CN109724615B - Method and system for verifying lane line identification result - Google Patents
Method and system for verifying lane line identification result Download PDFInfo
- Publication number
- CN109724615B CN109724615B CN201910149900.5A CN201910149900A CN109724615B CN 109724615 B CN109724615 B CN 109724615B CN 201910149900 A CN201910149900 A CN 201910149900A CN 109724615 B CN109724615 B CN 109724615B
- Authority
- CN
- China
- Prior art keywords
- lane line
- vehicle
- precision map
- camera
- recognition result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The invention discloses a method and a system for checking lane line recognition results, which are characterized in that after lane line information of the position of a vehicle in a high-precision map is acquired, the lane line information is subjected to coordinate conversion to obtain the position of a lane line point in the high-precision map listed under a vehicle coordinate system, and then the accuracy of the lane line recognition results output by a camera of the vehicle is judged by utilizing the position of the lane line point in the high-precision map listed under the vehicle coordinate system. According to the scheme for verifying the lane line recognition result output by the camera based on the high-precision map, the lane line recognition result can be verified on line, a laser radar does not need to be installed in a vehicle, and only the high-precision map needs to be downloaded in the vehicle, so that the cost is low, and the method can be applied to mass production vehicles.
Description
Technical Field
The invention relates to the technical field of intelligent driving positioning, in particular to a method and a system for verifying lane line identification results.
Background
The existing vision processing chip, related algorithm and the like can identify lane line information. However, most of the current methods rely on manual verification or laser radar verification to verify whether the lane line identification result is correct or not.
Whether the Lane line identification result is correct or not is manually verified, so that on one hand, the efficiency is low, time and labor are wasted, on the other hand, functions such as Lane Keeping (LKA) and the like realized according to the Lane line identification result cannot be fed back to an automobile system in time even if the Lane line identification result is found to be wrong manually, and the Lane line identification result cannot be verified on line. Although the lane line identification result can be checked on line by using the laser radar, the laser radar needs to be installed in the vehicle, and the laser radar cannot be applied to mass production vehicles due to high price.
Disclosure of Invention
In view of the above, the invention discloses a method and a system for checking lane line identification results, so as to realize on-line checking of lane line identification results, and a laser radar does not need to be installed in a vehicle, and only a high-precision map needs to be downloaded in the vehicle, so that the price is low, and the method and the system can be applied to mass production vehicles.
A method for verifying lane line identification results comprises the following steps:
acquiring lane line information of the position of the vehicle in the high-precision map;
performing coordinate conversion on the lane line information based on high-precision positioning to obtain the positions of lane line points in a high-precision map under a vehicle coordinate system;
and comparing the position of the lane line point array in the high-precision map under the vehicle coordinate system with the lane line recognition result output by the camera of the vehicle by taking the position of the lane line point array in the high-precision map as a reference, and judging the correctness of the lane line recognition result by the camera of the vehicle.
Optionally, the coordinate conversion is performed on the lane line information based on the high-precision positioning to obtain the position of the lane line point row in the high-precision map in the vehicle coordinate system, and the method specifically includes:
wherein (A)ix，Aiy) For the ith point A in the lane line point column in the high-precision mapiCoordinates in the vehicle coordinate System, (A)ilat，Ailon) The longitude and the latitude of the ith point in the lane line point column in the high-precision map are (O)lat，Olon) Is the latitude and longitude, theta, of the vehicleoIs the included angle between the heading and the longitude of the vehicle.
Optionally, the comparing, with the lane line recognition result output by the vehicle camera, the position of the lane line point in the high-precision map listed in the vehicle coordinate system as a reference to determine the correctness of the lane line recognition result by the vehicle camera specifically includes:
calculating the sum of squares of residual errors of a lane line point column in the high-precision map and a lane line recognition result output by the camera of the vehicle, wherein the lane line recognition result is a lane line curve;
judging whether the sum of the squares of the residual errors is smaller than a preset threshold value or not;
if so, judging that the lane line is correctly identified by the camera of the vehicle.
Optionally, the comparing, with the lane line recognition result output by the vehicle camera, the position of the lane line point in the high-precision map listed in the vehicle coordinate system as a reference to determine the correctness of the lane line recognition result by the vehicle camera specifically includes:
performing curve fitting on the lane line point column in the high-precision map to obtain a lane line equation in the high-precision map, which is as follows:
y＝M0+M1x+M2x2；
in the formula, M0For left and right distances of the vehicle from the lane line, M, obtained using a high-precision map1For obtaining the angle between the vehicle and the lane line using a high-precision map, M2The curvature of the lane line is obtained by using a high-precision map, wherein x is a coordinate along the longitudinal direction of the vehicle under a vehicle coordinate system, and y is a coordinate along the transverse direction of the vehicle under the vehicle coordinate system;
acquiring a lane line equation corresponding to a lane line recognition result output by the camera of the vehicle, wherein the lane line equation comprises the following steps:
y＝C0+C1x+C2x2；
in the formula, C0Left and right distance of vehicle from lane line, C, output for camera1Angle between vehicle and lane line output by camera, C2A curvature representing a lane line output for the camera;
and when the following inequality is satisfied, judging that the lane line is correctly identified by the camera of the vehicle, wherein the inequality is as follows:
in the formula, P0Allowing a threshold value for the left-right distance difference, P1Is the threshold value of the included angle between the vehicle and the lane line, P2Is the curvature threshold.
