WO2020063813A1 - 用于提取环境目标的特征点的方法及装置 - Google Patents

用于提取环境目标的特征点的方法及装置 Download PDF

Info

Publication number
WO2020063813A1
WO2020063813A1 PCT/CN2019/108399 CN2019108399W WO2020063813A1 WO 2020063813 A1 WO2020063813 A1 WO 2020063813A1 CN 2019108399 W CN2019108399 W CN 2019108399W WO 2020063813 A1 WO2020063813 A1 WO 2020063813A1
Authority
WO
WIPO (PCT)
Prior art keywords
point
environmental target
vehicle
target
area
Prior art date
Application number
PCT/CN2019/108399
Other languages
English (en)
French (fr)
Inventor
刘洪亮
葛建勇
甄龙豹
韩汝涛
李卫
赵俊鹏
曾荣林
田超凯
邓伟峰
Original Assignee
长城汽车股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 长城汽车股份有限公司 filed Critical 长城汽车股份有限公司
Priority to US17/281,560 priority Critical patent/US11928870B2/en
Priority to KR1020217013022A priority patent/KR102534412B1/ko
Priority to JP2021517849A priority patent/JP7299976B2/ja
Priority to EP19865835.3A priority patent/EP3859594A4/en
Publication of WO2020063813A1 publication Critical patent/WO2020063813A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • 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
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
    • B60W40/06—Road conditions
    • 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
    • G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes

Definitions

  • the present invention relates to the field of vehicles, and in particular, to a method and device for extracting feature points of an environmental target.
  • an autonomous vehicle needs to use sensors installed on the vehicle body to obtain information about the surrounding environment targets, and combine the surrounding environment lane line information to segment the targets to further obtain the target's area attributes.
  • Targets in different areas can be used for longitudinal control and lane change decisions, while lane line information can provide vehicles with travel path information for lateral control for lane change and lane keeping.
  • the present invention aims to propose a method for extracting feature points of an environmental target, for at least solving the extraction of feature points of an environmental target.
  • a method for extracting feature points of an environmental target comprising: acquiring a position of each corner point of the environmental target; and determining the environmental target according to the position of each corner point of the environmental target Characteristic points.
  • the feature points of the environmental target include fixed area feature points and / or follow-up area feature points
  • the fixed area feature points include one or more of the following: a fixed area lateral nearest point, and the fixed area lateral nearest point Refers to the horizontally closest point on the environmental target to the reference line of the traveling coordinate system X F O F Y F ; the horizontally farthest point in the fixed area, the horizontally furthest point in the fixed area refers to the horizontal distance from the reference line on the environmental target The farthest point; the longitudinal closest point of the fixed area, the longitudinal closest point of the fixed area refers to the longitudinal closest point on the environmental target to the coordinate origin O F of the driving coordinate system X F O F Y F ; the fixed area 3.75 feature points , wherein the fixation point is 3.75 area of the intersection from the lane lines on both sides of the lane of the target environment where the vehicle longitudinal coordinate point nearest the origin O F; 2.8 feature point fixing region, the fixing region 2.8 The feature point refers to a point longitudinally closest to the
  • the lane lines on both sides of the lane are parallel and are respectively virtual lines with a predetermined lateral distance from one of the lane lines on both sides; the following area is located in front of the driving direction of the vehicle, parallel to the road direction, and has a predetermined width.
  • the characteristic point of the following region refers to a point longitudinally closest to the coordinate origin O F among intersection points of the environmental target and the following region; wherein the driving coordinate system X F O F Y F is The side line of the road where the own vehicle is located is the reference line, the coordinate origin O F is the point with the shortest distance from the reference line to the characteristic point of the own vehicle, the horizontal axis X F is parallel to the direction of the road guide line, and the vertical axis Y F and the direction of the road guideline follow the left-hand or right-hand rule.
  • determining the characteristic points of the environmental target according to the position of each corner of the environmental target includes: selecting, from each corner of the environmental target, a lateral nearest point of the fixed area, the fixed area, The farthest point in the horizontal direction and the closest point in the longitudinal direction of the fixed area; calculate all the intersections of each line segment consisting of two adjacent corner points of the environmental target with the lane lines on both sides of the lane where the vehicle is located, The point closest to the coordinate origin O F in the longitudinal direction is the 3.75 feature point of the fixed area; calculating all the intersections of each line segment consisting of two adjacent corner points connecting the environmental target with the two virtual lines, respectively, Using the point closest to the coordinate origin O F in all intersections as the 2.8 feature point of the fixed area; and calculating each line segment consisting of two adjacent corner points connecting the environmental target with the follower, respectively All regions of the intersections, the intersections of all the coordinate origin O F from the nearest point as the longitudinal follower region feature points.
  • the method further includes: acquiring coordinates of each feature point of the environmental target in a driving coordinate system.
  • the method includes determining the coordinates of any feature point of the environmental target in the driving coordinate system according to the following steps: determining a shortest distance to the any feature point on the reference line and the A point on the reference line corresponding to the shortest distance; using the size of the shortest distance as the ordinate size of the any feature point in the driving coordinate system; determining the distance from the reference line to the any longitudinal arc length between the feature points of the shortest distance point and the coordinate origin O F; and the longitudinal size of the arc length as the abscissa of any size at a feature point of the road coordinate system, wherein according to any one of the feature point in the road coordinate system to determine the subphase any feature point of the ordinate and the abscissa position of the origin O F negative for the coordinates.
  • the extension of the centerline of the following area passes through the front center point of the own vehicle, the length of the following area ranges from 150m to 250m, and / or the width of the following area ranges from 2.2 m to 3.4m.
  • the predetermined lateral distance ranges from 0.445m to 0.505m.
  • the method for extracting feature points of an environmental target according to the present invention has the following advantages:
  • the problem of inaccurate target recognition can be solved, and it is especially suitable for the recognition of larger targets (for example, trucks, etc.).
  • By accurately extracting the feature points of environmental targets that the decision system pays more attention to it reduces the information requirements of the decision system on the target and simplifies the calculation of the decision system.
  • the decision-making system performs horizontal and vertical control according to the extracted target feature points, which makes the implementation of the control strategy simpler, and the control accuracy and control effect will be more ideal.
  • Another object of the present invention is to provide a device for extracting feature points of an environmental target.
  • the device includes: a memory and a processor.
  • the memory stores instructions, and the instructions are used to enable the processor to The above method for extracting feature points of an environmental target is performed.
  • the device for extracting feature points of an environmental target has the same advantages as the above-mentioned method for extracting feature points of an environmental target over the prior art, and is not repeated here.
  • Another object of the present invention is to provide a machine-readable storage medium having instructions stored on the machine-readable storage medium, which are used to enable a machine to execute the above-mentioned method for extracting feature points of an environmental target.
  • FIG. 1 is a schematic flowchart of a method for extracting feature points of an environmental target according to an embodiment of the present invention
  • FIG. 2 shows a relative schematic diagram of the vehicle and environmental targets
  • Figure 3 shows a schematic diagram of a driving coordinate system
  • FIG. 4 shows a schematic diagram of determining an origin of a driving coordinate system
  • FIG. 5 shows a schematic diagram of 3.75 feature points in a fixed area
  • Figure 6 shows a schematic diagram of 2.8 feature points in a fixed area
  • FIG. 7 shows a schematic diagram of feature points of a follow-up area
  • FIG. 8 shows a schematic diagram of calculating coordinates of any feature point of an environmental target in a driving coordinate system
  • FIG. 9 shows a structural block diagram of an apparatus for extracting feature points of an environmental target according to an embodiment of the present invention.
  • FIG. 1 is a schematic flowchart of a method for extracting feature points of an environmental target according to an embodiment of the present invention.
  • an embodiment of the present invention provides a method for extracting feature points of an environmental target.
  • the method can be applied to any vehicle, for example, it can be applied to an autonomous vehicle.
  • the method may include: step S110, obtaining a position of each corner point of the environmental target; and step S120, determining a feature point of the environment target according to the position of each corner point of the environmental target.
  • the “environmental target” may refer to any moving or stationary object around the vehicle, for example, a vehicle, a person, a building, etc. In the embodiment of the present invention, the “environmental target” is used as an example for the vehicle.
  • FIG. 2 shows the relative schematic diagram of the host vehicle and environmental targets.
  • the environmental target is a vehicle
  • the environmental target is in the left lane of the lane in which the vehicle is traveling.
  • the sensors on the vehicle can detect the position of the rear end center point A of the environmental target, the angle between the centerline of the environmental target and the horizontal or vertical axis of the vehicle, the length of the environmental target, and the width of the environmental target.
  • the output information of the sensors on the vehicle is usually given based on the vehicle coordinate system based on the vehicle.
  • the vehicle's standard system X H O H Y H O H usually selects the vehicle center point, and X H points to the direction of the vehicle's longitudinal axis.
  • Y H points to the direction of the vehicle's horizontal axis, and follows the right-hand rule, which is positive counterclockwise.
  • the position of the rear center point A is the coordinates of the rear center point A in the vehicle coordinate system.
  • the vehicle sensor outputs the angle ⁇ between the center line of the environmental target and the longitudinal axis of the vehicle (ie, the X H axis).
  • the coordinates of the rear end center point A of the environmental target in the vehicle coordinate system, the angle ⁇ between the center line of the environmental target and the longitudinal axis of the vehicle, the length of the environmental target, and the width of the environmental target can calculate each angle of the environmental target The position coordinates of the point in the vehicle coordinate system.
  • what is acquired in the above step S110 may be the position coordinates of each corner point of the environmental target in the vehicle coordinate system. Since the vehicle coordinate system and the global coordinate system can be converted to each other, the position coordinates of each corner point of the obtained environmental target in the vehicle coordinate system can also be converted to the position coordinates in the global coordinate system. What is obtained in the above step S110 may be the position coordinates of each corner point of the environmental target in the global coordinate system, which is not specifically limited in this embodiment of the present invention.
  • FIG. 3 shows a schematic diagram of a driving coordinate system.
  • X F O F Y F represents a driving coordinate system
  • X H O H Y H is a vehicle coordinate system
  • X G O G Y G is a global coordinate system.
  • the global coordinate system X G O G Y G is based on the geodetic coordinate system, X G points north, Y G points east, the angle direction is clockwise positive, and the angle range is [0,360 °].
  • the map lane line information is usually Given based on the global coordinate system.
  • the vehicle coordinate system X H O H Y H is based on the own vehicle.
  • the coordinate origin O H is usually the vehicle center point.
  • X H points to the longitudinal axis of the vehicle and Y H points to the transverse axis of the vehicle. It follows the right-hand rule and it is positive counterclockwise. .
  • the output information of sensors such as camera, lidar, millimeter wave radar on the vehicle is usually given based on the vehicle coordinate system.
  • the driving coordinate system X F O F Y F according to the embodiment of the present invention is based on the road sideline.
  • the road sideline can be the leftmost lane sideline or the rightmost lane sideline of the road on which the vehicle is located.
  • the coordinate origin O F is at The point with the shortest distance from the reference line to the fourth characteristic point of the vehicle, the horizontal axis X F may be parallel to the direction of the road guide line, for example, the horizontal axis X F may point to the direction of the road guide line or may be guided away from the road Line direction.
  • the vertical axis Y F and the direction of the road guideline follow the left-hand or right-hand rule.
  • the fourth feature point may be any point of the vehicle.
  • the fourth feature point may be a vehicle center point, that is, the fourth feature point may coincide with the origin of the vehicle coordinate system.
  • FIG. 2 shows a schematic diagram of the vertical axis Y F and the direction of the road guide line following the left-hand rule.
  • the embodiment of the present invention also mainly uses the vertical axis Y F and the direction of the road guide line to follow the left-hand rule as an example.
  • the vertical axis Y The situation where F and the road guideline follow the right-hand rule is similar, and will not be described again.
