WO2023088486A1 - 车道线提取方法、装置、车辆及存储介质 - Google Patents
车道线提取方法、装置、车辆及存储介质 Download PDFInfo
- Publication number
- WO2023088486A1 WO2023088486A1 PCT/CN2022/133480 CN2022133480W WO2023088486A1 WO 2023088486 A1 WO2023088486 A1 WO 2023088486A1 CN 2022133480 W CN2022133480 W CN 2022133480W WO 2023088486 A1 WO2023088486 A1 WO 2023088486A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- lane line
- data
- length
- predicted
- current
- Prior art date
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 31
- 238000000034 method Methods 0.000 claims abstract description 30
- 239000000284 extract Substances 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 6
- 230000004044 response Effects 0.000 claims 4
- 238000002360 preparation method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 238000013500 data storage Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 239000004020 conductor Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
Definitions
- the embodiments of the present application relate to the technical field of lane line detection, for example, to a lane line extraction method, device, vehicle, and storage medium.
- the lane line recognition system based on machine vision mainly collects images of the road ahead through image sensors such as cameras installed in front of the vehicle, and then recognizes and extracts lane lines from the images.
- image sensors such as cameras installed in front of the vehicle
- the lane line image collected by the image sensor has a serious problem of incomplete lane lines, which seriously affects the lane line recognition. the accuracy.
- Embodiments of the present application provide a lane line extraction method, device, vehicle, and storage medium, so as to reduce constraints for lane line recognition and improve lane line recognition accuracy.
- An embodiment of the present application provides a lane line extraction method, including: acquiring the current location point of the vehicle and high-precision map data, the high-precision map data including lane line data and the planned path data of the vehicle; based on the current location points and the planned route data, extracting target lane line data within the set lane line length in front of the lane where the vehicle is located from the lane line data; performing curve fitting on the target lane line data to obtain a lane line.
- the embodiment of the present application also provides a lane line extraction device, the device includes: an acquisition module, configured to acquire the current location point of the vehicle and high-precision map data, the high-precision map data includes lane line data and the planned path of the vehicle data; an extraction module, configured to extract target lane line data within the set lane line length in front of the lane where the vehicle is located from the lane line data of the high-precision map data based on the current position point and the planned route data ; A fitting module, configured to perform curve fitting on the target lane line data to obtain the lane line.
- the embodiment of the present application also provides a vehicle, including a memory, a processor, and a computer program stored on the memory and operable on the processor.
- a vehicle including a memory, a processor, and a computer program stored on the memory and operable on the processor.
- the processor executes the program, any The lane line extraction method described above.
- the embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the lane line extraction method described in any one of the embodiments of the present application is implemented.
- Fig. 1 is a flowchart of a lane line extraction method in Embodiment 1 of the present application
- Fig. 2 is a flowchart of a lane line extraction method in Embodiment 2 of the present application
- Fig. 3 is a schematic structural diagram of a lane line extraction device in Embodiment 3 of the present application.
- Fig. 4 is a schematic structural diagram of a vehicle in Embodiment 4 of the present application.
- Figure 1 is a flow chart of a lane line extraction method provided in Embodiment 1 of the present application. This embodiment is applicable to the case of extracting lane lines based on high-precision maps.
- This method can be implemented by the lane line extraction device in the embodiment of the present application. For execution, the device may be implemented in software and/or hardware.
- the method includes the following steps.
- a high-precision map refers to a high-precision, finely defined map whose accuracy can reach the decimeter level and can distinguish multiple lanes.
- Refined definition means that high-precision maps use a formatted storage method to store various traffic elements in traffic scenes, that is, high-precision map data can include road network data, lane network data, lane line data, and traffic signs of traditional maps. data.
- the formatted storage method of the high-precision map data is: based on the changes in the straight length of the lane, diversion, merging, curvature, etc., the lane line data is stored in the high-precision map as a layer , and assign lane identification numbers to lanes in each layer.
- the 0-500m and 500m-1000m lane data of a straight lane with a length of 1000m are stored in lane layers with lane identification numbers R01 and R02 respectively; the length of the right side of the R02 lane is 400
- the right-turn lane of m is stored in the lane layer with the lane identification number R03
- the left-turn lane with a length of 300 meters on the left side of the R02 lane is stored in the lane layer with the lane identification number R04.
- the lane line layer is also stored in the HD map, which is used to store the lane line data on both sides of the lane.
- the high-precision map data includes lane line data and vehicle planning path data.
- the lane line data is a sequence of boundary points on the boundary line of the lane
- the planning path data is a sequence of vehicle planning track points.
- the current position of the vehicle is the two-dimensional position coordinates of the vehicle in the HD map. Any method in the related art may be used to determine the vehicle's current location point and planned route, and this application does not set any limitation thereto.
- the target lane line data is a sequence composed of boundary points on the boundary line of the lane within the set lane line length extracted from the lane line data of the high-definition map data.
- the length of the lane line is set as the length of the lane line to be extracted in each cycle time, which can be set according to the actual situation of the road or the needs of the user.
- the length of the set lane line is determined according to at least one of the following conditions: the speed of the vehicle and the degree of curvature change of the road.
- the vehicle speed is relatively fast and the change in road curvature is relatively small, and the length of the lane line to be fitted is relatively large, so the length of the lane line is set to be the first lane line length; in a low-speed scene, the vehicle speed It is relatively slow and the change of road curvature is relatively large, so fitting a shorter lane line can meet the requirements of vehicle control, and there is no need to waste computing power by fitting too long lane lines, so set the length of the set lane line as the first The length of the second lane line, the length of the second lane line is less than the length of the first lane line.
- the target lane line data can be described as a series of points in vehicle coordinates (x 0 , y 0 ), (x 1 , y 1 ), (x 2 , y 2 ),...., (x n , y n ), where n is the number of data points in the target lane line data, and n can be determined by the length of the target lane line and the sampling frequency of the high-precision map data.
- Any curve fitting method in the related art can be used to perform curve fitting on the target lane line data to obtain the lane line, such as calling a curve fitting function or a custom curve fitting algorithm, which is not limited in this embodiment of the present application.
- the high-precision map data includes lane line data and planned path data of the vehicle; based on the current location point and the planned path data, Extracting the target lane line data within the set lane line length in front of the lane where the vehicle is located from the lane line data; performing curve fitting on the target lane line data to obtain the lane line, and extracting the lane line based on the high-precision map, Solve the problem that the traditional method of identifying lane lines on the ground through image acquisition equipment is limited by weather, ground light reflection and lane line wear, and the accuracy of lane line recognition is not high. It can be used alone or with traditional lane line recognition. The method is used in combination to reduce the constraints of lane line recognition and improve the accuracy of lane line recognition.
- Fig. 2 is a flow chart of a lane line extraction method in Embodiment 2 of the present application. This embodiment is based on the above-mentioned embodiment, for step S120, based on the current position point and the planned path data, from the lane The target lane line data within the set lane line length in front of the lane where the vehicle is located is extracted from the line data for illustration.
- the method of this embodiment includes the following steps.
- the current lane where the vehicle is located is determined according to the current location point and the planned route data, and the current lane line layer data corresponding to the current lane is extracted from the lane line data based on the current lane ID.
- the data size corresponding to the current lane line layer data is determined by the storage method of the high-precision map, and the lane line layer data sizes corresponding to different lane IDs can be different.
- the end position point in the current lane line layer data refers to a position point farthest from the current position point in the current vehicle forward direction stored in the current lane line layer data.
- the current lane line layer data is a sequence of position points, determine the end position point of the current lane line layer data, according to the number of data points between the current position point and the end position point, and the lane line layer in the high-precision map Data sampling frequency (unit, m/s), calculate the distance between the current location point and the end location point, and determine the distance as the current lane line length.
