one step may include obtaining a lane center line for a lane located near the current vehicle by using a local dynamic map (LDM), and a blind spot formed by a first vehicle and a point where the lane center line overlaps with each other. Obtaining a 3D coordinate value for the target, converting the obtained 3D coordinate value into a reference coordinate system for the current vehicle, generating a blind spot display section, and displaying the generated blind spot display section to a user; can do.

The at least one step may include detecting lane center points for two lanes forming the lane before obtaining the lane center line, and averaging lane center points detected in each of the two lanes with each other. The method may further include detecting a center point, combining the detected center points to detect the lane center line, and updating the LDM using the detected lane center line.

After acquiring the center line of the lane, acquiring a three-dimensional coordinate value of a point to be sensed by using the three-dimensional coordinate value of the obtained lane center line, and referring the acquired three-dimensional coordinate value to the current vehicle. The method may include converting to a coordinate system, acquiring sensing data by detecting a point corresponding to the converted 3D coordinate value using a sensor, and verifying the acquired sensing data.

The detecting of the center points of the lane may include obtaining point cloud data measured on a road surface on which the two lanes are painted by using a LIDAR device, and centering an input point on the obtained point cloud data. Extracting points within a constant radius, obtaining a lane direction vector through principal component analysis of the extracted points, and separating the extracted points into points located on a plane perpendicular to the direction vector. And detecting the lane center point by analyzing the reflection intensity with respect to the separated points.

The at least one step may include, after generating the blind spot display section, obtaining a three-dimensional coordinate value for the blind spot display section and transmitting the obtained three-dimensional coordinate value to the first vehicle. It may further include.

The at least one step may include: obtaining position coordinate values of a second vehicle positioned in the blind spot display section from the first vehicle after transmitting the obtained three-dimensional coordinate values to the first vehicle; The method may further include converting the acquired position coordinate value into a reference coordinate system for the current vehicle, and displaying the second vehicle to the user using the converted position coordinate value.

When using the method and apparatus for providing information on the blind spot based on the lane using the LDM (Local Dynamic Map) according to the present invention as described above, it is possible to call attention by displaying the lane based blind spot to the user have.

In addition, there is an advantage that can accurately detect the three-dimensional coordinate information based on the lane by using the lane center line obtained from the LDM.

In addition, V2V communication has the advantage of being able to determine the position of other vehicles or obstacles located in the blind spot.

In addition, by correcting the sensing data sensed by the vehicle by using the information obtained from the LDM, it is possible to reduce the blind spot of the autonomous vehicle and to increase the operation reliability.

1 is an exemplary diagram for describing a process of forming a blind spot in a curved section.
2 is a conceptual diagram illustrating a method of generating a blind spot display section according to an exemplary embodiment of the present invention.
3 is an exemplary diagram for describing input data used to detect a lane center point for the LDM according to an embodiment of the present invention.
4 is an exemplary diagram for describing a process of detecting a lane direction vector in a process of detecting a lane center point for an LDM according to an embodiment of the present invention.
FIG. 5 is an exemplary diagram for describing a process of separating input data into a plane perpendicular to a lane direction vector in a process of detecting a lane center point for an LDM according to an embodiment of the present invention.
FIG. 6 is an exemplary view illustrating a result of separating input data into a plane perpendicular to the lane direction vector according to an embodiment of the present invention. FIG.
FIG. 7 is an exemplary diagram for describing a process of approximating a parabola to point cloud data located in a plane perpendicular to a lane direction vector according to an embodiment of the present invention.
FIG. 8 is an exemplary diagram for describing a method of determining a lane center point using a plurality of parabolas derived according to FIG. 7.
9 is a flowchart illustrating a process of detecting a lane center point according to an embodiment of the present invention.
FIG. 10 is a first flowchart of a method of providing information on a blind spot based on a lane using LDM according to an embodiment of the present invention.
FIG. 11 is a second flowchart of a method of providing information on a blind spot on a lane basis using LDM according to an embodiment of the present invention.
12 is a conceptual diagram illustrating a method of determining whether an obstacle exists and correcting sensing data according to an embodiment of the present invention.
FIG. 13 is a flowchart illustrating an embodiment of a method of identifying presence of an obstacle and correcting sensing data using an LDM according to an embodiment of the present invention.
14 is a hardware block diagram of an apparatus for providing information on a blind spot based on a lane using LDM according to an embodiment of the present invention.

As the invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. In describing the drawings, similar reference numerals are used for similar elements.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component. The term and / or includes a combination of a plurality of related items or any item of a plurality of related items.