A system for verifying lane line recognition results, comprising:
the acquisition unit is used for acquiring lane line information of the position of the vehicle in the high-precision map;
the coordinate conversion unit is used for carrying out coordinate conversion on the lane line information based on high-precision positioning to obtain the positions of lane line points in the high-precision map under a vehicle coordinate system;
and the checking unit is used for comparing the position of the lane line point array in the high-precision map under the vehicle coordinate system as a reference with the lane line recognition result output by the camera of the vehicle and judging the correctness of the lane line recognition result by the camera of the vehicle.
Optionally, the coordinate transformation unit is specifically configured to:
wherein (A)ix，Aiy) For the ith point A in the lane line point column in the high-precision mapiCoordinates in the vehicle coordinate System, (A)ilat，Ailon) The longitude and the latitude of the ith point in the lane line point column in the high-precision map are (O)lat，Olon) Is the latitude and longitude, theta, of the vehicleoIs the included angle between the heading and the longitude of the vehicle.
Optionally, the verification unit is specifically configured to:
calculating the sum of squares of residual errors of a lane line point column in the high-precision map and a lane line recognition result output by the camera of the vehicle, wherein the lane line recognition result is a lane line curve;
judging whether the sum of the squares of the residual errors is smaller than a preset threshold value or not;
if so, judging that the lane line is correctly identified by the camera of the vehicle.
Optionally, the verification unit is specifically configured to:
performing curve fitting on the lane line point column in the high-precision map to obtain a lane line equation in the high-precision map, which is as follows:
y＝M0+M1x+M2x2；
in the formula, M0For left and right distances of the vehicle from the lane line, M, obtained using a high-precision map1For obtaining the angle between the vehicle and the lane line using a high-precision map, M2The curvature of the lane line is obtained by using a high-precision map, wherein x is a coordinate along the longitudinal direction of the vehicle under a vehicle coordinate system, and y is a coordinate along the transverse direction of the vehicle under the vehicle coordinate system;
acquiring a lane line equation corresponding to a lane line recognition result output by the camera of the vehicle, wherein the lane line equation comprises the following steps:
y＝C0+C1x+C2x2；
in the formula, C0Left and right distance of vehicle from lane line, C, output for camera1Angle between vehicle and lane line output by camera, C2A curvature representing a lane line output for the camera;
and when the following inequality is satisfied, judging that the lane line is correctly identified by the camera of the vehicle, wherein the inequality is as follows:
in the formula, P0Allowing a threshold value for the left-right distance difference, P1Is the threshold value of the included angle between the vehicle and the lane line, P2Is the curvature threshold.
According to the technical scheme, after the lane line information of the position of the vehicle in the high-precision map is obtained, the lane line information is subjected to coordinate conversion to obtain the position of the lane line points in the high-precision map under the vehicle coordinate system, and then the correctness of the lane line recognition result output by the camera of the vehicle is judged by utilizing the position of the lane line points in the high-precision map under the vehicle coordinate system. According to the scheme for verifying the lane line recognition result output by the camera based on the high-precision map, the lane line recognition result can be verified on line, a laser radar does not need to be installed in a vehicle, and only the high-precision map needs to be downloaded in the vehicle, so that the cost is low, and the method can be applied to mass production vehicles.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the disclosed drawings without creative efforts.
Fig. 1 is a flowchart of a method for verifying a lane line identification result according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a position of a lane line point row in a high-precision map in a vehicle coordinate system, which is obtained by performing coordinate transformation on lane line information based on high-precision positioning according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a position relationship between a lane line dot row in a high-precision map and a lane line recognition result output by a camera of a vehicle according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a system for verifying a lane line identification result according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a method and a system for checking lane line recognition results, which are characterized in that after lane line information of the position of a vehicle in a high-precision map is obtained, the lane line information is subjected to coordinate conversion to obtain the position of lane line points in the high-precision map under a vehicle coordinate system, and then the correctness of the lane line recognition results output by a camera of the vehicle is judged by utilizing the position of the lane line points in the high-precision map under the vehicle coordinate system. According to the scheme for verifying the lane line recognition result output by the camera based on the high-precision map, the lane line recognition result can be verified on line, a laser radar does not need to be installed in a vehicle, and only the high-precision map needs to be downloaded in the vehicle, so that the cost is low, and the method can be applied to mass production vehicles.
Referring to fig. 1, a flowchart of a method for verifying a lane line identification result according to an embodiment of the present invention includes:
s101, obtaining lane line information of the position of the vehicle in the high-precision map;
it should be noted that the method for verifying the lane line identification result provided by the present invention relies on high-precision positioning, and generally, high-precision positioning refers to centimeter-level positioning precision.