  • Lane lines and environmental targets can have both the vehicle coordinate system and the driving coordinate system.
  • FIG. 4 shows a schematic diagram of determining an origin of a driving coordinate system.
  • the leftmost lane line of the vehicle in FIG. 4 is determined as the reference line, and the reference line of the driving coordinate system X F O F Y F is divided into a plurality of points (reference line points shown in FIG. 4), each adjacent two The points may have the same predetermined distance, and the predetermined distance may be arbitrarily set as required, for example, the range of the predetermined distance may be 0.05m to 0.3m.
  • the point corresponding to the shortest distance on the reference line is taken as the origin O F of the driving coordinate system. With the movement of the vehicle, driving the coordinate system origin position O F are changing.
  • the feature points of the environmental target in the embodiments of the present invention are mainly described with respect to the driving coordinate system.
  • the feature points of the environmental target may include fixed area feature points and / or follow-up area feature points.
  • the fixed area feature points may include one or more of the following: a fixed area lateral nearest point, which refers to a point on the environmental target that is laterally closest to the reference line of the driving coordinate system X F O F Y F ; The farthest point in the horizontal direction of the fixed area.
  • the farthest point in the fixed area refers to the horizontally furthest point on the environmental target from the reference line of the driving coordinate system X F O F Y F.
  • the closest point refers to the longitudinal closest point on the environmental target to the coordinate origin O F of the driving coordinate system X F O F Y F ;
  • the fixed area 3.75 characteristic point, the fixed area 3.75 characteristic point refers to the environmental target and the local
  • the intersection of the lane lines on both sides of the lane where the car is located is the closest point longitudinally from the coordinate origin O F ;
  • the fixed area 2.8 feature point refers to the two virtual lines in the lane where the vehicle is located the intersection points from the coordinate origin O F longitudinal nearest point, the imaginary line is parallel to the two sides of the lane where the vehicle lane line and each side of the lane lines on both sides with a predetermined lane line distance in transverse Virtual line distance.
  • the environmental target is a vehicle
  • the environmental target has four corner points, and according to the definition of the feature points, a fixed area horizontal nearest point, a fixed area horizontal furthest point, and a fixed area vertical can be selected from the four corner points of the environmental target. Closest point, of which the vertical closest point in the fixed area is the closest feature point to the vehicle, and it is also a feature point that the autonomous vehicle is more concerned about, especially when the target vehicle is located in the area directly in front of the vehicle.
  • the decision system controls the speed, acceleration, and lane of the vehicle.
  • Figure 5 shows a schematic diagram of the 3.75 feature points of the fixed area.
  • the characteristic point of the fixed area 3.75 is a point B longitudinally closest to the coordinate origin O F among intersections of the environmental target and the lane lines on both sides of the lane where the vehicle is located.
  • all intersections of each line segment consisting of two adjacent corner points of the environmental target and all lane lines on both sides of the lane where the vehicle is located can be calculated, for example, in the vehicle coordinate system, the equations of the line segments and the equations of the lane lines are linked together, and the intersection coordinates are obtained by solving the equations.
  • Driving select from the coordinate system X F O F Y F from all the intersections of the longitudinal coordinate origin O F B nearest point as the fixed point feature area 3.75.
  • FIG. 6 shows a schematic diagram of the feature points of the fixed area 2.8.
  • the fixed point 2.8 feature point refers to the coordinate origin X of the driving coordinate system X F O F Y F at the intersection of the environmental target and two virtual lines L2.8 and R2.8 in the lane where the vehicle is located.
  • F vertical closest point C.
  • the virtual line L2.8 is a virtual line parallel to the lane lines on both sides of the lane where the vehicle is located and a predetermined lateral distance from the left lane line of the lane where the vehicle is located.
  • the virtual line R2.8 is a virtual line that is parallel to the lane lines on both sides of the lane where the own vehicle is located and a predetermined lateral distance from the right lane line of the lane where the own vehicle is located.
  • the range of the predetermined lateral distance may be 0.445m to 0.505m.
  • the predetermined distance may be 0.475m.
  • the lane width is 3.75m, and the width between two virtual lines L2.8 and R2.8 is 2.8m when the predetermined distance is 0.475m.
  • all intersections of each line segment consisting of each adjacent two corner points of the environmental target and two virtual lines L2.8 and R2.8 may be calculated, for example,
  • the equations of the line segments and the equations of the virtual lines can be linked together, and the intersection coordinates can be obtained by solving the equations.
  • Driving then select from the coordinate system X F O F Y F from all the intersections of the longitudinal coordinate origin O F C nearest point as the fixed point feature area 2.8.
  • the feature points of the target vehicle may only include the fixed-area lateral closest point, the fixed-area lateral furthest point, and the fixed-area longitudinal closest point, without including the fixed-area 3.75 feature points and / or the fixed-area 2.8 feature points. If it is calculated that the target vehicle has a fixed area 3.75 feature point and / or a fixed area 2.8 feature point, it means that a part of the target vehicle is located in the area in front of the lane where the vehicle is located.
  • the decision system makes a lane change decision, The positions of the 3.75 feature points in the fixed area and / or the 2.8 feature points in the fixed area are taken into account in order to arrive at a more accurate and safer lane change plan.
  • the decision system will more easily judge the important level of the target, making the entire system more secure and stable.
  • FIG. 7 shows a schematic diagram of feature points of a follow-up area.
  • the following area is an area that is located in front of the vehicle in the driving direction (for example, it may be directly in front) and is parallel to the road direction and has a predetermined width. Center point.
  • FIG. 7 is a schematic diagram of a follow-up area under a straight road. Under a curve condition, the follow-up area parallel to the road direction is also curved.
  • the two virtual following area lines FL2.8 and FR2.8 of the following area can be obtained by shifting the virtual line passing through the front center point of the vehicle and parallel to the lane line by a predetermined distance to the left and right,
  • the range of the predetermined distance may be, for example, 1.1 m to 1.7 m, and may be, for example, 1.4 m.
  • the length of the following area may be 150m to 250m, and / or the width of the following area may be 2.2m to 3.4m.
  • the width range of the following area may preferably be greater than or equal to the vehicle width.
  • the following area moves with the movement of the vehicle, but is always located in front of the vehicle. For example, when the vehicle changes lanes laterally, the following area is still located in front of the direction of travel of the vehicle.
  • the target vehicle has the characteristic points of the follow-up area, it can be determined that at least a part of the target vehicle is located in the follow-up area.
  • the speed and acceleration of the lane change should be adjusted in real time according to the position of the follow-up area feature points. Wait.
  • the method for extracting feature points of an environmental target may further include obtaining coordinates of each feature point of the environmental target in a driving coordinate system.
  • FIG. 8 shows a schematic diagram of calculating the coordinates of any feature point of an environmental target in a driving coordinate system.
  • FIG. 8 illustrates that the horizontal axis X F of the driving coordinate system points to the direction of the road guide line, and the vertical axis Y F and the direction of the road guide line follow the left-hand rule as an example.
  • the coordinates of any feature point M on the environmental target in the vehicle coordinate system have been calculated.
  • the shortest distance from the reference line to the point M in the driving coordinate system For example, you can traverse forward or backward from the origin of the driving coordinate system on the reference line. The forward or backward traversal depends on whether the point M is in front of or behind the vehicle center point. This can simply be done by passing the point M in the vehicle coordinate system. The coordinates are determined. If the point M is in front of the center point of the host vehicle, the points on the reference line are traversed forward from the origin of the driving coordinate system on the reference line to determine the shortest distance to the point M.
  • each point on the reference line is traversed backward from the origin of the driving coordinate system on the reference line to determine the shortest distance to the point M and the point N corresponding to the shortest distance on the reference line.
  • the magnitude of the shortest distance can be taken as the magnitude of the ordinate of the point M in the driving coordinate system.
  • the positive and negative ordinate of the point M in the driving coordinate system can be determined according to the position of the point M relative to the reference line or coordinate origin O F. If the point M is positive to the right of the reference line or coordinate origin O F , the left The side is negative.
  • the size of the longitudinal size of the arc length between the arc length of the longitudinal size of the point M in the road coordinate system abscissa point between the reference line and the coordinate origin O F N, the reference line and the point coordinate origin O F N the distance between can be divided between the reference line and the point coordinate origin O F N the dots accumulated calculated.
  • a negative point in the road coordinate system with respect to the ordinate location can coordinate origin O F is determined in accordance with point M, if M was in front of the point coordinate origin O F of positive, negative, compared with the rear.
  • the determination of the coordinates of any characteristic point of the environmental target in the driving coordinate system is similar to the determination method of the coordinates of the point M in the driving coordinate, which will not be repeated here.
  • the range of use of the target feature point information is extended, so that it can be used in both straight and curved conditions.
  • the distance calculated in the driving coordinate system is the arc length rather than the straight line distance under the curve conditions, which improves the accuracy of the distance information of the characteristic points of the environmental target under special conditions.
  • an embodiment of the present invention further provides a machine-readable storage medium, where the machine-readable storage medium stores instructions, and the instructions are used to enable a machine to execute the foregoing method for extracting feature points of an environmental target.
  • the machine-readable storage medium may be, for example, various programs that can store programs such as a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disk.
  • the medium of the code is, for example, various programs that can store programs such as a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disk.
  • FIG. 9 shows a structural block diagram of an apparatus for extracting feature points of an environmental target according to an embodiment of the present invention.
  • an embodiment of the present invention further provides a device for extracting feature points of an environmental target.
  • the device may include a memory 910 and a processor 920.
  • the memory 910 may store an instruction that causes the processor 920 can execute a method for extracting feature points of an environmental target according to any embodiment of the present invention.
  • the processor 920 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), ready-made Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • CPU central processing unit
  • DSP digital signal processors
  • ASIC application specific integrated circuits
  • FPGA Field-Programmable Gate Array
  • the memory 910 may be configured to store the computer program instructions, and the processor implements the data fusion device for a vehicle sensor by running or executing computer program instructions stored in the memory, and calling data stored in the memory.
  • the memory 910 may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, an internal memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, Flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
  • a non-volatile memory such as a hard disk, an internal memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, Flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
  • the program is stored in a storage medium and includes several instructions to make a single chip, chip or processor (processor) executes all or part of the steps of the method described in each embodiment of the present application.
  • the foregoing storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks and other media that can store program codes .