- the first current lane line data whose length is the set lane line length is obtained from the current lane line layer data along the vehicle's forward direction as the target lane line data.
- the target lane line data is used to determine the target lane line, that is, to obtain the target lane line whose length is the set lane line length.
- the predicted lane refers to the lane determined by the planned path ahead of the vehicle.
- the length of the current lane line is less than the set length of the lane line, it means that the length of the current lane line is not enough to determine the lane line. It is necessary to determine the predicted lane of the vehicle in the forward direction according to the current position point and the planned path data, and extract the lane from the predicted lane line, so that the length of the target lane line data reaches the set lane line length. However, there may be a situation that there is no road ahead, so it is judged whether the predicted lane identification number corresponding to the predicted lane is empty, and the target lane line data is obtained from the lane line data according to the judgment result.
- the high-precision map data includes lane line data and planned path data of the vehicle; based on the current location point and the planned path data, Extracting the target lane line data within the set lane line length in front of the lane where the vehicle is located from the lane line data; performing curve fitting on the target lane line data to obtain the lane line, and extracting the lane line based on the high-precision map, Solve the problem that the traditional method of identifying lane lines on the ground through image acquisition equipment is limited by weather, ground light reflection and lane line wear, and the accuracy of lane line recognition is not high. It can be used alone or with traditional lane line recognition. The method is used in combination to achieve the effects of reducing the constraints of lane line recognition and improving the accuracy of lane line recognition.
- step S240 acquiring the first current lane line data whose length is the set lane line length from the current lane line layer data includes: determining the current position point in the current lane line layer data Corresponding projection point, the projection point is the boundary point projected from the current position point onto the current lane boundary line; starting from the projection point in obtaining the current lane line layer data, the length is the set lane The boundary points included in the current lane boundary line of the line length; determining the sequence formed by the boundary points as the first current lane line data.
- Project the current position point of the vehicle onto the lane line determine the projection point of the current position point in the current lane line layer data, take the projection point as the starting point, and intercept the current lane line layer data whose length is the set lane line length
- the boundary points included in the lane boundary line, the sequence formed by the boundary points is determined as the first current lane line data.
- the coordinates of the current position point are (x 0 , y 0 )
- the length of the lane line is set to 100 meters
- the boundary points in the current lane line layer data are obtained: (x 1 , y 1 ), (x 2 , y 2 ),..., (x m , y m ).
- step S250: acquiring the target lane line data from the lane line data based on the judgment result includes the following steps.
- Step S251 If the judgment result is that the predicted lane identification number is empty, obtain the second current lane line data whose length is the length of the current lane line from the current lane line layer data, and convert the second current lane line data The lane line data is determined as the target lane line data;
- the predicted lane identification number is empty, it means that there is no predicted road in front of the current lane, and the second current lane line data whose length is the length of the current lane line is obtained from the current lane line layer data, and the second The current lane marking data is determined as the target lane marking data.
- the current lane line layer data is the position coordinates of the boundary point on the current lane line boundary line with a length of 100 meters
- the current position point is the data point at the 45th meter in the current lane line layer data
- the predicted lane line layer data refers to the data stored in the predicted lane line layer, which is the position coordinates of the boundary points on the boundary line of the predicted lane.
- the predicted lane identification number is not empty, it means that there is a predicted road ahead of the current lane, then extract the predicted lane line layer data from the lane line data, and determine the predicted lane line length corresponding to the predicted lane line layer data.
- the predicted lane line length is determined based on the number of predicted lane line layer data and the data sampling frequency of the predicted lane line layer.
- step S253 Determine whether the sum of the predicted lane line length and the current lane line length is less than the set lane line length; if the sum of the predicted lane line length and the current lane line length is not less than the set If the length of the lane marking is smaller than the set length of the lane marking, then step S254 is executed.
- S254 Obtain first predicted lane line data with a length of a first length from the predicted lane line layer data, where the first length is the length difference between the set lane line length and the current lane line length , determining the set of the second current lane marking data and the first predicted lane marking data as the target lane marking data, and ending the judging process.
- the predicted lane line length is 150 meters
- the current lane line length is 45 meters
- the set lane line length is 100 meters
- the sum of the predicted lane line length and the current lane line length is greater than or equal to the set lane line length
- the predicted lane line data of 0-55 meters is obtained from the predicted lane line layer data.
- the current lane line data and the predicted lane line data are combined to obtain the target lane line data corresponding to the lane line with a length of 100 meters.
- step S255 Determine the pre-predicted lane of the vehicle based on the planned route data, extract the pre-predicted lane line layer data corresponding to the pre-predicted lane from the lane line data, and determine the pre-predicted lane line layer data The corresponding pre-predicted lane line length; adding the predicted lane line length and the prepared predicted lane length to obtain a total predicted lane line length; determining the total predicted lane line length as the predicted lane line length until the If the sum of the total predicted lane line length and the current lane line length is greater than the set lane line length, step S256 is executed.
- the predicted lane length is 10 meters
- the current lane length is 45 meters
- the set lane length is 100 meters
- the sum of the predicted lane length and the current lane length is less than the set lane length
- then Determine the pre-predicted lane of the vehicle based on the planned route obtain the data of the pre-predicted lane line layer, and determine the length of the pre-predicted lane line corresponding to the lane line layer data. If the length of the pre-predicted lane line is 50 meters, the predicted The predicted lane lengths are accumulated to obtain a total predicted lane line length of 60 meters, that is, the predicted lane line length is updated to 60 meters. If the sum of the total predicted lane length and the current lane length is 105 meters, which is greater than the set lane length of 100 meters, step S254 is executed to obtain the target lane data corresponding to the lane length of 100 meters.
- the predicted lane line length is 10 meters
- the current lane line length is 45 meters
- the set lane line length is 100 meters
- the sum of the predicted lane line length and the current lane line length is less than the set lane line length
- step S255 execute step S255 again: determine the second preliminary predicted lane of the vehicle based on the planned route, obtain the layer data of the second preliminary predicted lane line, and determine the lane line
- the second preliminary predicted lane line length corresponding to the layer data if the second preliminary predicted lane line length is 40 meters, the predicted lane line length and the second preliminary predicted lane length are accumulated to obtain a total predicted lane line length of 80 meters, said The sum of the total predicted lane line length and the current lane line length is 125 meters, which is 100 meters greater than the set lane line length.
- Step S256 Obtain second predicted lane line data whose length is the length of the predicted lane line from the predicted lane line layer data; acquire a second-length preliminary predicted lane line data from the preliminary predicted lane line layer data data; the second length is the set lane line length minus the sum of the current lane line length and the second forecast lane line length; the second current lane line data, the second forecast A set of lane line data and the preliminary predicted lane line data is determined as the target lane line data.
- the predicted lane line length is 10 meters
- the first preliminary predicted lane line length is 30 meters
- the second preliminary predicted lane line length is 40 meters
- set If the length of the lane line is 100 meters, then obtain the second predicted lane line data of 10 meters in length from the predicted lane line layer data; First preliminary predicted lane line data; acquiring second preliminary predicted lane line data with a length of 15 meters from the second preliminary predicted lane line layer data.
- a set of the second current lane line data, the predicted lane line data and the preliminary predicted lane line data is determined as target lane line data.
- the method before extracting the preliminary prediction lane line layer data corresponding to the preliminary prediction lane from the lane line data, the method further includes: if the lane identification number corresponding to the preliminary prediction lane is empty, then Obtain predicted lane line data whose length is the length of the predicted lane line from the predicted lane line layer data; determine the set of the second current lane line data and the predicted lane line data as the target lane line data.