The lane line information refers to a position of a lane line in a vehicle coordinate system, and in practical application, the position can be represented by a lane line equation, which can be referred to an existing mature scheme specifically, and is not described herein again.
Specifically, a high-precision positioning system is adopted to perform high-precision positioning on the vehicle to obtain a high-precision positioning result, and a combined navigation system usually provided with an RTK (Real-time kinematic) can achieve centimeter-level positioning precision. The high-precision Positioning System used in this embodiment may be an RTK + GPS (Global Positioning System) System or a combined navigation System, such as a GPS + IMU (Inertial Measurement Unit) combined navigation System.
And inquiring the lane line information of the position of the vehicle according to the high-precision positioning result and a pre-stored high-precision map.
Step S102, carrying out coordinate conversion on the lane line information based on high-precision positioning to obtain the positions of lane line points in a high-precision map under a vehicle coordinate system;
specifically, referring to fig. 2, in an embodiment of the present invention, coordinate conversion is performed on lane line information based on high-precision positioning to obtain a schematic diagram of positions of lane line points in a high-precision map under a vehicle coordinate system, where 11 in fig. 2 indicates a host vehicle, and a positioning position of the host vehicle by a high-precision positioning system is: (O)lat，Olon)，OlatIs the latitude of the vehicle, OlonIs the longitude of the host vehicle; the dot columns of lane lines in the high-precision map are marked as A1，A2，...，AnWherein, any AiThe positions of the points are: (A)ilat，Ailon) (ii) a Lane line point aiPosition in the vehicle coordinate System (A)ix，Aiy) The following were used:
in the formula, thetaoIs the included angle between the heading and the longitude of the vehicle (A)ilat，Ailon) Is the ith point A in the lane line point column in the high-precision mapiLongitude and latitude of (A)ix，Aiy) Is the ith point A in the lane line point column in the high-precision mapiCoordinates in the vehicle coordinate system.
And step S103, comparing the position of the lane line point array in the high-precision map under the vehicle coordinate system as a reference with the lane line recognition result output by the camera of the vehicle, and judging the correctness of the lane line recognition result by the camera of the vehicle.
In summary, the method for verifying lane line recognition results disclosed by the present invention performs coordinate transformation on lane line information after obtaining lane line information of a position of a host vehicle in a high-precision map, to obtain a position of a lane line point array in the high-precision map under a vehicle coordinate system, and then determines the correctness of the lane line recognition results output by a camera of the host vehicle by using the position of the lane line point array in the high-precision map under the vehicle coordinate system. According to the scheme for verifying the lane line recognition result output by the camera based on the high-precision map, the lane line recognition result can be verified on line, a laser radar does not need to be installed in a vehicle, and only the high-precision map needs to be downloaded in the vehicle, so that the cost is low, and the method can be applied to mass production vehicles.
In practical application, in step S103, the position of the lane line point in the high-precision map listed under the vehicle coordinate system is used as a reference, and compared with the lane line recognition result output by the vehicle camera, to determine the correctness of the lane line recognition result by the vehicle camera, and the specific implementation methods may be as follows:
the method one, the position relation of the lane line point row in the high-precision map and the lane line recognition result output by the camera of the vehicle is directly compared, see fig. 3, the lane line recognition result is specifically a lane line curve, the lane line point row in the high-precision map refers to each point which is positioned on the lane line curve or near the lane line curve in fig. 3, the position relation of the lane line point row and the lane line curve in the high-precision map can be quantized by using a method of solving the square sum of residual errors, when the square sum of the residual errors of the lane line point row and the lane line curve in the high-precision map is smaller than a preset threshold value, the lane line point row is considered to be positioned near the lane line recognized by the camera of the vehicle, and at the moment, the camera of the vehicle is judged to correctly recognize the lane line; on the contrary, when the sum of squares of residuals of the lane line point rows and the lane line curves in the high-precision map is not less than the preset threshold, the lane line point rows are considered to be located at a position far from the lane line identified by the camera of the vehicle, and at the moment, it is determined that the camera of the vehicle identifies the lane line incorrectly.
Therefore, step S103 may specifically include:
calculating the sum of squares of residual errors of lane line point rows in the high-precision map and lane line recognition results output by the camera of the vehicle;
judging whether the sum of the squares of the residual errors is smaller than a preset threshold value or not;
if so, judging that the lane line is correctly identified by the camera of the vehicle.
And secondly, judging the correctness of the lane line recognition result of the vehicle camera by comparing the curve equation with the lane line recognition result output by the vehicle camera.
In practical application, there are many implementation ways for the specific process of obtaining the curve equation by using the curve fitting method for the lane line point column in the high-precision map, for example, the curve fitting is performed on the lane line point column in the high-precision map by using the least square method to obtain the curve equation. The present invention is not limited herein, particularly depending on the actual needs.
In this embodiment, when a curve equation obtained by curve fitting the lane line point rows in the high-precision map is compared with the lane line recognition result output by the vehicle camera, the lane line recognition result is specifically a lane line equation.