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Mathematical Physics (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Traffic Control Systems (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
  • Navigation (AREA)

Abstract

一种用于提取环境目标的特征点的方法及装置，属于车辆领域。所述方法包括：获取所述环境目标的每个角点的位置（S110）；以及根据所述环境目标的每个角点的位置而确定所述环境目标的特征点（S120）。其可以解决目标识别不精确的问题，尤其适用于针对较大目标的识别。

Description

用于提取环境目标的特征点的方法及装置 技术领域
本发明涉及车辆领域，具体地，涉及一种用于提取环境目标的特征点的方法及装置。
背景技术
自动驾驶车辆在行驶过程中需要利用安装在车身上的传感器来获取周围环境目标的信息，并结合周围环境车道线信息对目标进行区域划分，以进一步得到目标的区域属性。不同区域的目标可为纵向控制和换道决策使用，同时车道线信息可以为车辆提供行驶路径信息，用于横向控制进行换道和车道保持。
由于道路环境中存在较大环境目标(如大货车等)，它的长度、宽度尺寸都非常大，因此无法用目标的一个中心点属性表示，这将造成车辆横、纵向控制效果欠佳，严重情况下会造成交通事故。因此需要更全面的提取障碍物的关注点信息，去掉一些不相关的目标属性，简化决策系统对目标选择的难度，同时使整个系统控制更加精准和安全。
发明内容
有鉴于此，本发明旨在提出一种用于提取环境目标的特征点的方法，用于至少解决对于环境目标的特征点的提取。
为了实现上述目的，本发明的技术方案是这样实现的：
一种用于提取环境目标的特征点的方法，所述方法包括：获取所述环境目标的每个角点的位置；以及根据所述环境目标的每个角点的位置而确定所述环境目标的特征点。
进一步的，所述环境目标的特征点包括固定区域特征点和/或随动区域特征点，所述固定区域特征点包括以下一者或多者：固定区域横向最近点，该固定区域横向最近点是指所述环境目标上距离行车坐标系X FO FY F的基准线横向最近的点；固定区域横向最远点，该固定区域横向最远点是指所述环境目标上距离基 准线横向最远的点；固定区域纵向最近点，所述固定区域纵向最近点是指所述环境目标上距离行车坐标系X FO FY F的坐标原点O F纵向最近的点；固定区域3.75特征点，所述固定区域3.75特征点是指所述环境目标与本车所在车道的两侧车道线的交点中距离所述坐标原点O F纵向最近的点；固定区域2.8特征点，所述固定区域2.8特征点是指所述环境目标与本车所在车道内的两条虚拟线的交点中距离所述坐标原点O F纵向最近的点，所述两条虚拟线是与本车所在车道的两侧车道线平行且分别与两侧车道线中的一侧车道线距离预定横向距离的虚拟线；所述随动区域为位于本车行驶方向的前方、与道路方向平行且具有预定宽度的区域，所述随动区域特征点是指所述环境目标与所述随动区域的交点中距离所述坐标原点O F纵向最近的点；其中所述行车坐标系X FO FY F以所述本车的所在道路的边线为基准线，坐标原点O F为在所述基准线上到所述本车的特征点距离最短的点，横轴X F与道路引导线方向平行，纵轴Y F与道路引导线方向遵循左手或右手定则。
进一步的，根据所述环境目标的每个角点的位置而确定所述环境目标的特征点包括：从所述环境目标的每个角点中选取所述固定区域横向最近点、所述固定区域横向最远点、所述固定区域纵向最近点；计算所述环境目标的相邻两个角点组成的每一线段分别与本车所在车道的两侧车道线的所有交点，将该所有交点中距离所述坐标原点O F纵向最近的点作为所述固定区域3.75特征点；计算连接所述环境目标的相邻两个角点组成的每一线段分别与所述两条虚拟线的所有交点，将该所有交点中距离所述坐标原点O F纵向最近的点作为所述固定区域2.8特征点；以及计算连接所述环境目标的相邻两个角点组成的每一线段分别与所述随动区域的所有交点，将该所有交点中距离所述坐标原点O F纵向最近的点作为所述随动区域特征点。
进一步的，所述方法还包括：获取所述环境目标的每个特征点在行车坐标系下的坐标。
进一步的，所述方法包括根据以下步骤来确定所述环境目标的任一特征点在所述行车坐标系下的坐标：确定在所述基准线上到所述任一特征点的最短距离以及所述基准线上与该最短距离对应的点；将所述最短距离的大小作为所述任一 特征点在所述行车坐标系下的纵坐标大小；确定在所述基准线上到所述任一特征点距离最短的点与所述坐标原点O F之间的纵向弧长；以及将所述纵向弧长的大小作为所述任一特征点在所述行车坐标系下的横坐标的大小，其中根据所述任一特征点在所述行车坐标系下相对于所述坐标原点O F的位置来确定所述任一特征点纵坐标和横坐标的正负。
进一步的，所述随动区域的中心线的延长线过所述本车的前端中心点，所述随动区域的长度范围为150m至250m，和/或所述随动区域的宽度范围为2.2m至3.4m。
进一步的，所述预定横向距离的范围为0.445m至0.505m。
相对于现有技术，本发明所述的用于提取环境目标的特征点的方法具有以下优势：
通过特征点的选取，可以解决目标识别不精确的问题，尤其适用于针对较大目标(例如，卡车等)的识别。通过准确的提取决策系统更加关注的环境目标的特征点，减少了决策系统对目标的信息需求，简化决策系统的计算量。决策系统根据所提取的目标特征点进行横纵向控制，使得控制策略实现更加简单，控制精度和控制效果会更加理想。
本发明的另一目的在于提出一种用于提取环境目标的特征点的装置，所述装置包括：存储器和处理器，所述存储器中存储有指令，所述指令用于使得所述处理器能够执行上述的用于提取环境目标的特征点的方法。
所述的用于提取环境目标的特征点的装置与上述的用于提取环境目标的特征点的方法相对于现有技术所具有的优势相同，在此不再赘述。
本发明的另一目的在于提出一种机器可读存储介质，所述机器可读存储介质上存储有指令，所述指令用于使得机器能够执行上述的用于提取环境目标的特征点的方法。
本发明实施例的其它特征和优点将在随后的具体实施方式部分予以详细说明。
附图说明
附图是用来提供对本发明实施例的进一步理解，并且构成说明书的一部分， 与下面的具体实施方式一起用于解释本发明实施例，但并不构成对本发明实施例的限制。在附图中：
图1示出了根据本发明一实施例的用于提取环境目标的特征点的方法的流程示意图；
图2示出了本车与环境目标的相对示意图；
图3示出了行车坐标系的一示意图；
图4示出了行车坐标系的原点确定示意图；
图5示出了固定区域3.75特征点的示意图；
图6示出了固定区域2.8特征点的示意图；
图7示出了随动区域特征点的示意图；
图8示出了计算环境目标的任一特征点在行车坐标系下的坐标的示意图；以及
图9示出了根据本发明一实施例的用于提取环境目标的特征点的装置的结构框图。
附图标记说明
910      存储器               920   处理器
具体实施方式