- the current lane line length is 45 meters
- the predicted lane line length is 10 meters
- the first preliminary predicted lane line length is 30 meters
- the set lane line length is 100 meters
- the second preliminary predicted lane line length is 100 meters.
- the second current lane line data with a length of 45 meters is obtained from the current lane line layer data
- the second predicted lane line with a length of 10 meters is obtained from the predicted lane line layer data
- a set of the second current lane marking data, the predicted lane marking data and the first preliminary predicted lane marking data is determined as the target lane marking data with a length of 85 meters.
- FIG. 3 is a schematic structural diagram of a lane line extraction device provided in Embodiment 3 of the present application. This embodiment is applicable to the situation of extracting lane lines based on a high-precision map.
- the device can be implemented in software and/or hardware, and the device can be integrated in a vehicle.
- the lane line extraction device includes : an acquisition module 310 , an extraction module 320 and a fitting module 330 .
- the acquisition module 310 is configured to acquire the current location point of the vehicle and the high-precision map data, and the high-precision map data includes lane line data and the planned path data of the vehicle; the extraction module 320 is configured to obtain the current location point and the planned path data based on the vehicle. Data, from the lane line data of the high-precision map data, extract the target lane line data within the set lane line length in front of the lane where the vehicle is located; the fitting module 330 is configured to perform curve fitting on the target lane line data Converge to get the lane line.
- the extraction module 320 includes: an extraction unit configured to extract current lane line layer data from the lane line data, the current lane line layer data is based on the current position point and the The lane line layer data corresponding to the current lane determined by the planning path data; the first determination unit is configured to determine the distance between the current position point and the termination position point in the current lane line layer data as the current Lane line length; the second determining unit is configured to obtain a length from the current lane line layer data as the set lane line length if the current lane line length is greater than or equal to the set lane line length The first current lane line data, the first current lane line data is determined as the target lane line data; the acquisition unit is configured to be configured as if the length of the current lane line is less than the set lane line length, then Determine the predicted lane of the vehicle based on the current position point and the planned route data, judge whether the predicted lane identification number corresponding to the predicted lane is empty, and obtain the target lane from the lane line data
- the second determination unit is configured to: determine a projection point corresponding to the current position point in the current lane line layer data, and the projection point is the projection of the current position point to the current lane boundary The boundary point on the line; obtain the boundary points included in the current lane boundary line whose length is the length of the set lane line from the projection point in the current lane line layer data; The sequence is determined as the first current lane marking data.
- the acquisition unit includes: a determination subunit, configured to obtain a length equal to the length of the current lane line from the current lane line layer data if the judgment result is that the predicted lane identification number is empty.
- the second current lane line data, the second current lane line data is determined as the target lane line data; the acquisition subunit is set to, if the judgment result is that the predicted lane identification number is not empty, then from the Extracting the predicted lane line layer data corresponding to the predicted lane from the lane line data; determining the predicted lane line length corresponding to the predicted lane line layer data, based on the predicted lane line length and the current lane line length from the The target lane line data is obtained from the predicted lane line layer data.
- the acquisition subunit is configured to: judge whether the sum of the predicted lane line length and the current lane line length is less than the set lane line length; if the predicted lane line length and the The sum of the current lane line lengths is greater than or equal to the set lane line length, then the first predicted lane line data with a length of a first length is obtained from the predicted lane line layer data, and the first length is the Setting the length difference between the length of the lane line and the length of the current lane line; determining the set of the second current lane line data and the first predicted lane line data as the target lane line data.
- the acquisition subunit is still set as:
- the preparation prediction lane line layer data corresponding to the preparation prediction lane determine the preparation prediction lane line length corresponding to the preparation prediction lane line layer data; add up the prediction lane line length and the preparation prediction lane length to obtain the total prediction Lane line length; judging whether the sum of the total predicted lane line length and the current lane line length is less than the set lane line length; when the sum of the total predicted lane line length and the current lane line length is less than the set
- repeat execution update the predicted lane line length to the total predicted lane line length, determine the pre-predicted lane of the vehicle based on the planned path data, and obtain Extract the corresponding preliminary predicted lane line layer data of the prepared predicted lane, determine the prepared predicted lane line length corresponding
- the acquisition subunit is still set to: if the lane identification number corresponding to the preliminary predicted lane is empty, then acquire the first lane whose length is the length of the predicted lane line from the predicted lane line layer data.
- Predicted lane line data determining a set of the second current lane line data and the second predicted lane line data as the target lane line data.
- the set lane line length is determined according to at least one of the following conditions: vehicle speed and road curvature change degree.
- the above-mentioned products can execute the method provided by any embodiment of the present application, and have corresponding functional modules for executing the method.
- FIG. 4 is a structural block diagram of a vehicle provided in Embodiment 4 of the present application.
- the vehicle includes a processor 410, a memory 420, an input device 430, and an output device 440; the number of processors 410 in the vehicle can be One or more, one processor 410 is taken as an example in FIG. 4; the processor 410, memory 420, input device 430 and output device 440 in the vehicle can be connected through a bus or other methods, and the connection through a bus is taken as an example in FIG. 4 .
- the memory 420 can be set to store software programs, computer-executable programs and modules, such as program instructions/modules corresponding to the lane line extraction method in the embodiment of the present application (for example, in the lane line extraction device).
- the processor 410 executes various functional applications and data processing of the vehicle by running the software programs, instructions and modules stored in the memory 420 , that is, implements the above-mentioned lane line extraction method.
- the memory 420 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application program required by at least one function; the data storage area may store data created according to the use of the terminal, and the like.
- the memory 420 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage devices.
- memory 420 may also include memory located remotely from processor 410 , which may be connected to the vehicle via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
- the input device 430 may be configured to receive input of numeric or character information, and generate key signal input related to user settings and function controls of the vehicle.
- the output device 440 may include a display device such as a display screen.
- Embodiment 5 of the present application provides a computer-readable storage medium on which a computer program is stored.
- the program is executed by a processor, the lane line extraction method provided in all embodiments of the present application is realized: obtaining the current position point of the vehicle and high-precision map data, the high-precision map data includes lane line data and planned path data of the vehicle; based on the current location point and the planned path data, extract the lane ahead of the vehicle from the lane line data Setting the target lane line data within the length of the lane line; performing curve fitting on the target lane line data to obtain the lane line.
- the computer readable medium may be a computer readable signal medium or a computer readable storage medium.
- a computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof.
- a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.
- a computer readable signal medium may include a data signal carrying computer readable program code in baseband or as part of a carrier wave. Such propagated data signals may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.
- a computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. .
- the program code contained on the computer readable medium can be transmitted by any appropriate medium, including but not limited to wireless, electric wire, optical cable, radio frequency (Radio Frequency, RF), etc., or any suitable combination of the above.
- any appropriate medium including but not limited to wireless, electric wire, optical cable, radio frequency (Radio Frequency, RF), etc., or any suitable combination of the above.
- Computer program codes for performing the operations of the present application may be written in one or more programming languages or combinations thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional A procedural programming language, such as the "C" language or similar programming language.
- the program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server.
- the remote computer may be connected to the user's computer via any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or it may be connected to an external computer (e.g., using an Internet service Provider via Internet connection).