The application range of the lane line equation corresponding to the lane line recognition result output by the camera is as follows: (x)start,xend) And x is the coordinate along the longitudinal direction of the vehicle in the vehicle coordinate system, from (x)start,xend) Selecting the lane line point column in the high-precision map within the range, and performing step S102 to obtain (x)start,xend) Into a position in the vehicle coordinate system.
How to judge the correctness of the lane line recognition result by the camera of the vehicle by using a curve fitting method is described in detail below.
That is, step S103 may specifically include:
(1) performing curve fitting on the lane line point columns in the high-precision map to obtain a lane line equation in the high-precision map, which is as follows:
y＝M0+M1x+M2x2 (1)；
in the formula, M0For left and right distances of the vehicle from the lane line, M, obtained using a high-precision map1For obtaining the angle between the vehicle and the lane line using a high-precision map, M2In order to obtain the curvature of the lane line using the high-precision map, x is a coordinate in the vehicle longitudinal direction in the vehicle coordinate system, and y is a coordinate in the vehicle lateral direction in the vehicle coordinate system, specifically, see x and y shown on the rightmost side of fig. 3.
(2) Acquiring a lane line equation corresponding to a lane line recognition result output by the camera of the vehicle, wherein the lane line equation comprises the following steps:
y＝C0+C1x+C2x2 (2)；
in the formula, C0Left and right distance of vehicle from lane line, C, output for camera1Angle between vehicle and lane line output by camera, C2The curvature of the lane line is represented by the curvature output by the camera, x is the coordinate in the longitudinal direction of the vehicle in the vehicle coordinate system, and y is the coordinate in the lateral direction of the vehicle in the vehicle coordinate system, specifically see x and y shown on the rightmost side of fig. 3.
(3) When the formula (1) and the formula (2) satisfy the inequality shown in the formula (3), it is determined that the lane line identification by the vehicle camera is correct, or the error of the lane line equation identified by the camera is within the allowable range, and the formula (3) is as follows:
in the formula, P0Allowing a threshold value for the left-right distance difference, P1Is the threshold value of the included angle between the vehicle and the lane line, P2Is the curvature threshold.
In summary, the method for verifying lane line recognition results disclosed by the present invention performs coordinate transformation on lane line information after obtaining lane line information of a position of a host vehicle in a high-precision map, to obtain a position of a lane line point array in the high-precision map under a vehicle coordinate system, and then determines the correctness of the lane line recognition results output by a camera of the host vehicle by using the position of the lane line point array in the high-precision map under the vehicle coordinate system. According to the scheme for verifying the lane line recognition result output by the camera based on the high-precision map, the lane line recognition result can be verified on line, a laser radar does not need to be installed in a vehicle, and only the high-precision map needs to be downloaded in the vehicle, so that the cost is low, and the method can be applied to mass production vehicles.
Corresponding to the embodiment of the method, the invention also discloses a system for verifying the lane line identification result.
Referring to fig. 4, a schematic structural diagram of a system for verifying lane line identification results disclosed in an embodiment of the present invention includes:
an obtaining unit 201, configured to obtain lane line information of a position of a host vehicle in a high-precision map;
it should be noted that the method for verifying the lane line identification result provided by the present invention relies on high-precision positioning, and generally, high-precision positioning refers to centimeter-level positioning precision.
The lane line information refers to a position of a lane line in a vehicle coordinate system, and in practical application, the position can be represented by a lane line equation, which can be referred to an existing mature scheme specifically, and is not described herein again.
Specifically, a high-precision positioning system is adopted to perform high-precision positioning on the vehicle to obtain a high-precision positioning result, and a combined navigation system usually provided with an RTK (Real-time kinematic) can achieve centimeter-level positioning precision. The high-precision Positioning System used in this embodiment may be an RTK + GPS (Global Positioning System) System or a combined navigation System, such as a GPS + IMU (Inertial Measurement Unit) combined navigation System.
And inquiring the lane line information of the position of the vehicle according to the high-precision positioning result and a pre-stored high-precision map.
The coordinate conversion unit 202 is used for performing coordinate conversion on the lane line information based on high-precision positioning to obtain the positions of lane line points in the high-precision map under a vehicle coordinate system;
specifically, referring to fig. 2, in an embodiment of the present invention, coordinate conversion is performed on lane line information based on high-precision positioning to obtain a schematic diagram of positions of lane line points in a high-precision map under a vehicle coordinate system, where 11 in fig. 2 indicates a host vehicle, and a positioning position of the host vehicle by a high-precision positioning system is: (O)lat，Olon)，OlatIs the latitude of the vehicle, OlonIs the longitude of the host vehicle; the dot columns of lane lines in the high-precision map are marked as A1，A2，...，AnWherein, any AiThe positions of the points are: (A)ilat，Ailon) (ii) a Lane line point aiPosition in the vehicle coordinate System (A)ix，Aiy) The following were used:
in the formula, thetaoIs the included angle between the heading and the longitude of the vehicle (A)ilat，Ailon) Is the ith point A in the lane line point column in the high-precision mapiLongitude and latitude of (A)ix，Aiy) Is the ith point A in the lane line point column in the high-precision mapiCoordinates in the vehicle coordinate system.