以下结合附图对本发明实施例的具体实施方式进行详细说明。应当理解的是，此处所描述的具体实施方式仅用于说明和解释本发明实施例，并不用于限制本发明实施例。
图1示出了根据本发明一实施例的用于提取环境目标的特征点的方法的流程示意图。如图1所示，本发明实施例提供一种用于提取环境目标的特征点的方法，所述方法可以适用于任意车辆，例如可适用于自动驾驶车辆等。所述方法可以包括：步骤S110，获取所述环境目标的每个角点的位置；以及步骤S120，根据所述环境目标的每个角点的位置而确定所述环境目标的特征点。“环境目标”可以指处于车辆周围的移动的或静止的任意物体，例如，车辆、人、建筑物等，本发明实施例是以“环境目标”为车辆进行举例说明的。
图2示出了本车与环境目标的相对示意图。如图2所示，环境目标为一车辆，且环境目标处于本车所行驶车道的左侧车道。本车上的传感器可以检测出环境目标的后端中心点A的位置、环境目标的中心线与本车横轴或纵轴的夹角、环境目标的长度和环境目标的宽度。本车上的传感器的输出信息通常是基于以本车为基准的车辆坐标系给出的，车辆做标系X HO HY H的O H通常选取车辆中心点，X H指向车辆纵轴方向，Y H指向车辆横轴方向，遵从右手定则，逆时针为正。后端中心点A的位置为后端中心点A在车辆坐标系下的坐标，以车辆传感器输出的是环境目标的中心线与本车纵轴(即X H轴)的夹角α，则使用环境目标的后端中心点A在车辆坐标系下的坐标、环境目标的中心线与本车纵轴的夹角α、环境目标的长度和环境目标的宽度就可以计算出环境目标的每个角点在车辆坐标系下的位置坐标。
也就是说，在上述步骤S110中所获取的可以是环境目标的每个角点在车辆坐标系下的位置坐标。由于车辆坐标系与全局坐标系可以互相转换，也可以将所获取的环境目标的每个角点在车辆坐标系下的位置坐标转换到全局坐标系下的位置坐标。在上述步骤S110中所获取的可以是环境目标的每个角点在全局坐标系下的位置坐标，本发明实施例对此并不进行特定限制。
下面首先结合图3和图4对本发明实施例中提到的行车坐标系进行介绍。图3示出了行车坐标系的一示意图。图3中X FO FY F表示行车坐标系，X HO HY H为车辆坐标系，X GO GY G为全局坐标系。其中全局坐标系X GO GY G是以大地坐标系为基准，X G指向北，Y G指向东，角度方向顺时针为正，角度范围[0,360°]，其中，地图车道线信息通常是基于全局坐标系给出。车辆坐标系X HO HY H以本车为基准，坐标原点O H通常选取车辆中心点，X H指向车辆纵轴方向，Y H指向车辆横轴方向，遵从右手定则，逆时针为正。车辆上的诸如摄像头、激光雷达、毫米波雷达等传感器的输出信息通常是基于车辆坐标系给出。本发明实施例提出的行车坐标系X FO FY F是以道路的边线为基准，道路的边线可以是车辆所在道路的最左侧车道边线或最右侧车道边线，坐标原点O F为在所述基准线上到所述车辆的 第四特征点距离最短的点，横轴X F可以与道路的引导线方向平行，例如，横轴X F可以指向道路引导线方向或者可以背向道路引导线方向。纵轴Y F与道路引导线方向遵循左手或右手定则。这里，第四特征点可以是车辆的任意一点，例如第四特征点可以选取车辆中心点，也就是说，第四特征点可以与车辆坐标系的原点重合。
在道路是笔直的情况下，横轴X F与笔直的道路引导线方向平行，在道路是弯曲的情况下，如车辆拐弯过程中，横轴X F与弯曲的道路引导线方向平行，即，横轴X F与道路引导线方向始终保持一致。图2中示出了纵轴Y F与道路引导线方向遵循左手定则的示意图，本发明实施例也主要以纵轴Y F与道路引导线方向遵循左手定则为例进行说明，纵轴Y F与道路引导线方向遵循右手定则的情况与之类似，将不再赘述。车道线和环境目标可同时拥有车辆坐标系和行车坐标系双重属性。
图4示出了行车坐标系的原点确定示意图。图4中车辆最左侧车道线被确定为基准线，行车坐标系X FO FY F的基准线被划分为多个点(图4中示出的基准线点)，每相邻两个点之间可以具有相同的预定距离，该预定距离可以根据需要而被任意设置，例如预定距离的范围可以是0.05m至0.3m。在车辆坐标系下寻找基准线上到车辆中心点的最短距离，例如，可以计算基准线上所划分的每个点或者预定范围的每个点与车辆中心的欧式距离，以寻找到车辆中心的最短距离。将基准线上与该最短距离对应的点作为行车坐标系的原点O F。随着车辆的移动，行车坐标系的原点位置O F也在不断变化。
本发明实施例中环境目标的特征点主要是相对于行车坐标系进行说明的。环境目标的特征点可以包括固定区域特征点和/或随动区域特征点。
固定区域特征点可以包括以下一者或多者：固定区域横向最近点，该固定区域横向最近点是指所述环境目标上距离行车坐标系X FO FY F的基准线横向最近的点；固定区域横向最远点，该固定区域横向最远点是指所述环境目标上距离行车坐标系X FO FY F的基准线横向最远的点；固定区域纵向最近点，该固定区域 纵向最近点是指所述环境目标上距离行车坐标系X FO FY F的坐标原点O F纵向最近的点；固定区域3.75特征点，所述固定区域3.75特征点是指所述环境目标与本车所在车道的两侧车道线的交点中距离所述坐标原点O F纵向最近的点；固定区域2.8特征点，所述固定区域2.8特征点是指与本车所在车道内的两条虚拟线的交点中距离所述坐标原点O F纵向最近的点，所述两条虚拟线是与本车所在车道的两侧车道线平行且分别与两侧车道线中的一侧车道线距离预定横向距离的虚拟线。
在环境目标为车辆的情况下，环境目标具有四个角点，可以根据特征点的定义从环境目标的四个角点中选取出固定区域横向最近点、固定区域横向最远点、固定区域纵向最近点，其中固定区域纵向最近点是距离本车最近的特征点，也是自动驾驶车辆比较关注的特征点，尤其在目标车辆位于本车正前方区域的情况下，通过获取固定区域最近点可以帮助决策系统控制本车的速度、加速度和行驶车道等。
图5示出了固定区域3.75特征点的示意图。如图5所示，固定区域3.75特征点是环境目标与本车所在车道的两侧车道线的交点中距离所述坐标原点O F纵向最近的点B。可选地，可以在车辆坐标系或全局坐标系下，计算环境目标的每相邻两个角点组成的每一线段分别与本车所在车道的两侧车道线的所有交点，例如，可以在车辆坐标系下，将线段的方程和车道线的方程联立，通过解方程组得到交点坐标。然后从所有交点中选择距离行车坐标系X FO FY F的坐标原点O F纵向最近的点B作为固定区域3.75特征点。
图6示出了固定区域2.8特征点的示意图。如图6所示，固定区域2.8特征点是指环境目标与本车所在车道内的两条虚拟线L2.8和R2.8的交点中距离行车坐标系X FO FY F的坐标原点O F纵向最近的点C。虚拟线L2.8是与本车所在车道的两侧车道线平行，且与本车所在车道的左侧车道线距离预定横向距离的虚拟线。虚拟线R2.8是与本车所在车道的两侧车道线平行，且与本车所在车道的右侧车道线距离预定横向距离的虚拟线。所述预定横向距离的范围可以是0.445m至0.505m，例如预定距离可以选用0.475m。通常，车道宽度为3.75m，则在预定距离为0.475m的情况下，两条虚拟线L2.8和R2.8之间的宽度为2.8m。可选地， 可以在车辆坐标系或全局坐标系下，计算环境目标的每相邻两个角点组成的每一线段分别与两条虚拟线L2.8和R2.8的所有交点，例如，可以在车辆坐标系下，将线段的方程和虚拟线的方程联立，通过解方程组得到交点坐标。然后从所有交点中选择距离行车坐标系X FO FY F的坐标原点O F纵向最近的点C作为固定区域2.8特征点。