- LAN Local Area Network
- WAN Wide Area Network
- Internet service Provider via Internet connection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
Abstract
本申请公开了一种车道线提取方法、装置、车辆及存储介质。该方法包括：获取车辆的当前位置点和高精地图数据，所述高精地图数据包括车道线数据和所述车辆的规划路径数据；基于所述当前位置点和所述规划路径数据，从所述车道线数据中提取所述车辆所在车道前方设定车道线长度内的目标车道线数据；对所述目标车道线数据进行曲线拟合得到车道线。
Description
本申请要求在2021年11月22日提交中国专利局、申请号为202111385472.X的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
本申请实施例涉及车道线检测技术领域，例如涉及一种车道线提取方法、装置、车辆及存储介质。
随着人工智能技术的发展，自动驾驶技术得到快速发展。在自动驾驶领域车辆自动采集环境信息并根据环境信息自动行驶，车道线检测是自动驾驶车辆中必不可少的基础功能。
基于机器视觉的车道线识别系统主要是通过安装在车辆前方的摄像头等图像传感器采集前方道路图像，对图像进行车道线的识别和提取。在车道线识别的过程中，受天气、地面光线反射或者车道线磨损程度大等原因的影响，图像传感器所采集到的车道线图像出现车道线残缺的问题较为严重，从而严重影响了车道线识别的准确度。
申请内容
本申请实施例提供一种车道线提取方法、装置、车辆及存储介质，以减少车道线识别的约束条件、提高车道线的识别准确率。
本申请实施例提供了一种车道线提取方法，包括：获取车辆的当前位置点和高精地图数据，所述高精地图数据包括车道线数据和所述车辆的规划路径数据；基于所述当前位置点和所述规划路径数据，从所述车道线数据中提取所述车辆所在车道前方设定车道线长度内的目标车道线数据；对所述目标车道线数据进行曲线拟合得到车道线。
本申请实施例还提供了一种车道线提取装置，该装置包括：获取模块，设置为获取车辆的当前位置点和高精地图数据，高精地图数据包括车道线数据和所述车辆的规划路径数据；提取模块，设置为基于所述当前位置点和所述规划路径数据，从所述高精地图数据的车道线数据中提取所述车辆所在车道前方设定车道线长度内的目标车道线数据；拟合模块，设置为对所述目标车道线数据进行曲线拟合得到车道线。
本申请实施例还提供了一种车辆，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现如本申请实施例中任一所述的车道线提取方法。
本申请实施例还提供了一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如本申请实施例中任一所述的车道线提取方法。
下面将对实施例中所需要使用的附图作简单地介绍，应当理解，以下附图仅示出了本申请的一些实施例，因此不应被看作是对范围的限定，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他相关的附图。
图1是本申请实施例一中的一种车道线提取方法的流程图；
图2是本申请实施例二中的一种车道线提取方法的流程图；
图3是本申请实施例三中的一种车道线提取装置的结构示意图；
图4是本申请实施例四中的一种车辆的结构示意图。
下面结合附图和实施例对本申请进行说明。可以理解的是，此处所描述的具体实施例仅仅用于解释本申请，而非对本申请的限定。另外还需要说明的是，为了便于描述，附图中仅示出了与本申请相关部分的结构。
应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦一项在一个附图中被定义，则在随后的附图中不需要对其进行定义和解释。同时，在本申请的描述中，术语“第一”、“第二”等仅用于区分描述，而不能理解为指示或暗示相对重要性。
实施例一
图1为本申请实施例一提供的一种车道线提取方法的流程图，本实施例可适用于基于高精地图提取车道线的情况，该方法可以由本申请实施例中的车道线提取装置来执行，该装置可采用软件和/或硬件的方式实现。
如图1所示，该方法包括如下步骤。
S110，获取车辆的当前位置点和高精地图数据，高精地图数据包括车道线数据和车辆的规划路径数据。
高精地图是指高精度、精细化定义的地图，其精度可达到分米级能够区分 多个车道。精细化定义是指高精地图采用格式化的存储方式存储交通场景中的多种交通要素，也就是高精地图数据可以包括传统地图的道路网数据、车道网络数据、车道线数据以及交通标志等数据。在车辆在高精地图中基于车辆的当前位置点、车辆将要达到的目的地以及道路环境信息确定规划路径后，高精地图数据中还包括车辆的规划路径数据，用于表示高精地图中车辆的规划路径对应的车道。
需要说明的，在高精地图中，高精地图数据的格式化存储方式为：基于车道直行长度、分流、合并、曲率等变化情况，将车道线数据以图层的方式存储在高精地图中，并为每个图层中的车道分配车道标识号。
示例性的，将长度为1000米的直行车道的0-500米和500米-1000米车道数据，分别存储在车道标识号为R01和R02的车道图层中；将R02车道右侧长度为400米的右转车道存储在车道标识号为R03的车道图层中，将R02车道左侧长度为300米的左转车道存储在车道标识号为R04的车道图层中。高精地图中还存储有车道线图层，用于存储车道两侧的车道线数据。
高精地图数据包括车道线数据和车辆的规划路径数据，车道线数据为车道的边界线上的边界点构成的序列，规划路径数据为车辆规划轨迹点构成的序列。车辆的当前位置点为车辆在高精地图中的二维位置坐标。可以采用相关技术中的任意方法确定车辆当前位置点和规划路径，本申请对此不设限制。
S120，基于当前位置点和规划路径数据，从高精地图数据的车道线数据中提取车辆所在车道前方设定车道线长度内的目标车道线数据。
目标车道线数据是从高精地图数据的车道线数据中所提取的设定车道线长度内的车道的边界线上的边界点构成的序列。
设定车道线长度为每个周期时间内所要提取车道线的长度，可以根据道路的实际情况或用户的需求设定。
基于规划路径数据和当前位置点确定规划路径对应的车道图层，基于车道图层对应的车道标识号(Identity，ID)，获取车道线图层存储的车道线图层数据，从车道线图层数据中截取设定车道线长度的车道线数据作为目标车道线数据。
可选的，设定车道线长度根据以下至少一个条件确定：车辆的车速、道路曲率变化程度。
示例性的，在高速场景下，车速相对较快且道路曲率变化相对较小，需要拟合的车道线长度较大，因此设定车道线长度为第一车道线长度；在低速场景下，车速相对较慢且道路曲率变化相对较大，故拟合较短的车道线就能满足车 辆控制的要求，不需要拟合太长的车道线而浪费算力，因此设设定车道线长度为第二车道线长度，第二车道线长度小于第一车道线长度。
S130，对目标车道线数据进行曲线拟合得到车道线。
目标车道线数据可以描述为一系列车辆坐标下的点构成的序列(x
0，y
0)，(x
1，y
1)，(x
2，y
2)，....，(x
n，y
n)，其中，n为目标车道线数据中数据点的个数，n可以由目标车道线的长度和高精地图数据的采样频率所确定。可以采用任何相关技术中的曲线拟合方法对目标车道线数据进行曲线拟合得到车道线，如调用曲线拟合函数或自定义曲线拟合算法，本申请实施例对此不设限制。
示例性的，采用三次曲线拟合方法，对目标车道线数据进行曲线拟合得到车道线，三次曲线拟合方程为y′＝C
0+C
1x+C
2x
2+C
3x
3，C
0、C
1、C
2、C
3均为拟合参数。将目标车道线数据的数据点依次代入次曲线拟合方程得到：
基于最小二乘法求该组拟合参数C
0、C
1、C
2、C
3，使∑|y
i-y′
i|最小，(x
i，y
i)为目标车道线数据中的第i个数据点的坐标，y
i为y
i对应的期望值。
本实施例的技术方案，通过获取车辆的当前位置点和高精地图数据，所述高精地图数据包括车道线数据和车辆的规划路径数据；基于所述当前位置点和所述规划路径数据，从所述车道线数据中提取所述车辆所在车道前方设定车道线长度内的目标车道线数据；对所述目标车道线数据进行曲线拟合得到车道线，能够基于高精地图提取车道线，解决传统的通过图像采集设备识别地面上的车道线的方法，受天气、地面光线反射和车道线磨损程度的约束，车道线识别准确度不高的问题，能够单独使用或与传统的车道线识别方法结合使用，减少车道线识别的约束条件、提高车道线的识别准确率。
实施例二