That is, the coordinate conversion unit 202 obtains the position of the lane line point row in the high-precision map in the vehicle coordinate system according to the above formula.
And the checking unit 203 is used for comparing the position of the lane line point array in the high-precision map under the vehicle coordinate system as a reference with the lane line recognition result output by the vehicle camera, and judging the correctness of the lane line recognition result by the vehicle camera.
In summary, the system for checking lane line recognition results disclosed by the invention performs coordinate conversion on lane line information after acquiring the lane line information of the position of the vehicle in the high-precision map, so as to obtain the position of the lane line points in the high-precision map under the vehicle coordinate system, and then determines the correctness of the lane line recognition results output by the camera of the vehicle by using the position of the lane line points in the high-precision map under the vehicle coordinate system. According to the scheme for verifying the lane line recognition result output by the camera based on the high-precision map, the lane line recognition result can be verified on line, a laser radar does not need to be installed in a vehicle, and only the high-precision map needs to be downloaded in the vehicle, so that the cost is low, and the method can be applied to mass production vehicles.
In practical application, the verification unit 203 compares the position of the lane line point in the high-precision map listed under the vehicle coordinate system with the lane line recognition result output by the vehicle camera as a reference, and determines the correctness of the lane line recognition result by the vehicle camera, and the specific implementation methods may be the following two methods:
the method one, the position relation of the lane line point row in the high-precision map and the lane line recognition result output by the camera of the vehicle is directly compared, see fig. 3, the lane line recognition result is specifically a lane line curve, the lane line point row in the high-precision map refers to each point which is positioned on the lane line curve or near the lane line curve in fig. 3, the position relation of the lane line point row and the lane line curve in the high-precision map can be quantized by using a method of solving the square sum of residual errors, when the square sum of the residual errors of the lane line point row and the lane line curve in the high-precision map is smaller than a preset threshold value, the lane line point row is considered to be positioned near the lane line recognized by the camera of the vehicle, and at the moment, the camera of the vehicle is judged to correctly recognize the lane line; on the contrary, when the sum of squares of residuals of the lane line point rows and the lane line curves in the high-precision map is not less than the preset threshold, the lane line point rows are considered to be located at a position far from the lane line identified by the camera of the vehicle, and at the moment, it is determined that the camera of the vehicle identifies the lane line incorrectly.
Therefore, the verification unit 203 may be specifically configured to:
calculating the residual square sum of a lane line point column in the high-precision map and a lane line recognition result output by the camera of the vehicle, wherein the lane line recognition result is a lane line curve;
judging whether the sum of the squares of the residual errors is smaller than a preset threshold value or not;
if so, judging that the lane line is correctly identified by the camera of the vehicle.
And secondly, judging the correctness of the lane line recognition result of the vehicle camera by comparing the curve equation with the lane line recognition result output by the vehicle camera.
In practical application, there are many implementation ways for the specific process of obtaining the curve equation by using the curve fitting method for the lane line point column in the high-precision map, for example, the curve fitting is performed on the lane line point column in the high-precision map by using the least square method to obtain the curve equation. The present invention is not limited herein, particularly depending on the actual needs.
In this embodiment, when a curve equation obtained by curve fitting the lane line point rows in the high-precision map is compared with the lane line recognition result output by the vehicle camera, the lane line recognition result is specifically a lane line equation.
Therefore, the verification unit 203 may be specifically configured to:
performing curve fitting on the lane line point columns in the high-precision map to obtain a lane line equation in the high-precision map, which is as follows:
y＝M0+M1x+M2x2；
in the formula, M0For left and right distances of the vehicle from the lane line, M, obtained using a high-precision map1For obtaining the angle between the vehicle and the lane line using a high-precision map, M2For the curvature of the lane line obtained using the high-precision map, x is the lower edge of the vehicle coordinate systemThe longitudinal coordinate of the vehicle, and y is the transverse coordinate of the vehicle in the vehicle coordinate system;
acquiring a lane line equation corresponding to a lane line recognition result output by the camera of the vehicle, wherein the lane line equation comprises the following steps:
y＝C0+C1x+C2x2；
in the formula, C0Left and right distance of vehicle from lane line, C, output for camera1Angle between vehicle and lane line output by camera, C2A curvature representing a lane line output for the camera;
when the following inequality is satisfied, the lane line is correctly identified by the camera of the vehicle, and the inequality is as follows:
in the formula, P0Allowing a threshold value for the left-right distance difference, P1Is the threshold value of the included angle between the vehicle and the lane line, P2Is the curvature threshold.