在一些情况下，目标车辆的特征点可能仅包括固定区域横向最近点、固定区域横向最远点、固定区域纵向最近点，而不包括固定区域3.75特征点和/或固定区域2.8特征点。如果计算出目标车辆具有固定区域3.75特征点和/或固定区域2.8特征点，则说明目标车辆有一部分是位于本车所在车道的前方区域内，则决策系统在进行换道决策的时候，可以将固定区域3.75特征点和/或固定区域2.8特征点的位置考虑进去，以便得出更准确安全的换道方案。此外，根据固定区域3.75特征点和固定区域2.8特征点这些特征点的求解，决策系统将更加容易判断目标的重要等级，使整个系统更加安全稳定。
图7示出了随动区域特征点的示意图。如图7所示，随动区域为位于车辆行驶方向的前方(例如，可以是正前方)、与道路方向平行且具有预定宽度的区域，随动区域的中心线的延长线可以过本车的前端中心点。在图7中示出的是笔直道路下随动区域的示意图，在弯道工况下，与道路方向的随动区域平行也是弯曲的。随动区域的两条虚拟随动区域线FL2.8和FR2.8可以是将过本车的前端中心点并与车道线平行的虚拟线向左和向右各偏移预定距离而得到的，该预定距离的范围例如可以是1.1m至1.7m，例如可以是1.4m。随动区域的长度范围可以是150m至250m，和/或所述随动区域的宽度范围可以是2.2m至3.4m。随动区域的宽度范围优选可以大于或等于车辆宽度。随动区域随着车辆的移动而移动，但始终位于车辆前方，例如在车辆横向换道时，随动区域仍然位于车辆行驶方向的前方。
随动区域特征点是指环境目标与随动区域的交点中距离行车坐标系X FO FY F的坐标原点O F纵向最近的点D。可选地，可以在车辆坐标系或全局坐标系下，计算环境目标的每相邻两个角点组成的每一线段分别与两条虚拟线FL2.8和FR2.8的所有交点。例如，可以在车辆坐标系下，将线段的方程和虚拟线的方程联立，通过解方程组得到交点坐标。然后从所有交点中选择距离行车坐 标系X FO FY F的坐标原点O F纵向最近的点D作为随动区域特征点。
如果计算出目标车辆具有随动区域特征点，则可以确定目标车辆至少有一部分位于随动区域内，在车辆换道过程中，应当根据随动区域特征点的位置实时调整换道的速度、加速度等。
进一步地，本发明实施例提供的用于提取环境目标的特征点的方法还可以包括获取环境目标的每个特征点在行车坐标系下的坐标。
图8示出了计算环境目标的任一特征点在行车坐标系下的坐标的示意图。图8以行车坐标系的横轴X F指向道路引导线方向、纵轴Y F与道路引导线方向遵循左手定则为例进行说明。环境目标上的任一特征点M在车辆坐标系下的坐标已计算得知。
在行车坐标系下寻找基准线上到点M的最短距离。例如可以从基准线上行车坐标系的原点开始向前或向后遍历，其中向前还是向后遍历取决于点M在车辆中心点的前方还是后方，这可以简单的通过点M在车辆坐标系下的坐标而确定出。如果点M在本车车辆中心点的前方，则从基准线上行车坐标系的原点开始向前遍历基准线上的各点以确定出到点M的最短距离。如果点M在车辆中心点的后方，则从基准线上行车坐标系的原点开始向后遍历基准线上的各点以确定出到点M最短距离以及基准线上与该最短距离对应的点N，该最短距离的大小可以作为点M在行车坐标系下的纵坐标的大小。点M在行车坐标系下的纵坐标的正负可以根据点M相对于基准线或坐标原点O F的位置来确定，如果点M在基准线或坐标原点O F的右侧则为正，左侧则为负。
点M在行车坐标系下的横坐标的大小为基准线上点N与坐标原点O F之间的纵向弧长的大小，基准线上点N与坐标原点O F之间的纵向弧长的大小可以使用基准线上点N与坐标原点O F之间的划分的点与点之间的距离累加计算得到。点A在行车坐标系下的纵坐标的正负可以根据点M相对于坐标原点O F的位置来确定，如果点M在坐标原点O F的前方则为正，后方则为负。
环境目标的任意特征点在行车坐标系下的坐标的确定与点M在行车坐标下的坐标的确定方式类似，这里将不再赘述。
通过计算出各特征点在行车坐标系下的坐标，拓展了目标特征点信息的使用范围，使得在直道工况和弯道工况下均可以使用。在行车坐标系下计算的距离 在弯道工况下是弧长而非直线距离，这提升了环境目标在特殊工况下特征点距离信息的准确度。
相应地，本发明实施例还提供一种机器可读存储介质，所述机器可读存储介质上存储有指令，所述指令用于使得机器能够执行上述的用于提取环境目标的特征点的方法。所述机器可读存储介质例如可以是U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
图9示出了根据本发明一实施例的用于提取环境目标的特征点的装置的结构框图。如图9所示，本发明实施例还提供一种用于提取环境目标的特征点的装置，所述装置可包括存储器910和处理器920，存储器910中可以存储有指令，该指令使得处理器920能够执行根据本发明任意实施例的用于提取环境目标的特征点的方法。
处理器920可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。
存储器910可用于存储所述计算机程序指令，所述处理器通过运行或执行存储在所述存储器内的计算机程序指令，以及调用存储在存储器内的数据，实现所述用于车辆传感器的数据融合装置的各种功能。存储器910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，例如硬盘、内存、插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)、至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。
本发明实施例提供的用于提取环境目标的特征点的装置的具体工作原理及益处与上述本发明实施例提供的用于提取环境目标的特征点的方法的具体工作原理及益处相类似，这里将不再赘述。
以上结合附图详细描述了本发明实施例的可选实施方式，但是，本发明实施例并不限于上述实施方式中的具体细节，在本发明实施例的技术构思范围内，可以对本发明实施例的技术方案进行多种简单变型，这些简单变型均属于本发明 实施例的保护范围。
另外需要说明的是，在上述具体实施方式中所描述的各个具体技术特征，在不矛盾的情况下，可以通过任何合适的方式进行组合。为了避免不必要的重复，本发明实施例对各种可能的组合方式不再另行说明。
本领域技术人员可以理解实现上述实施例方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成，该程序存储在一个存储介质中，包括若干指令用以使得单片机、芯片或处理器(processor)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
此外，本发明实施例的各种不同的实施方式之间也可以进行任意组合，只要其不违背本发明实施例的思想，其同样应当视为本发明实施例所公开的内容。