图2为本申请实施例二中的一种车道线提取方法的流程图，本实施例以上述实施例为基础对步骤S120，基于所述当前位置点和所述规划路径数据，从所述车道线数据中提取所述车辆所在车道前方设定车道线长度内的目标车道线数据进行说明。
如图2所示，本实施例的方法包括如下步骤。
S210，获取车辆的当前位置点和高精地图数据，高精地图数据包括车道线数据和车辆的规划路径数据。
S220，从车道线数据中提取当前车道线图层数据，当前车道线图层数据是基于当前位置点和规划路径数据所确定的当前车道对应的车道线图层数据。
根据当前位置点和规划路径数据确定车辆所处的当前车道，基于当前车道ID从车道线数据中提取当前车道对应的当前车道线图层数据。
S230，将当前位置点与当前车道线图层数据中的终止位置点之间的距离确定为当前车道线长度。
当前车道线图层数据对应的数据大小由高精地图的存储方式所确定，不同车道ID对应的车道线图层数据大小都可以不相同。当前车道线图层数据中的终止位置点是指当前车道线图层数据中存储的当前车辆前进方向上距离当前位置点最远的一个位置点。
当前车道线图层数据为一系列位置点构成的序列，确定当前车道线图层数据的终止位置点，根据当前位置点和终止位置点之间的数据点数，以及高精地图中车道线图层数据的采样频率(单位，米/秒)，计算得到当前位置点和终止位置点之间的距离，将该距离确定为当前车道线长度。
S240，若当前车道线长度大于或等于设定车道线长度，则从当前车道线图层数据中获取长度为设定车道线长度的第一当前车道线数据，将第一当前车道线数据确定为目标车道线数据。
若当前车道线长度大于或等于设定车道线长度，从当前车道线图层数据中沿车辆前进方向获取长度为设定车道线长度的第一当前车道线数据，作为目标车道线数据。目标车道线数据用于确定目标车道线，即得到长度为设定车道线长度的目标车道线。
S250，若当前车道线长度小于设定车道线长度，则基于当前位置点和规划路径数据确定车辆的预测车道，判断预测车道对应的预测车道标识号是否为空，基于判断结果从车道线数据中获取目标车道线数据。
预测车道是指车辆前方由规划路径所确定的车道。
若当前车道线长度小于设定车道线长度，说明当前车道线的长度不足够用于确定车道线，需要根据当前位置点和规划路径数据确定车辆在前进方向的预测车道，从预测车道中提取车道线，以使目标车道线数据的长度达到设定车道线长度。但是可能存在前方没有道路的情况，因此，对预测车道对应的预测车道标识号是否为空进行判断，根据判断结果从车道线数据中获取目标车道线数据。
S260，对目标车道线数据进行曲线拟合得到车道线。
本实施例的技术方案，通过获取车辆的当前位置点和高精地图数据，所述高精地图数据包括车道线数据和车辆的规划路径数据；基于所述当前位置点和所述规划路径数据，从所述车道线数据中提取所述车辆所在车道前方设定车道 线长度内的目标车道线数据；对所述目标车道线数据进行曲线拟合得到车道线，能够基于高精地图提取车道线，解决传统的通过图像采集设备识别地面上的车道线的方法，受天气、地面光线反射和车道线磨损程度的约束，车道线识别准确度不高的问题，能够单独使用或与传统的车道线识别方法结合使用，实现减少车道线识别的约束条件、提高车道线的识别准确率的效果。
可选的，步骤S240，从所述当前车道线图层数据中获取长度为设定车道线长度的第一当前车道线数据，包括：确定所述当前车道线图层数据中所述当前位置点对应的投影点，所述投影点为所述当前位置点投影到当前车道边界线上的边界点；获取所述当前车道线图层数据中以所述投影点起，长度为所述设定车道线长度的当前车道边界线所包含的边界点；将所述边界点构成的序列确定为所述第一当前车道线数据。
将车辆的当前位置点投影到车道线上，确定当前位置点在当前车道线图层数据中的投影点，以投影点为起点，截取当前车道线图层数据中长度为设定车道线长度的车道边界线所包含的边界点，所述边界点构成的序列确定为第一当前车道线数据。
示例性的，若当前位置点的坐标为(x
0，y
0)，则当前车道线图层数据中当前位置点对应的投影点的坐标为(x
1，y
1)，且y
0＝y
1。若设定车道线长度为100米，则获取当前车道线图层数据中的边界点：(x
1，y
1)，(x
2，y
2)，......，(x
m，y
m)。(x
m，y
m)为第一当前车道线数据中的终点。若当前车道线图层数据的数据采样频率为f
s，则(m-1)f
s＝100米。
可选的，步骤S250：基于判断结果从所述车道线数据中获取所述目标车道线数据，包括如下步骤。
步骤S251：若判断结果为所述预测车道标识号为空，则从所述当前车道线图层数据中获取长度为所述当前车道线长度的第二当前车道线数据，将所述第二当前车道线数据确定为所述目标车道线数据；
若预测车道标识号为空，表示当前车道的前方不存在预测道路，将当前车道线图层数据中获取以投影点为起点，长度为当前车道线长度的第二当前车道线数据，将第二当前车道线数据确定为目标车道线数据。
示例性的，若当前车道线图层数据为长度为100米的当前车道边界线上的边界点的位置坐标，当前位置点为当前车道线图层数据中的处于第45米处的数据点，则获取当前车道线图层数据中第45米至第100米的当前车道边界线的边界点的位置坐标构成的第二当前车道线数据，并将第二当前车道线数据确定为目标车道线数据。
S252：若判断结果为所述预测车道标识号不为空，则从所述车道线数据中提取所述预测车道对应的预测车道线图层数据；确定所述预测车道线图层数据对应的预测车道线长度。
预测车道线图层数据是指存储于预测车道线图层中的数据，该数据为预测车道的边界线上的边界点的位置坐标。
若预测车道标识号不为空，表示当前车道的前方存在预测道路，则从车道线数据中提取预测车道线图层数据，确定预测车道线图层数据对应的预测车道线长度。所述预测车道线长度是基于预测车道线图层数据的个数和预测车道线图层的数据采样频率所确定。
S253：判断所述预测车道线长度和所述当前车道线长度之和是否小于所述设定车道线长度；若所述预测车道线长度和所述当前车道线长度之和不小于所述设定车道线长度，则执行步骤S254，若所述预测车道线长度和所述当前车道线长度之和小于所述设定车道线长度，则执行S255。
S254：从所述预测车道线图层数据中获取长度为第一长度的第一预测车道线数据，所述第一长度为所述设定车道线长度和所述当前车道线长度的长度差值，将所述第二当前车道线数据和所述第一预测车道线数据的集合确定为所述目标车道线数据，并结束判断过程。
示例性的，若预测车道线长度为150米，当前车道线长度为45米，设定车道线长度为100米，预测车道线长度和当前车道线长度之和大于或等于设定车道线长度，则确定设定车道线长度和当前车道线长度的长度差值为55米，从预测车道线图层数据中获取0-55米的预测车道线数据。将当前车道线数据和预测车道线数据合并得到，长度为100米的车道线对应的目标车道线数据。
S255：基于所述规划路径数据确定所述车辆的预备预测车道，从所述车道线数据中提取所述预备预测车道对应的预备预测车道线图层数据，确定所述预备预测车道线图层数据对应的预备预测车道线长度；将所述预测车道线长度和所述预备预测车道长度累加得到总预测车道线长度；将所述总预测车道线长度确定为所述预测车道线长度，直到所述总预测车道线长度和所述当前车道线长度之和大于所述设定车道线长度，执行步骤S256。
在一个例子中，若预测车道线长度为10米，当前车道线长度为45米，设定车道线长度为100米，预测车道线长度和当前车道线长度之和小于设定车道线长度，则基于规划路径确定车辆的预备预测车道，获取预备预测车道线图层数据，确定车道线图层数据对应的预备预测车道线长度，若预备预测车道线长度为50米，将预测车道线长度和预备预测车道长度累加得到总预测车道线长度 为60米，即预测车道线长度更新为60米。所述总预测车道线长度和所述当前车道线长度之和为105米，大于设定车道线长度100米，则执行步骤S254，从而长度为100米的车道线对应的目标车道线数据。
在另一个例子中，若预测车道线长度为10米，当前车道线长度为45米，设定车道线长度为100米，预测车道线长度和当前车道线长度之和小于设定车道线长度，则基于规划路径确定车辆的第一预备预测车道，获取第一预备预测车道线图层数据，确定车道线图层数据对应的第一预备预测车道线长度，若第一预备预测车道线长度为30米，将预测车道线长度和第一预备预测车道长度累加得到总预测车道线长度为40米。判断预测车道线长度和所述当前车道线长度之和为85米，则再次执行步骤S255：基于规划路径确定车辆的第二预备预测车道，获取第二预备预测车道线图层数据，确定车道线图层数据对应的第二预备预测车道线长度，若第二预备预测车道线长度为40米，将预测车道线长度和第二预备预测车道长度累加得到总预测车道线长度为80米，所述总预测车道线长度和所述当前车道线长度之和为125米，大于设定车道线长度100米。
步骤S256：从所述预测车道线图层数据中获取长度为所述预测车道线长度的第二预测车道线数据；从所述预备预测车道线图层数据中获取第二长度的预备预测车道线数据；所述第二长度为所述设定车道线长度减去所述当前车道线长度和所述第二预测车道线长度的和；将所述第二当前车道线数据、所述第二预测车道线数据和所述预备预测车道线数据的集合确定为所述目标车道线数据。
示例性的，在上述例子中，若当前车道线长度为45米，预测车道线长度为10米，第一预备预测车道线长度为30米，第二预备预测车道线长度为40米，设定车道线长度为100米，则从所述预测车道线图层数据中获取长度为10米的第二预测车道线数据；从所述第一预备预测车道线图层数据中获取长度为30米的第一预备预测车道线数据；从所述第二预备预测车道线图层数据中获取长度为15米的第二预备预测车道线数据。将所述第二当前车道线数据、所述预测车道线数据和所述预备预测车道线数据的集合确定为目标车道线数据。
可选的，在从所述车道线数据中提取所述预备预测车道对应的预备预测车道线图层数据之前，该方法还还包括：若所述预备预测车道对应的车道标识号为空，则从所述预测车道线图层数据中获取长度为所述预测车道线长度的预测车道线数据；将所述第二当前车道线数据和所述预测车道线数据的集合确定为所述目标车道线数据。
示例性的，在上述例子中，若当前车道线长度为45米，预测车道线长度为10米，第一预备预测车道线长度为30米，设定车道线长度为100米，第二预备 预测车道对应的车道标识号为空，则从当前车道线图层数据中获取长度为45米的第二当前车道线数据，从预测车道线图层数据中获取长度为10米的第二预测车道线数据；从第一预备预测车道线图层数据中获取长度为30米的第一预备预测车道线数据。将第二当前车道线数据、预测车道线数据和第一预备预测车道线数据的集合确定为长度为85米的目标车道线数据。
实施例三
图3为本申请实施例三提供的一种车道线提取装置的结构示意图。本实施例可适用于基于高精地图提取车道线的情况，该装置可采用软件和/或硬件的方式实现，该装置可集成在车辆中，如图3所示，所述车道线提取装置包括：获取模块310、提取模块320和拟合模块330。
获取模块310，设置为获取车辆的当前位置点和高精地图数据，高精地图数据包括车道线数据和车辆的规划路径数据；提取模块320，设置为基于所述当前位置点和所述规划路径数据，从所述高精地图数据的车道线数据中提取所述车辆所在车道前方设定车道线长度内的目标车道线数据；拟合模块330，设置为对所述目标车道线数据进行曲线拟合得到车道线。
可选的，所述提取模块320，包括：提取单元，设置为从所述车道线数据中提取当前车道线图层数据，所述当前车道线图层数据是基于所述当前位置点和所述规划路径数据所确定的当前车道对应的车道线图层数据；第一确定单元，设置为将所述当前位置点与所述当前车道线图层数据中的终止位置点之间的距离确定为当前车道线长度；第二确定单元，设置为若所述当前车道线长度大于或等于所述设定车道线长度，则从所述当前车道线图层数据中获取长度为所述设定车道线长度的第一当前车道线数据，将所述第一当前车道线数据确定为所述目标车道线数据；获取单元，用于设置为若所述当前车道线长度小于所述设定车道线长度，则基于所述当前位置点和所述规划路径数据确定所述车辆的预测车道，判断所述预测车道对应的预测车道标识号是否为空，基于判断结果从所述车道线数据中获取所述目标车道线数据。
可选的，所述第二确定单元，是设置为：确定所述当前车道线图层数据中所述当前位置点对应的投影点，所述投影点为所述当前位置点投影到当前车道边界线上的边界点；获取所述当前车道线图层数据中以所述投影点起，长度为所述设定车道线长度的当前车道边界线所包含的边界点；将所述边界点构成的序列确定为所述第一当前车道线数据。
可选的，所述获取单元，包括：确定子单元，设置为若判断结果为所述预测车道标识号为空，则从所述当前车道线图层数据中获取长度为所述当前车道线长度的第二当前车道线数据，将所述第二当前车道线数据确定为所述目标车 道线数据；获取子单元，设置为若判断结果为所述预测车道标识号不为空，则从所述车道线数据中提取所述预测车道对应的预测车道线图层数据；确定所述预测车道线图层数据对应的预测车道线长度，基于所述预测车道线长度和所述当前车道线长度从所述预测车道线图层数据中获取所述目标车道线数据。
可选的，所述获取子单元，是设置为：判断所述预测车道线长度和所述当前车道线长度之和是否小于所述设定车道线长度；若所述预测车道线长度和所述当前车道线长度之和大于或等于所述设定车道线长度，则从所述预测车道线图层数据中获取长度为第一长度的第一预测车道线数据，所述第一长度为所述设定车道线长度和所述当前车道线长度的长度差值；将所述第二当前车道线数据和所述第一预测车道线数据的集合确定为所述目标车道线数据。
可选的，所述获取子单元，还是设置为：
若所述预测车道线长度和所述当前车道线长度之和小于所述设定车道线长度，则基于所述规划路径数据确定所述车辆的预备预测车道，从所述车道线数据中提取所述预备预测车道对应的预备预测车道线图层数据，确定所述预备预测车道线图层数据对应的预备预测车道线长度；将所述预测车道线长度和所述预备预测车道长度累加得到总预测车道线长度；判断所述总预测车道线长度和所述当前车道线长度之和是否小于所述设定车道线长度；在所述总预测车道线长度和所述当前车道线长度之和小于所述设定车道线长度的情况下，重复执行：更新所述预测车道线长度为所述总预测车道线长度，基于所述规划路径数据确定所述车辆的预备预测车道，从所述车道线数据中提取所述预备预测车道对应的预备预测车道线图层数据，确定所述预备预测车道线图层数据对应的预备预测车道线长度；将所述预测车道线长度和所述预备预测车道长度累加得到总预测车道线长度；在所述总预测车道线长度和所述当前车道线长度之和大于所述设定车道线长度的情况下，从所述预测车道线图层数据中获取长度为所述预测车道线长度的第二预测车道线数据；从所述预备预测车道线图层数据中获取长度为第二长度的预备预测车道线数据；所述第二长度为所述设定车道线长度减去所述当前车道线长度和所述第二预测车道线长度的和；将所述第二当前车道线数据、所述第二预测车道线数据和所述预备预测车道线数据的集合确定为所述目标车道线数据。
可选的，所述获取子单元，还是设置为：若所述预备预测车道对应的车道标识号为空，则从所述预测车道线图层数据中获取长度为所述预测车道线长度的第二预测车道线数据；将所述第二当前车道线数据和所述第二预测车道线数据的集合确定为所述目标车道线数据。
可选的，所述设定车道线长度根据以下至少一个条件确定：车辆的车速、 道路曲率变化程度。
上述产品可执行本申请任意实施例所提供的方法，具备执行方法相应的功能模块。
实施例四
图4为本申请实施例四提供的一种车辆的结构框图，如图4所示，该车辆包括处理器410、存储器420、输入装置430和输出装置440；车辆中处理器410的数量可以是一个或多个，图4中以一个处理器410为例；车辆中的处理器410、存储器420、输入装置430和输出装置440可以通过总线或其他方式连接，图4中以通过总线连接为例。
存储器420作为一种计算机可读存储介质，可设置为存储软件程序、计算机可执行程序以及模块，如本申请实施例中的车道线提取方法对应的程序指令/模块(例如，车道线提取装置中的获取模块310、提取模块320和拟合模块330)。处理器410通过运行存储在存储器420中的软件程序、指令以及模块，从而执行车辆的多种功能应用以及数据处理，即实现上述的车道线提取方法。
存储器420可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序；存储数据区可存储根据终端的使用所创建的数据等。此外，存储器420可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在一些实例中，存储器420还可包括相对于处理器410远程设置的存储器，这些远程存储器可以通过网络连接至车辆。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
输入装置430可设置为接收输入的数字或字符信息，以及产生与车辆的用户设置以及功能控制有关的键信号输入。输出装置440可包括显示屏等显示设备。
实施例五
本申请实施例五提供了一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如本申请所有申请实施例提供的车道线提取方法：获取车辆的当前位置点和高精地图数据，所述高精地图数据包括车道线数据和车辆的规划路径数据；基于所述当前位置点和所述规划路径数据，从所述车道线数据中提取所述车辆所在车道前方设定车道线长度内的目标车道线数据；对所述目标车道线数据进行曲线拟合得到车道线。
可以采用一个或多个计算机可读的介质的任意组合。计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质。计算机可读存储介质例如可 以是但不限于电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：具有一个或多个导线的电连接、便携式计算机磁盘、硬盘、随机存取存储器(Random Access Memory，RAM)、只读存储器(Read-Only Memory，ROM)、可擦式可编程只读存储器(Erasable Programmable Read-Only Memory，EPROM)或闪存、光纤、便携式紧凑磁盘只读存储器(Compact Disc Read-Only Memory，CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本文件中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。
计算机可读的信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读的信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。
计算机可读介质上包含的程序代码可以用任何适当的介质传输，包括但不限于无线、电线、光缆、射频(Radio Frequency，RF)等，或者上述的任意合适的组合。
可以以一种或多种程序设计语言或其组合来编写用于执行本申请操作的计算机程序代码，所述程序设计语言包括面向对象的程序设计语言，诸如Java、Smalltalk、C++，还包括常规的过程式程序设计语言，诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络包括局域网(Local Area Network，LAN)或广域网(Wide AreaNetwork，WAN)连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。
Claims (11)
- 一种车道线提取方法，包括：获取车辆的当前位置点和高精地图数据，所述高精地图数据包括车道线数据和所述车辆的规划路径数据；基于所述当前位置点和所述规划路径数据，从所述车道线数据中提取所述车辆所在车道前方设定车道线长度内的目标车道线数据；对所述目标车道线数据进行曲线拟合得到车道线。
- 根据权利要求1所述的方法，其中，所述基于所述当前位置点和所述规划路径数据，从所述车道线数据中提取所述车辆所在车道前方设定车道线长度内的目标车道线数据，包括：从所述车道线数据中提取当前车道线图层数据，所述当前车道线图层数据是基于所述当前位置点和所述规划路径数据所确定的当前车道对应的车道线图层数据；将所述当前位置点与所述当前车道线图层数据中的终止位置点之间的距离确定为当前车道线长度；在所述当前车道线长度大于或等于所述设定车道线长度的情况下，从所述当前车道线图层数据中获取长度为所述设定车道线长度的第一当前车道线数据，将所述第一当前车道线数据确定为所述目标车道线数据；在所述当前车道线长度小于所述设定车道线长度的情况下，基于所述当前位置点和所述规划路径数据确定所述车辆的预测车道，判断所述预测车道对应的预测车道标识号是否为空，基于判断结果从所述车道线数据中获取所述目标车道线数据。
- 根据权利要求2所述的方法，其中，所述从所述当前车道线图层数据中获取长度为所述设定车道线长度的第一当前车道线数据，包括：确定所述当前车道线图层数据中所述当前位置点对应的投影点，所述投影点为所述当前位置点投影到当前车道边界线上的边界点；获取所述当前车道线图层数据中以所述投影点起，长度为所述设定车道线长度的当前车道边界线所包含的边界点；将所述边界点构成的序列确定为所述第一当前车道线数据。
- 根据权利要求2所述的方法，其中，所述基于判断结果从所述车道线数据中获取所述目标车道线数据，包括：响应于所述判断结果为所述预测车道标识号为空，从所述当前车道线图层 数据中获取长度为所述当前车道线长度的第二当前车道线数据，将所述第二当前车道线数据确定为所述目标车道线数据；响应于所述判断结果为所述预测车道标识号不为空，从所述车道线数据中提取所述预测车道对应的预测车道线图层数据；确定所述预测车道线图层数据对应的预测车道线长度，基于所述预测车道线长度和所述当前车道线长度从所述预测车道线图层数据中获取所述目标车道线数据。
- 根据权利要求4所述的方法，其中，基于所述预测车道线长度和所述当前车道线长度从所述预测车道线图层数据中获取所述目标车道线数据，包括：判断所述预测车道线长度和所述当前车道线长度之和是否小于所述设定车道线长度；响应于所述预测车道线长度和所述当前车道线长度之和大于或等于所述设定车道线长度，从所述预测车道线图层数据中获取长度为第一长度的第一预测车道线数据，所述第一长度为所述设定车道线长度和所述当前车道线长度的长度差值；将所述第二当前车道线数据和所述第一预测车道线数据的集合确定为所述目标车道线数据。
- 根据权利要求5所述的方法，其中，响应于所述预测车道线长度和所述当前车道线长度之和小于所述设定车道线长度，基于所述规划路径数据确定所述车辆的预备预测车道，从所述车道线数据中提取所述预备预测车道对应的预备预测车道线图层数据，确定所述预备预测车道线图层数据对应的预备预测车道线长度；将所述预测车道线长度和所述预备预测车道长度累加得到总预测车道线长度；判断所述总预测车道线长度和所述当前车道线长度之和是否小于所述设定车道线长度；在所述总预测车道线长度和所述当前车道线长度之和小于所述设定车道线长度的情况下，重复执行：更新所述预测车道线长度为所述总预测车道线长度，基于所述规划路径数据确定所述车辆的预备预测车道，从所述车道线数据中提取所述预备预测车道对应的预备预测车道线图层数据，确定所述预备预测车道线图层数据对应的预备预测车道线长度；将所述预测车道线长度和所述预备预测车道长度累加得到总预测车道线长度；在所述总预测车道线长度和所述当前车道线长度之和大于或等于所述设定 车道线长度的情况下，从所述预测车道线图层数据中获取长度为所述预测车道线长度的第二预测车道线数据；从所述预备预测车道线图层数据中获取长度为第二长度的预备预测车道线数据；其中，所述第二长度为所述设定车道线长度减去所述当前车道线长度和所述第二预测车道线长度的和；将所述第二当前车道线数据、所述第二预测车道线数据和所述预备预测车道线数据的集合确定为所述目标车道线数据。
- 根据权利要求6所述的方法，在所述从所述车道线数据中提取所述预备预测车道对应的预备预测车道线图层数据之前，还包括：在所述预备预测车道对应的车道标识号为空的情况下，从所述预测车道线图层数据中获取长度为所述预测车道线长度的第二预测车道线数据；将所述第二当前车道线数据和所述第二预测车道线数据的集合确定为所述目标车道线数据。
- 根据权利要求1-7任一所述的方法，其中，所述设定车道线长度根据以下至少一个条件确定：车辆的车速、道路曲率变化程度。
- 一种车道线提取装置，包括：获取模块，设置为获取车辆的当前位置点和高精地图数据，高精地图数据包括车道线数据和所述车辆的规划路径数据；提取模块，设置为基于所述当前位置点和所述规划路径数据，从所述高精地图数据的车道线数据中提取所述车辆所在车道前方设定车道线长度内的目标车道线数据；拟合模块，设置为对所述目标车道线数据进行曲线拟合得到车道线。
- 一种车辆，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现如权利要求1-8中任一所述的车道线提取方法。
- 一种计算机可读存储介质，其上存储有计算机程序，其中，该程序被处理器执行时实现如权利要求1-8中任一所述的车道线提取方法。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP22895015.0A EP4345773A1 (en) | 2021-11-22 | 2022-11-22 | Lane line extraction method and apparatus, vehicle and storage medium |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111385472.X | 2021-11-22 | ||
| CN202111385472.XA CN114092909A (zh) | 2021-11-22 | 2021-11-22 | 一种车道线提取方法、装置、车辆及存储介质 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023088486A1 true WO2023088486A1 (zh) | 2023-05-25 |
Family
ID=80302680
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2022/133480 WO2023088486A1 (zh) | 2021-11-22 | 2022-11-22 | 车道线提取方法、装置、车辆及存储介质 |
Country Status (3)
| Country | Link |
|---|---|
| EP (1) | EP4345773A1 (zh) |
| CN (1) | CN114092909A (zh) |
| WO (1) | WO2023088486A1 (zh) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114092909A (zh) * | 2021-11-22 | 2022-02-25 | 中国第一汽车股份有限公司 | 一种车道线提取方法、装置、车辆及存储介质 |
| CN116793369B (zh) * | 2023-02-10 | 2024-03-08 | 北京斯年智驾科技有限公司 | 路径规划方法、装置、设备及计算机可读存储介质 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108007471A (zh) * | 2016-10-28 | 2018-05-08 | 北京四维图新科技股份有限公司 | 车辆引导区块获取方法和装置以及自动驾驶方法和系统 |
| US20190130182A1 (en) * | 2017-11-01 | 2019-05-02 | Here Global B.V. | Road modeling from overhead imagery |
| CN111815742A (zh) * | 2020-09-14 | 2020-10-23 | 蘑菇车联信息科技有限公司 | 一种车道线生成方法及系统 |
| CN114092909A (zh) * | 2021-11-22 | 2022-02-25 | 中国第一汽车股份有限公司 | 一种车道线提取方法、装置、车辆及存储介质 |
-
2021
- 2021-11-22 CN CN202111385472.XA patent/CN114092909A/zh active Pending
-
2022
- 2022-11-22 WO PCT/CN2022/133480 patent/WO2023088486A1/zh active Application Filing
- 2022-11-22 EP EP22895015.0A patent/EP4345773A1/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108007471A (zh) * | 2016-10-28 | 2018-05-08 | 北京四维图新科技股份有限公司 | 车辆引导区块获取方法和装置以及自动驾驶方法和系统 |
| US20190130182A1 (en) * | 2017-11-01 | 2019-05-02 | Here Global B.V. | Road modeling from overhead imagery |
| CN111815742A (zh) * | 2020-09-14 | 2020-10-23 | 蘑菇车联信息科技有限公司 | 一种车道线生成方法及系统 |
| CN114092909A (zh) * | 2021-11-22 | 2022-02-25 | 中国第一汽车股份有限公司 | 一种车道线提取方法、装置、车辆及存储介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114092909A (zh) | 2022-02-25 |
| EP4345773A1 (en) | 2024-04-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11105638B2 (en) | Method, apparatus, and computer readable storage medium for updating electronic map | |
| WO2023088486A1 (zh) | 车道线提取方法、装置、车辆及存储介质 | |
| JP7351487B2 (ja) | トポロジーマップに基づくインテリジェントナビゲーションの方法及びシステム | |
| WO2017041390A1 (zh) | 高精地图数据的处理方法、装置、存储介质和设备 | |
| US20220343758A1 (en) | Data Transmission Method and Apparatus | |
| US20220076038A1 (en) | Method for controlling vehicle and electronic device | |
| US11783588B2 (en) | Method for acquiring traffic state, relevant apparatus, roadside device and cloud control platform | |
| EP4016130A2 (en) | Method for outputting early warning information, device, storage medium and program product | |
| CN112883236B (zh) | 一种地图更新方法、装置、电子设备及存储介质 | |
| WO2021253245A1 (zh) | 识别车辆变道趋势的方法和装置 | |
| WO2022021982A1 (zh) | 可行驶区域判定的方法、智能驾驶系统和智能汽车 | |
| CN111380546A (zh) | 基于平行道路的车辆定位方法、装置、电子设备和介质 | |
| WO2023155580A1 (zh) | 一种对象识别方法和装置 | |
| WO2024131436A1 (zh) | 自动驾驶车辆的能量回收方法、装置及系统 | |
| CN113722342A (zh) | 高精地图要素变更检测方法、装置、设备及自动驾驶车辆 | |
| CN113496182B (zh) | 基于遥感影像的道路提取方法及装置、存储介质及设备 | |
| WO2020135325A1 (zh) | 移动设备定位方法、装置、系统及移动设备 | |
| US20230126172A1 (en) | Method of outputting prompt information, device, medium, and vehicle | |
| CN112987707A (zh) | 一种车辆的自动驾驶控制方法及装置 | |
| CN113450794B (zh) | 导航播报的检测方法、装置、电子设备和介质 | |
| WO2021008500A1 (zh) | 图像处理方法和装置 | |
| US20230274557A1 (en) | Method for determining line pressing state of a vehicle, electronic device, and non-transitory computer-readable storage medium | |
| JP2023011912A (ja) | 走行環境における交通オブジェクトの融合及び関連付け方法、装置並びにエッジコンピューティングデバイス | |
| KR20220119167A (ko) | 차량의 차선 이탈 식별 방법 및 장치，전자 기기, 및 저장매체 | |
| CN115631468A (zh) | 一种面向智慧车联服务生态的多模态数据智能监控方法 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22895015 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2022895015 Country of ref document: EP |
|
| ENP | Entry into the national phase |
Ref document number: 2022895015 Country of ref document: EP Effective date: 20231229 |