In summary, the system for checking lane line recognition results disclosed by the invention performs coordinate conversion on lane line information after acquiring the lane line information of the position of the vehicle in the high-precision map, so as to obtain the position of the lane line points in the high-precision map under the vehicle coordinate system, and then determines the correctness of the lane line recognition results output by the camera of the vehicle by using the position of the lane line points in the high-precision map under the vehicle coordinate system. According to the scheme for verifying the lane line recognition result output by the camera based on the high-precision map, the lane line recognition result can be verified on line, a laser radar does not need to be installed in a vehicle, and only the high-precision map needs to be downloaded in the vehicle, so that the cost is low, and the method can be applied to mass production vehicles.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (4)
1. A method for verifying lane line recognition results is characterized by comprising the following steps:
acquiring lane line information of the position of the vehicle in the high-precision map;
performing coordinate conversion on the lane line information based on high-precision positioning to obtain the positions of lane line points in a high-precision map under a vehicle coordinate system;
taking the position of the lane line points in the high-precision map listed under the vehicle coordinate system as a reference, comparing the position with a lane line recognition result output by the camera of the vehicle, and judging the correctness of the lane line recognition result by the camera of the vehicle, wherein the lane line points in the high-precision map are listed as all points on a lane line curve or near the lane line curve;
the step of comparing the position of the lane line point in the high-precision map in the vehicle coordinate system with the lane line recognition result output by the vehicle camera to judge the correctness of the lane line recognition result by the vehicle camera includes:
calculating the sum of squares of residual errors of a lane line point column in the high-precision map and a lane line recognition result output by the camera of the vehicle, wherein the lane line recognition result is a lane line curve;
judging whether the sum of the squares of the residual errors is smaller than a preset threshold value or not;
if so, judging that the lane line is correctly identified by the camera of the vehicle;
alternatively, the first and second electrodes may be,
the step of comparing the position of the lane line point in the high-precision map in the vehicle coordinate system with the lane line recognition result output by the vehicle camera to judge the correctness of the lane line recognition result by the vehicle camera includes:
performing curve fitting on the lane line point column in the high-precision map to obtain a lane line equation in the high-precision map, which is as follows:
y＝M0+M1x+M2x2；
in the formula, M0For left and right distances of the vehicle from the lane line, M, obtained using high-precision maps1For obtaining the angle between the vehicle and the lane line using a high-precision map, M2In order to obtain the curvature of the lane line by using a high-precision map, x is a coordinate along the longitudinal direction of the vehicle under a vehicle coordinate system, and y is a coordinate along the transverse direction of the vehicle under the vehicle coordinate system;
acquiring a lane line equation corresponding to a lane line recognition result output by the camera of the vehicle, wherein the lane line equation comprises the following steps:
y＝C0+C1x+C2x2；
in the formula, C0Left and right distance of vehicle from lane line, C, output for camera1Angle between vehicle and lane line output by camera, C2A curvature representing a lane line output for the camera;
and when the following inequality is satisfied, judging that the lane line is correctly identified by the camera of the vehicle, wherein the inequality is as follows:
in the formula, P0Allowing a threshold value for the left-right distance difference, P1Is the threshold value of the included angle between the vehicle and the lane line, P2Is the curvature threshold.
2. The verification method according to claim 1, wherein the coordinate conversion of the lane line information based on the high-precision positioning to obtain the position of the lane line point row in the high-precision map in the vehicle coordinate system specifically comprises:
wherein (A)ix，Aiy) For the ith point A in the lane line point column in the high-precision mapiCoordinates in the vehicle coordinate System, (A)ilat，Ailon) The longitude and the latitude of the ith point in the lane line point column in the high-precision map are (O)lat，Olon) Is the latitude and longitude, theta, of the vehicleoIs the included angle between the heading and the longitude of the vehicle.
3. A system for verifying lane line recognition results, comprising:
the acquisition unit is used for acquiring lane line information of the position of the vehicle in the high-precision map;
the coordinate conversion unit is used for carrying out coordinate conversion on the lane line information based on high-precision positioning to obtain the positions of lane line points in the high-precision map under a vehicle coordinate system;
the checking unit is used for comparing the position of the lane line points in the high-precision map under the vehicle coordinate system as a reference with the lane line recognition result output by the camera of the vehicle, and judging the correctness of the lane line recognition result by the camera of the vehicle, wherein the lane line points in the high-precision map are points on or near a lane line curve;
wherein the verification unit is specifically configured to:
calculating the sum of squares of residual errors of a lane line point column in the high-precision map and a lane line recognition result output by the camera of the vehicle, wherein the lane line recognition result is a lane line curve;
judging whether the sum of the squares of the residual errors is smaller than a preset threshold value or not;
if so, judging that the lane line is correctly identified by the camera of the vehicle;
alternatively, the first and second electrodes may be,
the verification unit is specifically configured to:
performing curve fitting on the lane line point column in the high-precision map to obtain a lane line equation in the high-precision map, which is as follows:
y＝M0+M1x+M2x2；
in the formula, M0For left and right distances of the vehicle from the lane line, M, obtained using high-precision maps1For obtaining the angle between the vehicle and the lane line using a high-precision map, M2In order to obtain the curvature of the lane line by using a high-precision map, x is a coordinate along the longitudinal direction of the vehicle under a vehicle coordinate system, and y is a coordinate along the transverse direction of the vehicle under the vehicle coordinate system;
acquiring a lane line equation corresponding to a lane line recognition result output by the camera of the vehicle, wherein the lane line equation comprises the following steps:
y＝C0+C1x+C2x2；
in the formula, C0Left and right distance of vehicle from lane line, C, output for camera1Angle between vehicle and lane line output by camera, C2A curvature representing a lane line output for the camera;
and when the following inequality is satisfied, judging that the lane line is correctly identified by the camera of the vehicle, wherein the inequality is as follows:
in the formula, P0Allowing a threshold value for the left-right distance difference, P1Is the threshold value of the included angle between the vehicle and the lane line, P2Is the curvature threshold.