Claims (10)

  1. 一种用于提取环境目标的特征点的方法，其特征在于，所述方法包括：
    获取所述环境目标的每个角点的位置；以及
    根据所述环境目标的每个角点的位置而确定所述环境目标的特征点。
  2. 根据权利要求1所述的方法，其特征在于，
    所述环境目标的特征点包括固定区域特征点和/或随动区域特征点，所述固定区域特征点包括以下一者或多者：固定区域横向最近点，该固定区域横向最近点是指所述环境目标上距离行车坐标系X FO FY F的基准线横向最近的点；固定区域横向最远点，该固定区域横向最远点是指所述环境目标上距离基准线横向最远的点；固定区域纵向最近点，所述固定区域纵向最近点是指所述环境目标上距离行车坐标系X FO FY F的坐标原点O F纵向最近的点；固定区域3.75特征点，所述固定区域3.75特征点是指所述环境目标与本车所在车道的两侧车道线的交点中距离所述坐标原点O F纵向最近的点；固定区域2.8特征点，所述固定区域2.8特征点是指所述环境目标与本车所在车道内的两条虚拟线的交点中距离所述坐标原点O F纵向最近的点，所述两条虚拟线是与本车所在车道的两侧车道线平行且分别与两侧车道线中的一侧车道线距离预定横向距离的虚拟线；
    所述随动区域为位于本车行驶方向的前方、与道路方向平行且具有预定宽度的区域，所述随动区域特征点是指所述环境目标与所述随动区域的交点中距离所述坐标原点O F纵向最近的点；
    其中所述行车坐标系X FO FY F以所述本车的所在道路的边线为基准线，坐标原点O F为在所述基准线上到所述本车的特征点距离最短的点，横轴X F与道路引导线方向平行，纵轴Y F与道路引导线方向遵循左手或右手定则。
  3. 根据权利要求1或2所述方法，其特征在于，根据所述环境目标的每个角点的位置而确定所述环境目标的特征点包括：
    从所述环境目标的每个角点中选取所述固定区域横向最近点、所述固定区域横向最远点、所述固定区域纵向最近点；
    计算所述环境目标的相邻两个角点组成的每一线段分别与本车所在车道的两侧车道线的所有交点，将该所有交点中距离所述坐标原点O F纵向最近的点作为所述固定区域3.75特征点；
    计算连接所述环境目标的相邻两个角点组成的每一线段分别与所述两条虚拟线的所有交点，将该所有交点中距离所述坐标原点O F纵向最近的点作为所述固定区域2.8特征点；以及
    计算连接所述环境目标的相邻两个角点组成的每一线段分别与所述随动区域的所有交点，将该所有交点中距离所述坐标原点O F纵向最近的点作为所述随动区域特征点。
  4. 根据权利要求1至3中任一项权利要求所述的方法，其特征在于，所述方法还包括：
    获取所述环境目标的每个特征点在行车坐标系下的坐标。
  5. 根据权利要求4所述的方法，其特征在于，所述方法包括根据以下步骤来确定所述环境目标的任一特征点在所述行车坐标系下的坐标：
    确定在所述基准线上到所述任一特征点的最短距离以及所述基准线上与该最短距离对应的点；
    将所述最短距离的大小作为所述任一特征点在所述行车坐标系下的纵坐标大小；
    确定在所述基准线上到所述任一特征点距离最短的点与所述坐标原点O F之间的纵向弧长；以及
    将所述纵向弧长的大小作为所述任一特征点在所述行车坐标系下的横坐标的大小，
    其中根据所述任一特征点在所述行车坐标系下相对于所述坐标原点O F的位置来确定所述任一特征点纵坐标和横坐标的正负。
  6. 根据权利要求2或3所述的方法，其特征在于，
    所述随动区域的中心线的延长线过所述本车的前端中心点，所述随动区域的长度范围为150m至250m，和/或所述随动区域的宽度范围为2.2m至3.4m。
  7. 根据权利要求2或3所述的方法，其特征在于，所述预定横向距离的范围为0.445m至0.505m。
  8. 根据权利要求1至7中任一项所述的方法，其特征在于，所述获取所述环境目标的每个角点的位置包括：
    获取所述环境目标的后端中心点的位置、所述环境目标的中心线与本车横轴或纵轴的夹角、所述环境目标的长度和所述环境目标的宽度；以及
    使用所述后端中心点的位置、所述夹角、所述环境目标的长度和所述环境目标的宽度来分别计算所述环境目标的每个角点的位置。
  9. 一种用于确定车辆的环境目标所处的区域的装置，其特征在于，所述装置包括存储器和处理器，所述存储器中存储有指令，所述指令用于使得所述处理器能够执行根据权利要求1至8中任一项所述的用于提取环境目标的特征点的方法。
  10. 一种机器可读存储介质，其特征在于，所述机器可读存储介质上存储有指令，所述指令用于使得机器能够执行根据权利要求1至8中任一项所述的用于提取环境目标的特征点的方法。
PCT/CN2019/108399 2018-09-30 2019-09-27 用于提取环境目标的特征点的方法及装置 WO2020063813A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US17/281,560 US11928870B2 (en) 2018-09-30 2019-09-27 Method and apparatus used for extracting feature point of environmental target
KR1020217013022A KR102534412B1 (ko) 2018-09-30 2019-09-27 환경 목표의 특징점을 추출하는 방법 및 장치
JP2021517849A JP7299976B2 (ja) 2018-09-30 2019-09-27 環境ターゲットの特徴点抽出方法及び装置
EP19865835.3A EP3859594A4 (en) 2018-09-30 2019-09-27 METHOD AND DEVICE FOR EXTRACTION OF A FEATURE POINT OF AN ENVIRONMENTAL DESTINATION

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811154061.8 2018-09-30
CN201811154061.8A CN110378177B (zh) 2018-09-30 2018-09-30 用于提取环境目标的特征点的方法及装置

Publications (1)

Publication Number Publication Date
WO2020063813A1 true WO2020063813A1 (zh) 2020-04-02

Family

ID=68243466

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/108399 WO2020063813A1 (zh) 2018-09-30 2019-09-27 用于提取环境目标的特征点的方法及装置

Country Status (6)

Country Link
US (1) US11928870B2 (zh)
EP (1) EP3859594A4 (zh)
JP (1) JP7299976B2 (zh)
KR (1) KR102534412B1 (zh)
CN (1) CN110378177B (zh)
WO (1) WO2020063813A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112308836A (zh) * 2020-11-02 2021-02-02 创新奇智(上海)科技有限公司 角点检测方法、装置、电子设备及可读存储介质

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112799055B (zh) * 2020-12-28 2024-07-09 深圳承泰科技有限公司 一种探测被测车辆的方法、装置以及电子设备
CN114132325B (zh) * 2021-12-14 2024-03-01 京东鲲鹏(江苏)科技有限公司 车辆的行驶方法和装置
CN117111491B (zh) * 2023-10-25 2024-01-19 北京赛目科技股份有限公司 一种自动驾驶仿真中物体最近点确定方法、装置及设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103065151A (zh) * 2012-11-04 2013-04-24 北京工业大学 一种基于深度信息的车辆识别方法
CN105956527A (zh) * 2016-04-22 2016-09-21 百度在线网络技术（北京）有限公司 用于无人驾驶车的障碍物检测结果评估方法和装置
US20180150082A1 (en) * 2016-11-30 2018-05-31 Samsung Electronics Co., Ltd. Method and apparatus for generating autonomous driving route
CN108280840A (zh) * 2018-01-11 2018-07-13 武汉理工大学 一种基于三维激光雷达的道路实时分割方法

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102254318B (zh) * 2011-04-08 2013-01-09 上海交通大学 基于图像透视投影变换的车辆道路交通视频测速方法
KR101227569B1 (ko) 2011-05-26 2013-01-29 한국과학기술연구원 골프 스윙 분석을 위한 신체 부위별 위치 추적 장치 및 방법
KR101455236B1 (ko) * 2013-04-19 2014-11-07 주식회사 제이엠랩 이동형 증강현실 제공 장치 및 그 방법
US9963144B2 (en) 2014-08-29 2018-05-08 Nissan Motor Co., Ltd. Travel control device and travel control method
CN106697042B (zh) * 2015-11-17 2019-04-19 比亚迪股份有限公司 车辆、车辆的转向传动机构和车辆的转向系统
JP6564713B2 (ja) * 2016-02-01 2019-08-21 三菱重工業株式会社 自動運転制御装置、車両及び自動運転制御方法
JP6323473B2 (ja) 2016-02-25 2018-05-16 トヨタ自動車株式会社 走行制御装置
KR102528002B1 (ko) * 2016-08-17 2023-05-02 현대모비스 주식회사 탑뷰 영상 생성 장치 및 그 방법
CN106548486B (zh) * 2016-11-01 2024-02-27 浙江大学 一种基于稀疏视觉特征地图的无人车位置跟踪方法
WO2018098607A1 (en) * 2016-11-29 2018-06-07 SZ DJI Technology Co., Ltd. Method and system of adjusting image focus
CN108286976A (zh) * 2017-01-09 2018-07-17 北京四维图新科技股份有限公司 一种点云数据的融合方法和装置、以及混合导航系统
CN107895375B (zh) * 2017-11-23 2020-03-31 南京莱斯电子设备有限公司 基于视觉多特征的复杂道路线提取方法
CN110962856B (zh) * 2018-09-30 2021-07-27 毫末智行科技有限公司 用于确定车辆的环境目标所处的区域的方法及装置
CN111563474A (zh) * 2020-05-18 2020-08-21 北京茵沃汽车科技有限公司 运动背景下的基于车载鱼眼镜头的障碍物检测方法、系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103065151A (zh) * 2012-11-04 2013-04-24 北京工业大学 一种基于深度信息的车辆识别方法
CN105956527A (zh) * 2016-04-22 2016-09-21 百度在线网络技术（北京）有限公司 用于无人驾驶车的障碍物检测结果评估方法和装置
US20180150082A1 (en) * 2016-11-30 2018-05-31 Samsung Electronics Co., Ltd. Method and apparatus for generating autonomous driving route
CN108280840A (zh) * 2018-01-11 2018-07-13 武汉理工大学 一种基于三维激光雷达的道路实时分割方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3859594A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112308836A (zh) * 2020-11-02 2021-02-02 创新奇智(上海)科技有限公司 角点检测方法、装置、电子设备及可读存储介质
CN112308836B (zh) * 2020-11-02 2022-10-04 创新奇智(上海)科技有限公司 角点检测方法、装置、电子设备及可读存储介质

Also Published As

Publication number Publication date
KR20210069687A (ko) 2021-06-11
US11928870B2 (en) 2024-03-12
EP3859594A4 (en) 2021-11-24
JP2022502311A (ja) 2022-01-11
US20210365695A1 (en) 2021-11-25
KR102534412B1 (ko) 2023-05-26
EP3859594A1 (en) 2021-08-04
JP7299976B2 (ja) 2023-06-28
CN110378177B (zh) 2022-01-28
CN110378177A (zh) 2019-10-25

Similar Documents

Publication Publication Date Title
WO2020063813A1 (zh) 用于提取环境目标的特征点的方法及装置
CN111750886B (zh) 局部路径规划方法及装置
US12007243B2 (en) Traffic lane line fitting method and system
CN110687539B (zh) 一种车位检测方法、装置、介质和设备
CN108227695A (zh) 自动驾驶控制装置、包括该装置的系统及其方法
JP2019108116A (ja) 車両の縦方向速度を制御する装置及び方法
JP6654923B2 (ja) 地図情報出力装置
US11449058B2 (en) Traveling track determination processing and automated drive device
JP7192610B2 (ja) 走行支援方法および走行支援装置
JP6297482B2 (ja) 情報処理装置
CN115077557B (zh) 一种路口掉头的路径规划方法
US20190180117A1 (en) Roadside object recognition apparatus
JP7043765B2 (ja) 車両走行制御方法及び装置
US11347240B2 (en) Method and apparatus for determining path
CN110962856B (zh) 用于确定车辆的环境目标所处的区域的方法及装置
CN110962857B (zh) 用于确定车辆的环境目标所处的区域的方法及装置
JP5682302B2 (ja) 走行道路推定装置、方法およびプログラム
CN110531347A (zh) 激光雷达的探测方法、装置及计算机可读存储介质
CN110967699B (zh) 用于确定车辆的环境目标所处的区域的方法及装置
JP2019215177A (ja) 軌道設定装置
JP2019049513A (ja) 車両走行制御方法及び装置
JP2023151393A (ja) 走行制御方法及び走行制御装置
CN115077542A (zh) 一种车辆行驶路径规划方法、装置及车辆控制器
CN118500432A (zh) 泊车路径规划方法、装置、电子设备及存储介质
JP2023151308A (ja) 先行車両判定方法及び先行車両判定装置

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: 19865835

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021517849

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2101001898

Country of ref document: TH

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 20217013022

Country of ref document: KR

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2019865835

Country of ref document: EP