When a component is said to be "connected" or "connected" to another component, it may be directly connected to or connected to that other component, but it may be understood that another component may be present in the middle. Should be. On the other hand, when a component is said to be "directly connected" or "directly connected" to another component, it should be understood that there is no other component in between.

The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting of the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, the terms "comprise" or "have" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art. Terms such as those defined in the commonly used dictionaries should be construed as having meanings consistent with the meanings in the context of the related art, and shall not be construed in ideal or excessively formal meanings unless expressly defined in this application. Do not.

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

1 is an exemplary diagram for describing a process of forming a blind spot in a curved section.

Referring to FIG. 1, when the current vehicle 10 is an autonomous vehicle, the current vehicle 10 may detect surrounding traffic conditions using various sensors mounted therein. For example, various sensors may be lidar, camera, or the like.

Here, the first vehicle 20 may be detected using sensors of the current vehicle 10. However, the sensors currently mounted on the vehicle 10 by the first vehicle 20 may not detect an area beyond the first vehicle 20. That is, as shown in FIG. 1, the blind spot 40 that is not detectable by the current vehicle 10 is formed in front of the first vehicle 20.

In addition to the curved line as shown in FIG. 1, when the slope is severe, the blind spot 40 may be formed because the front area located on the same lane as the current vehicle 10 is not detected by the sensors of the current vehicle 10. have.

As such, since the blind spot 40 may be formed in various ways according to the terrain or lane shape, the general autonomous vehicle recognizes the second vehicle 30 located in the blind spot 40, various objects and obstacles, and the like. It cannot be displayed to the user.

In addition, since the current vehicle 10 cannot detect the second vehicle 30 located in the blind spot 40, the second vehicle 30 that is driving the same lane as the current vehicle 10 suddenly brakes. In various dangerous situations, including the current vehicle 10 is difficult to cope quickly.

In order to solve this problem, the blind spot 40 can be clearly displayed to the user to draw attention of the user and to prompt the user to cope with an emergency situation. In addition, as an autonomous vehicle, the blind spot 40 may be explicitly checked to perform an operation of additionally identifying various vehicles, obstacles, and the like located in the blind spot 40.

Hereinafter, a method of recognizing and displaying the blind spot 40 and the vehicle 30 located in the blind spot 40 will be described.

2 is a conceptual diagram illustrating a method of generating a blind spot display section according to an exemplary embodiment of the present invention.

In general, when the entire area not detected by the sensor is set to the blind spot, an excessively wide area may be set to the blind spot, and there is a problem that it is difficult to clearly identify the detailed position of the set blind spot.

Since autonomous vehicles traveling on roads generally drive on roads with lanes, it may be advantageous to set blind spots among areas located in lanes. In addition, rather than simply setting the area located in the lane to the blind spot, the current vehicle is obtained by acquiring three-dimensional coordinates corresponding to the centerline by the lane and converting the acquired three-dimensional coordinates into a reference coordinate system for the current vehicle. As a reference, the coordinates of the blind spot can be easily identified.

Referring to FIG. 2, a method of displaying a blind spot by a current vehicle 10 that runs along a curved section, first, the current vehicle 10 is a lane centerline 50 for a lane adjacent to a lane on which the current vehicle 10 travels. Can be detected. For example, the lane centerline 50 may be detected in a lane in which the current vehicle 10 travels or in a lane located to the left or right of the lane in which the current vehicle 10 travels. Here, the lanes may be sections formed by lanes each painted one to the right. Therefore, if the three-dimensional coordinate values of the lane center point can be detected in each of the two lanes located on both sides of the lane, the three-dimensional coordinate value of the lane center point is calculated by averaging the three-dimensional coordinate values of the detected two lane center points. can do. The set of coordinate values of the center point of the lane calculated here may be defined as the lane centerline 50.

In this case, the lane center point may mean a coordinate located at the center of coordinates in which the lane exists in the assumed coordinate system, assuming a coordinate system showing a cross section perpendicular to the ground surface on which the lane is drawn. A method of detecting a lane center point for a local dynamic map (LDM) may refer to FIGS. 3 to 9 to be described later.

The lane center line 50 may be pre-detected using a vehicle in which various sensors are embedded and stored in advance as additional information in the LDM. The LDM in which the lane center line 50 is added may be stored and managed in an external LDM server. It may also be stored in the internal memory of each individual autonomous vehicle. Accordingly, the autonomous vehicle may receive three-dimensional coordinates of the lane centerline 50 from an external LDM server or directly obtain three-dimensional coordinates of the lane centerline 50 through the LDM stored in the internal memory. have.

Here, LDM (Local Dynamic Map) includes a fixed map that forms a permanent map with primary information, and is secondary information on a fixed map, and various road facilities, structures, traffic signs, and lane centerlines belonging to a temporary fixed body. And the like, and may include information such as road condition, traffic volume, signal type of the current intersection, real-time location and moving direction and speed of various moving objects, and distance from the car as the third information (or dynamic information). .

Meanwhile, when the lane center line 50 is detected, coordinate values overlapping the blind spot may be obtained from the three-dimensional coordinate values of the detected lane center line 50. For example, when the coordinate plane of the ground surface on which the vehicle 10 currently runs is represented by (x, y), the blind spot may be represented by a range of the (x, y) coordinate system. At this time, since the lane centerline 50 further has a vertical coordinate axis (referred to as the z-axis) of the ground, the (x, y, z) of the set of (x, y, z) coordinate values for the lane centerline 50 is excluded (x, y) Three-dimensional coordinate values whose coordinates belong to the blind spot can be extracted. The three-dimensional coordinate value extracted here may correspond to a linear three-dimensional coordinate located in the blind spot among the coordinate values for the center line 50 by the lane.

The blind spot display section 45 may be generated by converting the extracted three-dimensional three-dimensional coordinates into a reference coordinate system for the current vehicle 10. In this case, the reference coordinate system may be a three-dimensional coordinate system centering on the current vehicle. Accordingly, the blind spot display section 45 may be represented by three-dimensional coordinates in the reference coordinate system or a three-dimensional distance and a direction angle Φ with the current vehicle 10 in the reference coordinate system.

Therefore, according to an embodiment of the present invention, since the blind spot based on the center line of the lane can be obtained as a three-dimensional coordinate value, the current vehicle 10 more accurately checks the spots where obstacles or vehicles located in the blind spot may exist. I can figure it out.

3 is an exemplary diagram for describing input data used to detect a lane center point for the LDM according to an embodiment of the present invention.

The input data obtained through the LiDAR device used to build the LDM consists of a number of irregular point clouds (referred to as point clouds or point groups) with three-dimensional spatial coordinates (X, Y, Z coordinates). Can be. Since such input data may have data beyond pixel data having existing two-dimensional coordinates, it is advantageous to configure LDM in the form of three-dimensional spatial information.

Referring to FIG. 3, an example of input data obtained for a road on which one or more lanes are painted may be identified using a LIDAR device. The input data obtained as in FIG. 3 may be a point cloud composed of numerous points having three-dimensional spatial coordinates. In this case, the individual points may have additional properties such as reflection intensity and color value in addition to having essentially three-dimensional spatial coordinates.

At this time, the lane painted on the road tends to be damaged from the outside of the lane because the vehicle enters from the outside of the lane. Therefore, the reflection intensity measured for the lane painted on the road surface may have a larger value as the vehicle moves toward the center of the lane and a smaller value as it approaches the boundary. Accordingly, an embodiment of the present invention proposes a method of accurately determining the lane center point by approximating the reflection intensity to a parabola by using this characteristic.

4 is an exemplary diagram for describing a process of detecting a lane direction vector in a process of detecting a lane center point for an LDM according to an embodiment of the present invention.

In order to detect a lane center point according to an embodiment of the present invention, first, an arbitrary point (or arbitrary spatial coordinates) may be selected for the input data according to FIG. 3. In this case, any point may be obtained as input data through a user's input means, or may be obtained using a random variable.

Referring to FIG. 4, an example in which an input point 10 received from a user is displayed on three-dimensional space coordinates may be confirmed. In FIG. 4, for convenience of description, the input point 60 is illustrated to be positioned at the center of the 3D space coordinates, but may also be represented by a specific space coordinate.

As shown in FIG. 4, when the input point 60 is obtained, points within a predetermined distance from the input point 60 may be extracted. For example, points (points shown in FIG. 4) within a predetermined radius r about the input point 60 may be extracted from the input data.

Next, points having a reflection intensity greater than or equal to a reference value may be extracted from points within a predetermined radius r. Here, the points whose reflection intensity is above the reference value can be expected to be points located on the painted lane. Thus, the reference value may be determined as a median or average value between the average reflection intensity for the painted lane and the average reflection intensity for portions other than the painted lane. Also, the reference value may be determined as a value corresponding to a preset ratio (for example, 80%) based on the average reflection intensity value for the painted lane.

When the points having the reflection intensity greater than or equal to the reference value are extracted, the lane direction vector 61 may be derived by performing principal component analysis (PCA). For example, the spatial coordinate distribution of points having a reflection intensity greater than or equal to a reference value may be analyzed, and the direction in which the analyzed spatial coordinate distribution is most widely (or long) distributed may be determined as the lane direction vector 61.

FIG. 5 is an exemplary diagram for describing a process of separating input data into a plane perpendicular to a lane direction vector in a process of detecting a lane center point for an LDM according to an embodiment of the present invention. FIG. 6 is an exemplary view illustrating a result of separating input data into a plane perpendicular to the lane direction vector according to an embodiment of the present invention. FIG.

If the points within the predetermined radius r shown in FIG. 4 are classified into points located on a plurality of planes 62 perpendicular to the lane direction vector 61, the points are as shown in FIG.

Referring to FIG. 5, the points located on the plane 62 perpendicular to the lane direction vector may be identified. Here, the points located on one plane 62 may be points located on one straight line because the points measured with respect to the road surface by the lidar apparatus. Referring to FIG. 6, the data obtained by classifying the point cloud data acquired on the road surface into a plurality of planes perpendicular to the lane direction vector may be confirmed. As shown in FIG. 5, the points located on the plane 62 perpendicular to the lane direction vector correspond to a straight line (for example, a cross section of the road plane perpendicular to the lane direction because the plane is measured on the actual road plane). Can be understood as a straight line). Thus, points located on one plane may be located on one straight line (or approximated straight line) in the spatial coordinate system.

FIG. 7 is an exemplary diagram for describing a process of approximating a parabola to point cloud data located in a plane perpendicular to a lane direction vector according to an embodiment of the present invention.

Referring to FIG. 7, the linear coordinates where the points located on the plane 62 perpendicular to the lane direction vectors derived from FIGS. 4 to 5 are located as the horizontal axis, and the reflection intensity values of the points are set as the vertical axis. You can check the graph. Referring to the graph shown in FIG. 7, it is possible to confirm the distribution of the reflection intensity of points located on one linear coordinate. The central the reflection intensity value is, the more the center is located, as in the general lane painting trend. It can be seen that the reflection intensity value is small. In this case, if the reflection intensity distribution according to FIG. 7 is approximated to a parabola, the approximate parabola 63 illustrated in FIG. 7 may be derived. The vertex coordinates of the approximate parabola 40 derived here may be expected to be close to the center point of the lane.

As described above, since an approximate parabola is derived based on the reflection intensity with respect to points located on one linear coordinate, the center point of the lane even if there is a reflection intensity error of some point or damage to the painted surface. Can be derived accurately.

FIG. 8 is an exemplary diagram for describing a method of determining a lane center point using a plurality of parabolas derived according to FIG. 7.

Since the approximate parabola derived in FIG. 7 is obtained by using points located on a plane perpendicular to the direction vector of the lane, it may be difficult to compensate for lane painting damage of all of the points. Therefore, according to an exemplary embodiment of the present invention, a more precise lane center point can be derived by deriving an approximate parabola for each individual plane and approximating the vertices of the derived parabolic parabola with one straight line.

Specifically, referring to FIG. 8, a straight line 64 approximating the vertices of the approximate parabola derived for each individual plane according to FIG. 5 with one straight line may be identified. The point where the vertical lines are connected to each other by the vertical line 64 from the input point 60 obtained in FIG. 4 may be a spatial coordinate position corresponding to the center point of the lane.

9 is a flowchart illustrating a process of detecting a lane center point according to an embodiment of the present invention.

Referring to FIG. 9, in the method for detecting a lane center point for LDM, obtaining point cloud data measured on a road surface on which a lane is painted using a Lidar device (S100), and obtaining the point cloud Extracting points within a predetermined radius of the input point from the data (S110), acquiring a lane direction vector through principal component analysis of the extracted points (S120), and extracting the extracted points from the direction vector. Separating into points located on a vertical plane (S130) and analyzing the reflection intensity for the separated points may include detecting the lane center point (S140).

The input point may be determined through a user input.

The obtaining of the direction vector of the lane (S120) may include extracting points having a reflection intensity greater than or equal to a reference value among the extracted points, and obtaining the direction vector through principal component analysis of the extracted points. can do.

The reference value may be an intermediate value or an average value between the average reflection intensity for the painted lane and the average reflection intensity for portions other than the painted lane.

The reference value may be a value corresponding to a preset ratio based on the average reflection intensity value for the painted lane.

The lane center point detection method for the LDM may be performed by a vehicle equipped with a lidar device.

In addition, the lane center point detection method for the LDM may further include updating the LDM using the detected lane center line.

In addition, the method for detecting the lane center point for the LDM may be applied to the method or the steps described with reference to FIGS.

FIG. 10 is a first flowchart of a method of providing information on a blind spot based on a lane using LDM according to an embodiment of the present invention. FIG. 11 is a second flowchart of a method of providing information on a blind spot on a lane basis using LDM according to an embodiment of the present invention.

Referring to FIG. 10, a method of providing information on a blind spot based on a lane using a local dynamic map (LDM) may include obtaining a lane center line for a lane currently located around the vehicle using the LDM (S200). ), Obtaining a three-dimensional coordinate value for the point where the blind spot formed by the first vehicle and the center line of the lane overlap (S210), by converting the obtained three-dimensional coordinate value into a reference coordinate system for the current vehicle A step of generating a blind spot display section (S220) and a step of displaying the generated blind spot display section to a user (S230).

Here, the first vehicle, as shown in FIGS. 1 and 2, may be a vehicle that is detected by the current vehicle and may form a blind spot for the current vehicle.

Obtaining the lane center line (S200) may include receiving the lane center line from an external local dynamic map (LDM) server. Alternatively, the obtaining of the lane center line (S200) may include obtaining the lane center line using an LDM stored in an internal memory.

The lane center line may be detected before the step of obtaining the lane center line (S200), and the LDM may be updated using the detected lane center line. For example, a method of providing information on a blind spot based on a lane may include detecting lane center points of two lanes forming the lane before obtaining the lane center line (S200). Detecting a lane center point by averaging the lane center points detected in each of the lanes, combining the detected lane center points to detect the lane center line, and updating the LDM using the detected lane center line The method may further include). For a process of detecting the lane center line from the lane center point, the description according to FIG. 2 may be referred to, and thus a detailed description thereof is omitted.

Detecting the lane center points may refer to the description of FIGS. 3 to 9. For example, the detecting of the lane center points may include obtaining point cloud data measured on a road surface on which the two lanes are painted by using a Lidar (LIDAR) device. Extracting points within a certain radius around the input point, obtaining a lane direction vector through principal component analysis of the extracted points, and extracting the points on a plane perpendicular to the direction vector And detecting the lane center point by analyzing the reflection intensity of the separated points.

On the other hand, the method for providing information on the blind spot based on the lane using the LDM according to an embodiment of the present invention not only generates a blind spot display section to display to the user, but also obstacles, vehicles located in the blind spot display section Etc. may be detected and displayed to the user through vehicle-to-vehicle (V2V) communication.

For example, after generating the blind spot display section (S220), obtaining a 3D coordinate value for the blind spot display section and transmitting the obtained 3D coordinate value to the first vehicle. It may further include. The three-dimensional coordinate value for the blind spot display section may be a three-dimensional coordinate value according to the previous step S210, or may be a three-dimensional coordinate value converted into a reference coordinate system according to the step S220. In this case, the transmitting of the 3D coordinate value to the first vehicle may be performed by a broadcasting method. For example, a method of providing information on a blind spot on a lane basis using LDM may generate and generate a transmission message including a 3D coordinate value for the blind spot display section and an identifier for the first vehicle. Broadcasted messages can be broadcast. When a vehicle other than the first vehicle receives the transmission message, the identification symbol for the first vehicle may be confirmed and discarded. On the other hand, the first vehicle may check the identification symbol included in the transmission message and obtain a 3D coordinate value included in the transmission message. In addition, the first vehicle may detect whether an obstacle, a vehicle, or the like exists by performing sensing using a sensor mounted on a position corresponding to the obtained 3D coordinate value. When the first vehicle detects an obstacle, a vehicle, or the like, the first vehicle may transmit three-dimensional coordinate values of the detected obstacle, vehicle, etc. to the preceding current vehicle.

Therefore, after the step of transmitting the obtained three-dimensional coordinate value to the first vehicle, obtaining a position coordinate value of the second vehicle located in the blind spot display interval from the first vehicle, the obtained position coordinate value The method may further include converting a to a reference coordinate system for the current vehicle and displaying the second vehicle to the user using the converted position coordinate value.

Here, the second vehicle may be a vehicle that is located on a blind spot display section that is derived on a lane basis and belongs to the blind spot formed by the first vehicle as shown in FIGS. 1 and 2.

After acquiring the center line of the lane (S200), acquiring a three-dimensional coordinate value of a point to be sensed using the three-dimensional coordinate value of the obtained lane center line, and obtaining the obtained three-dimensional coordinate value from the current vehicle. The method may include converting a reference coordinate system into a reference coordinate system, obtaining sensing data by detecting a point corresponding to the converted 3D coordinate value using a sensor, and verifying the acquired sensing data.

Method for providing information on the blind spot on a lane basis using the LDM according to an embodiment of the present invention is an apparatus for providing information on the lane-based blind spot mounted on an autonomous vehicle, a semi-autonomous vehicle or a vehicle ( 13) to be described later.

12 is a conceptual diagram illustrating a method of determining whether an obstacle exists and correcting sensing data according to an embodiment of the present invention.

When the autonomous vehicle (or the current vehicle) detects a vehicle or various objects at a specific point in front of the vehicle using various sensors such as a camera (for example, following a vehicle in front of the vehicle using cruise control) It is essential to accurately identify the point ahead and to obtain sensing data using the sensors for the identified point.

In this case, when there is an obstacle due to another vehicle or terrain between a specific point in front of the current vehicle and the like, the sensing data may be data obtained for the obstacle instead of the vehicle or object located at the specific point.

For example, referring to FIG. 12, a scenario in which the current vehicle 10 detects (or follows) the second vehicle 30 using various sensors such as a camera may be considered. In this case, the position of the second vehicle 30 located in front of the vehicle may be specified by using the curvature and the slope of the lane in which the vehicle 10 is currently located, but there is a high possibility of error when the terrain or the lane is not constant. Referring to FIG. 12, although the second vehicle 30 is a vehicle that is located in front of a path on which the vehicle 10 currently runs, the position of the second vehicle 30 is difficult to be clearly identified since the lane is rapidly rotating. . In order to solve this problem, in one embodiment of the present invention, the current vehicle 10 may obtain a three-dimensional coordinate value for the point where the second vehicle 30 exists. In this case, the 3D coordinate value of the point where the second vehicle 30 exists may be obtained through the LDM, may be obtained through step S320 of FIG. 11, or may be obtained based on the center line 50 by the lane. .

For example, based on the lane center line 50, first, as in step S200, a lane center line for a lane located near the current vehicle 10 (shown in the same lane as the current vehicle in FIG. 12) may be obtained. . In this case, the three-dimensional coordinate values for the lane centerline may be obtained through the methods described above and then uploaded (or updated) to the LDM. Next, three-dimensional coordinate values of the position of the second vehicle 30 may be obtained by acquiring coordinate values corresponding to the position of the second vehicle 30 among the three-dimensional coordinate values of the obtained center line. .

When the 3D coordinate value for the point where the second vehicle 30 exists is obtained, the obtained 3D coordinate value may be converted into a reference coordinate system for the current vehicle 10. Next, the sensing data may be obtained by detecting a point corresponding to the three-dimensional coordinate value previously converted into the reference coordinate system using sensors currently mounted in the vehicle 10. At this time, since the first vehicle 20 is located in the same sensing target direction as the second vehicle 30, the obtained sensing data is data sensed for the first vehicle 20 instead of the second vehicle 30. Can be. In particular, in the case of a sensor such as a camera there is a high possibility of such a problem. In order to solve this problem, according to an embodiment of the present invention, the sensing data may be verified by checking the presence of an obstacle (here, the first vehicle 20). For example, by using the LDM (or received from the LDM server), it is determined whether an obstacle exists between the current vehicle 10 and the second vehicle 30 to be sensed. The presence possibility can be displayed to the user or the sensing data can be corrected for an obstacle rather than a second vehicle.

Therefore, when the autonomous vehicle wants to detect a specific point in front of the vehicle, by additionally providing whether there is an obstacle between the point to be detected and the current vehicle, the error of the sensing data is detected in advance so that the autonomous vehicle can be detected. Reduce blind spots.

FIG. 13 is a flowchart illustrating an embodiment of a method of identifying presence of an obstacle and correcting sensing data using an LDM according to an embodiment of the present invention.

Referring to FIG. 13, a method of determining the presence of an obstacle and modifying sensing data by using a local dynamic map (LDM) may include obtaining a lane center line for a lane currently located around the vehicle using the LDM ( S400, obtaining a three-dimensional coordinate value for the point to be detected using the three-dimensional coordinate value for the center line with the obtained car (S410), converting the obtained three-dimensional coordinate value to the reference coordinate system for the current vehicle In operation S420, the method may include obtaining sensing data by detecting a point corresponding to the converted 3D coordinate value using a sensor in operation S430, and verifying the obtained sensing data in operation S440.

Here, steps S410 to S440 may also be interpreted as being performed after step S200 according to FIG. 10. In addition, if it is assumed that the point to be sensed as the second vehicle according to FIG. 11, it may be interpreted that steps S420 to S440 are performed after step S310 of FIG. 11.

The verifying of the obtained sensing data (S440) may include detecting the presence of an obstacle and correcting the sensing data according to the detected result. The detecting of the presence of an obstacle may include determining whether an obstacle exists between the current vehicle and the point to be detected by using an LDM, and displaying an existence of the obstacle to a user if the obstacle exists. .

In addition, it should be interpreted that the method of detecting the presence of an obstacle using LDM and correcting the sensing data may include the steps or methods described with reference to FIG. 12.

14 is a hardware block diagram of an apparatus for providing information on a blind spot on a lane basis using LDM according to an embodiment of the present invention.

Referring to FIG. 14, an apparatus 100 for providing information on a blind spot based on a lane using a local dynamic map (LDM) may include at least one processor 110 and at least one processor. It may include a memory (120) for storing instructions (instructions) instructing to perform the step of.

Here, the at least one processor 110 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to embodiments of the present invention are performed. Can be. Each of the memory 120 and the storage device 160 may be configured as at least one of a volatile storage medium and a nonvolatile storage medium. For example, the memory 120 may be configured as at least one of a read only memory (ROM) and a random access memory (RAM).

In addition, the apparatus 100 for providing the information on the blind spot based on the lane using the LDM, the transceiver 130 for performing communication through a vehicle-to-everything (V2X) based wireless network (transceiver) (130) It may include. In addition, the apparatus 100 for providing information on the blind spot based on the lane using the LDM may further include an input interface device 140, an output interface device 150, and a storage device 160. Each component included in the apparatus 100 that provides lane-based information on the blind spot using the LDM may be connected by a bus 170 to communicate with each other.

The at least one step may include obtaining a lane center line for a lane located near the current vehicle by using LDM, and a three-dimensional coordinate value for a point where the blind spot formed by the first vehicle and the lane center line overlap each other. The method may include acquiring, generating a blind spot display section by converting the obtained three-dimensional coordinate value into a reference coordinate system for the current vehicle, and displaying the generated blind spot display section to the user.

After acquiring the center line of the lane, acquiring a three-dimensional coordinate value of a point to be sensed by using the three-dimensional coordinate value of the obtained lane center line, and referring the acquired three-dimensional coordinate value to the current vehicle. The method may include converting to a coordinate system, acquiring sensing data by detecting a point corresponding to the converted 3D coordinate value using a sensor, and verifying the acquired sensing data.

The at least one step may include detecting lane center points for two lanes forming the lane before obtaining the lane center line, and averaging lane center points detected in each of the two lanes with each other. The method may further include detecting a center point, combining the detected center points to detect the lane center line, and updating the LDM using the detected lane center line.

The detecting of the center points of the lane may include obtaining point cloud data measured on a road surface on which the two lanes are painted by using a LIDAR device, and centering an input point on the obtained point cloud data. Extracting points within a constant radius, obtaining a lane direction vector through principal component analysis of the extracted points, and separating the extracted points into points located on a plane perpendicular to the direction vector. And detecting the lane center point by analyzing the reflection intensity with respect to the separated points.

The at least one step may include, after generating the blind spot display section, obtaining a three-dimensional coordinate value for the blind spot display section and transmitting the obtained three-dimensional coordinate value to the first vehicle. It may further include.

The at least one step may include: acquiring a position coordinate value of a second vehicle positioned in the blind spot display section from the first vehicle after transmitting the obtained three-dimensional coordinate value to the first vehicle; The method may further include converting the acquired position coordinate value into a reference coordinate system for the current vehicle, and displaying the second vehicle to the user using the converted position coordinate value.

In addition, although not shown in the drawing, an apparatus 100 for providing information on a blind spot based on a lane using an LDM may include an object located on a nearby road such as a GPS, a camera, a LIDAR, a radar, and the like. It may further include a sensing module for detecting the presence or location of the vehicle, obstacles and the like.

In addition, the apparatus 100 for providing information on the blind spot based on the lane using the LDM, the surrounding road situation, the surrounding detected or received from a sensing module, an external LDM server, a vehicle, a roadside unit, etc. A vehicle that manages a vehicle, an object, an LDM, generates a control signal for controlling the vehicle driving based on the blind spot display section or the detected position of the second vehicle, and transmits the generated control signal to the driving controller inside the vehicle. It may further include an LDM module.

The methods according to the invention can be implemented in the form of program instructions that can be executed by various computer means and recorded on a computer readable medium. Computer-readable media may include, alone or in combination with the program instructions, data files, data structures, and the like. The program instructions recorded on the computer readable medium may be those specially designed and constructed for the present invention, or may be known and available to those skilled in computer software.

Examples of computer readable media may include hardware devices that are specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions may include high-level language code that can be executed by a computer using an interpreter, as well as machine code such as produced by a compiler. The hardware device described above may be configured to operate with at least one software module to perform the operations of the present invention, and vice versa.

In addition, the above-described method or apparatus may be implemented by combining all or part of the configuration or function, or may be implemented separately.

Although described above with reference to a preferred embodiment of the present invention, those skilled in the art various modifications and variations of the present invention without departing from the spirit and scope of the invention described in the claims below I can understand that you can.

Claims (10)

A method of providing blind spot information on a lane basis using a local dynamic map (LDM), which is performed in an autonomous vehicle,
Acquiring a lane centerline for the lane located near the current vehicle using the LDM;
Obtaining a three-dimensional coordinate value for a point where the blind spot formed by the first vehicle and the center line of the lane overlap;
Generating a blind spot display section by converting the obtained three-dimensional coordinate values into a reference coordinate system for the current vehicle; And
And displaying the generated blind spot display section to the user, the information on the blind spot based on the lane using the LDM. In claim 1,
Prior to acquiring the centerline with the car,
Detecting lane center points for the two lanes forming the lane;
Detecting lane center points by averaging lane center points detected in each of the two lanes with each other;
Combining the detected lane center points to detect the lane center line; And
And updating the LDM using the detected lane centerline. 11. The method of claim 1, further comprising updating the LDM using lane detection. In claim 1,
After acquiring a centerline with the car,
Obtaining a three-dimensional coordinate value for a point to be sensed using the three-dimensional coordinate value for the center line with the obtained difference;
Converting the obtained three-dimensional coordinate value into a reference coordinate system for the current vehicle;
Acquiring sensing data by detecting a point corresponding to the converted three-dimensional coordinate value using a sensor; and
And verifying the acquired sensing data, providing information on the blind spot on a lane basis using LDM. In claim 1,
After generating the blind spot display interval,
Obtaining a three-dimensional coordinate value for the blind spot display section; And
And transmitting the obtained three-dimensional coordinate values to the first vehicle, the information on the blind spot based on the lane using the LDM. In claim 4,
After the step of transmitting the obtained three-dimensional coordinate value to the first vehicle,
Obtaining a position coordinate value of a second vehicle positioned in the blind spot display section from the first vehicle;
Converting the acquired position coordinate value into a reference coordinate system for the current vehicle; And
And displaying the second vehicle to the user by using the converted position coordinate value. A device that provides information on blind spots based on lanes using a local dynamic map (LDM).
At least one processor; And
A memory storing instructions instructing the at least one processor to perform at least one step,
The at least one step,
Acquiring a lane centerline for the lane located near the current vehicle using the LDM;
Obtaining a three-dimensional coordinate value for a point where the blind spot formed by the first vehicle and the center line of the lane overlap;
Generating a blind spot display section by converting the obtained three-dimensional coordinate values into a reference coordinate system for the current vehicle; And
Apparatus for providing information on the blind spot on a lane basis using the LDM, comprising the step of displaying the generated blind spot display section to the user. In claim 6,
Prior to acquiring the centerline with the car,
Detecting lane center points for the two lanes forming the lane;
Detecting lane center points by averaging lane center points detected in each of the two lanes with each other;
Combining the detected lane center points to detect the lane center line; And
And updating the LDM using the detected lane centerline. 11. The apparatus of claim 1, further comprising updating the LDM using a lane line. In claim 6,
After acquiring a centerline with the car,
Obtaining a three-dimensional coordinate value for a point to be sensed using the three-dimensional coordinate value for the center line with the obtained difference;
Converting the obtained three-dimensional coordinate value into a reference coordinate system for the current vehicle;
Acquiring sensing data by detecting a point corresponding to the converted three-dimensional coordinate value using a sensor; and
The method for providing information on the blind spot on a lane basis using the LDM, further comprising the step of verifying the obtained sensing data. In claim 6,
After generating the blind spot display interval,
Obtaining a three-dimensional coordinate value for the blind spot display section; And
And transmitting the obtained three-dimensional coordinate values to the first vehicle. The apparatus for providing blind spot information on a lane basis using LDM. In claim 9,
After the step of transmitting the obtained three-dimensional coordinate value to the first vehicle,
Obtaining a position coordinate value of a second vehicle positioned in the blind spot display section from the first vehicle;
Converting the acquired position coordinate value into a reference coordinate system for the current vehicle; And
And displaying the second vehicle to the user by using the converted position coordinate value. The apparatus for providing the blind spot based on the lane using LDM.

Images