4. The verification system of claim 3, wherein the coordinate transformation unit is specifically configured to:
wherein (A)ix，Aiy) For the ith point A in the lane line point column in the high-precision mapiCoordinates in the vehicle coordinate System, (A)ilat，Ailon) The longitude and the latitude of the ith point in the lane line point column in the high-precision map are (O)lat，Olon) Is the latitude and longitude, theta, of the vehicleoIs the included angle between the heading and the longitude of the vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910149900.5A CN109724615B (en) | 2019-02-28 | 2019-02-28 | Method and system for verifying lane line identification result |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910149900.5A CN109724615B (en) | 2019-02-28 | 2019-02-28 | Method and system for verifying lane line identification result |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109724615A CN109724615A (en) | 2019-05-07 |
| CN109724615B true CN109724615B (en) | 2021-06-29 |
Family
ID=66300138
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910149900.5A Active CN109724615B (en) | 2019-02-28 | 2019-02-28 | Method and system for verifying lane line identification result |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109724615B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112172810A (en) * | 2019-06-18 | 2021-01-05 | 广州汽车集团股份有限公司 | Lane keeping device, method and system and automobile |
| CN112419404A (en) * | 2019-08-21 | 2021-02-26 | 北京初速度科技有限公司 | Map data acquisition method and device |
| CN110645996B (en) * | 2019-09-17 | 2021-07-16 | 武汉中海庭数据技术有限公司 | Method and system for extracting perception data |
| CN110595494B (en) * | 2019-09-17 | 2021-06-22 | 百度在线网络技术（北京）有限公司 | Map error determination method and device |
| CN110906953A (en) * | 2019-11-26 | 2020-03-24 | 武汉中海庭数据技术有限公司 | Relative position precision evaluation method and device for automatic driving positioning |
| CN111141311B (en) * | 2019-12-31 | 2022-04-08 | 武汉中海庭数据技术有限公司 | Evaluation method and system of high-precision map positioning module |
| CN111121849B (en) * | 2020-01-02 | 2021-08-20 | 大陆投资(中国)有限公司 | Automatic calibration method for orientation parameters of sensor, edge calculation unit and roadside sensing system |
| WO2021208110A1 (en) * | 2020-04-18 | 2021-10-21 | 华为技术有限公司 | Method for determining lane line recognition abnormal event, and lane line recognition apparatus and system |
| CN111516673B (en) * | 2020-04-30 | 2022-08-09 | 重庆长安汽车股份有限公司 | Lane line fusion system and method based on intelligent camera and high-precision map positioning |
| CN111998860B (en) * | 2020-08-21 | 2023-02-17 | 阿波罗智能技术(北京)有限公司 | Automatic driving positioning data verification method and device, electronic equipment and storage medium |
| CN112560680A (en) * | 2020-12-16 | 2021-03-26 | 北京百度网讯科技有限公司 | Lane line processing method and device, electronic device and storage medium |
| CN112966059B (en) * | 2021-03-02 | 2023-11-24 | 北京百度网讯科技有限公司 | Data processing method and device for positioning data, electronic equipment and medium |
| CN113175937B (en) * | 2021-06-29 | 2021-09-28 | 天津天瞳威势电子科技有限公司 | Method and device for evaluating lane line sensing result |
| CN113587940A (en) * | 2021-07-30 | 2021-11-02 | 重庆长安汽车股份有限公司 | Lane line checking method and system based on vehicle turning radius and vehicle |
| CN113569800A (en) * | 2021-08-09 | 2021-10-29 | 北京地平线机器人技术研发有限公司 | Lane recognition and verification method and device, readable storage medium and electronic equipment |
| CN114323033B (en) * | 2021-12-29 | 2023-08-29 | 北京百度网讯科技有限公司 | Positioning method and equipment based on lane lines and feature points and automatic driving vehicle |
| CN114252082B (en) * | 2022-03-01 | 2022-05-17 | 苏州挚途科技有限公司 | Vehicle positioning method and device and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102013201941A1 (en) * | 2013-02-06 | 2014-08-07 | Bayerische Motoren Werke Aktiengesellschaft | Method for determining lane course for vehicle, involves determining resultant respective path includes traffic lane boundary depending on predetermined allocation map which represents predetermined area around vehicle |
| CN105260699A (en) * | 2015-09-10 | 2016-01-20 | 百度在线网络技术（北京）有限公司 | Lane line data processing method and lane line data processing device |
| CN107782321A (en) * | 2017-10-10 | 2018-03-09 | 武汉迈普时空导航科技有限公司 | A kind of view-based access control model and the Combinated navigation method of high-precision map lane line constraint |
| CN109084782A (en) * | 2017-06-13 | 2018-12-25 | 蔚来汽车有限公司 | Lane line map constructing method and building system based on camera sensing device |
| CN109186615A (en) * | 2018-09-03 | 2019-01-11 | 武汉中海庭数据技术有限公司 | Lane side linear distance detection method, device and storage medium based on high-precision map |
| CN109297499A (en) * | 2018-08-20 | 2019-02-01 | 武汉中海庭数据技术有限公司 | Lane model building method, device and computer can storage mediums |
-
2019
- 2019-02-28 CN CN201910149900.5A patent/CN109724615B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102013201941A1 (en) * | 2013-02-06 | 2014-08-07 | Bayerische Motoren Werke Aktiengesellschaft | Method for determining lane course for vehicle, involves determining resultant respective path includes traffic lane boundary depending on predetermined allocation map which represents predetermined area around vehicle |
| CN105260699A (en) * | 2015-09-10 | 2016-01-20 | 百度在线网络技术（北京）有限公司 | Lane line data processing method and lane line data processing device |
| CN109084782A (en) * | 2017-06-13 | 2018-12-25 | 蔚来汽车有限公司 | Lane line map constructing method and building system based on camera sensing device |
| CN107782321A (en) * | 2017-10-10 | 2018-03-09 | 武汉迈普时空导航科技有限公司 | A kind of view-based access control model and the Combinated navigation method of high-precision map lane line constraint |
| CN109297499A (en) * | 2018-08-20 | 2019-02-01 | 武汉中海庭数据技术有限公司 | Lane model building method, device and computer can storage mediums |
| CN109186615A (en) * | 2018-09-03 | 2019-01-11 | 武汉中海庭数据技术有限公司 | Lane side linear distance detection method, device and storage medium based on high-precision map |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109724615A (en) | 2019-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109724615B (en) | Method and system for verifying lane line identification result | |
| CN113538919B (en) | Lane departure recognition method, device, equipment and storage medium | |
| CN110462343B (en) | Method and system for navigating a vehicle using automatically marked images | |
| US20210207977A1 (en) | Vehicle position estimation device, vehicle position estimation method, and computer-readable recording medium for storing computer program programmed to perform said method | |
| CN106918342B (en) | Method and system for positioning driving path of unmanned vehicle | |
| CN111696160B (en) | Automatic calibration method and equipment for vehicle-mounted camera and readable storage medium | |
| CN110954112B (en) | Method and device for updating matching relation between navigation map and perception image | |
| US20230143687A1 (en) | Method of estimating three-dimensional coordinate value for each pixel of two-dimensional image, and method of estimating autonomous driving information using the same | |
| US20210229280A1 (en) | Positioning method and device, path determination method and device, robot and storage medium | |
| CN109115231B (en) | Vehicle positioning method and device and automatic driving vehicle | |
| JP6454726B2 (en) | Own vehicle position estimation device | |
| CN111507130B (en) | Lane-level positioning method and system, computer equipment, vehicle and storage medium | |
| CN111507129B (en) | Lane-level positioning method and system, computer equipment, vehicle and storage medium | |
| JP6509361B2 (en) | Parking support device and parking support method | |
| CN110969055B (en) | Method, apparatus, device and computer readable storage medium for vehicle positioning | |
| JP6776707B2 (en) | Own vehicle position estimation device | |
| CN108573611B (en) | Speed limit sign fusion method and speed limit sign fusion system | |
| CN110555885B (en) | Calibration method and device of vehicle-mounted camera and terminal | |
| KR102331312B1 (en) | 3D vehicular navigation system using vehicular internal sensor, camera, and GNSS terminal | |
| CN103171560B (en) | Lane recognition device | |
| US20220187095A1 (en) | Landmark location estimation apparatus and method, and computer-readable recording medium storing computer program programmed to perform method | |
| CN110906953A (en) | Relative position precision evaluation method and device for automatic driving positioning | |
| CN114396957B (en) | Positioning pose calibration method based on vision and map lane line matching and automobile | |
| JP3288566B2 (en) | Travel lane recognition device | |
| CN103163543A (en) | Method of detecting location of opposing vehicle using GPS information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 4 / F, building 1, No.14 Jiuxianqiao Road, Chaoyang District, Beijing 100020 Applicant after: Beijing Jingwei Hengrun Technology Co., Ltd Address before: 8 / F, block B, No. 11, Anxiang Beili, Chaoyang District, Beijing 100101 Applicant before: Beijing Jingwei HiRain Technologies Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |