CN113479191B - Lane-line-free lane boundary detection system and method for parking and vehicle - Google Patents
Lane-line-free lane boundary detection system and method for parking and vehicle Download PDFInfo
- Publication number
- CN113479191B CN113479191B CN202110742580.1A CN202110742580A CN113479191B CN 113479191 B CN113479191 B CN 113479191B CN 202110742580 A CN202110742580 A CN 202110742580A CN 113479191 B CN113479191 B CN 113479191B
- Authority
- CN
- China
- Prior art keywords
- lane
- parking
- vehicle
- gridmap
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
- B60W30/06—Automatic manoeuvring for parking
Abstract
The invention discloses a lane boundary detection system without lane lines for parking, a method and a vehicle, comprising the following steps: a) Acquiring obstacle point information by using an ultrasonic sensor; b) Acquiring 3D visual point cloud information by using a visual sensor; c) Fusing obstacle point information and visual point cloud information, and filtering dynamic obstacles; d) Projecting the information to a two-dimensional plane to generate a Gridmap grid map, wherein the Gridmap grid map is used for representing a local static environment; e) Converting the Gridmap grid graph into a binary image by using a threshold value; f) Performing linear extraction by a Hough linear detection method provided by opencv; g) F, screening and filtering the line segments extracted in the step f by using the heading information of the vehicle, and outputting 1 boundary at most on the left and right; h) And finally, filtering the boundary line. The invention can realize the detection of the lane boundary under the scene without lane lines, so that the parking system can realize the function of automatically searching the parking spaces.
Description
Technical Field
The invention belongs to the technical field of automatic parking, and particularly relates to a lane boundary detection system and method without lane lines for parking and a vehicle.
Background
In recent years, the problem of difficulty in parking has become more prominent as the amount of automobiles kept has exponentially increased. This problem has attracted much attention from the outside world, especially in large host plants. Nowadays, intelligent driving is emerging, and vehicles equipped with autonomous parking systems are more and more popular. Along with automatic system's function of parking constantly upgrades, more and more host computer factories release the automatic search parking stall function, when the driver finds the place ahead has the parking stall, the driver can push down after the parking key can get off the car or wait in the car, the vehicle can seek the parking stall forward by oneself, it just can stop into by oneself to judge that the parking stall is suitable when the system is listened to the convenience of parkking has been promoted. The control process of the vehicle to automatically advance and search the parking space mainly depends on the advance of the lane center line, so the key step is to know the lane center line. Most roadways have no lane center line, and even some lane-connecting lines. If the system knows the lane boundaries, then the lane centerline can be abstracted and the vehicle can proceed with reference to the lane centerline. However, most of the existing host plants which release the automatic parking space searching function require that a lane line is arranged on a lane, and the host plants do not have the automatic parking space searching function aiming at scenes without the lane line.
In the field of automatic parking space searching, for example, a parking lot management system and method for automatically searching for vacant parking spaces disclosed in patent document CN104282173A, a parking space occupancy detector is provided and respectively arranged on each parking space for detecting whether a vehicle is parked in the parking space, and information on whether the parking space is occupied is transmitted to a parking lot controller in real time; the parking lot controller is used for receiving the parking space occupation information sent by each parking space occupation detector, storing and updating the parking space idle information in real time and displaying the parking space idle information on the entrance screen. The method is mainly used for reminding a driver of the remaining free parking spaces and depends on an intelligent parking lot. As another example, patent document CN107776570A discloses a full-automatic parking method and a full-automatic parking system, but it is not proposed how to search for a parking space by automatically advancing. In the field of lane boundary detection,
therefore, there is a need to develop a new lane boundary detection system, method, vehicle and vehicle for parking without lane lines.
Disclosure of Invention
The invention aims to provide a lane boundary detection system and method without lane lines for parking and a vehicle, which can realize the extraction of lane boundaries under the scene without the lane lines.
In a first aspect, the present invention provides a lane boundary detection method for parking without lane lines, comprising:
a) Acquiring obstacle point information by using an ultrasonic sensor;
b) Acquiring 3D visual point cloud information by using a visual sensor;
c) Fusing the obstacle point information and the visual point cloud information, and filtering out dynamic obstacles;
d) Projecting the information to a two-dimensional plane to generate a Gridmap grid map;
e) Converting the Gridmap grid map into a binary image by using a threshold value;
f) Performing linear extraction by a Hough linear detection method provided by opencv;
g) F, screening and filtering the line segments extracted in the step f by using the heading information of the vehicle, and outputting 1 boundary at most on the left and right;
h) And finally, filtering the boundary line.
Alternatively, when the vehicle is moving, the currently active raster image region should always cover the ROI around the vehicle body.
Optionally, the Gridmap grid map is always only translated and not rotated relative to a global coordinate system, where the global coordinate system uses a rear axle center of the vehicle as an origin of coordinates, an X-axis is directly in front of the vehicle, and a Y-axis is left of the vehicle.
Optionally, each grid of the Gridmap grid map contains three probabilities of free, occupied, and unknown, and the sum of the three values is equal to 1.
Optionally, the Gridmap raster map update output frequency is not less than 5fps.
Optionally, the Gridmap grid map is 200 × 200 in size, with a grid resolution of 10cm × 10cm.
In a second aspect, the present invention provides a lane boundary detection system for parking without a lane line, including:
the ultrasonic sensor is used for acquiring obstacle point information;
the visual sensor is used for acquiring 3D visual point cloud information;
a memory having a computer readable program stored therein;
a controller capable of executing the steps of the lane boundary detection method for a driverless vehicle parking according to any one of claims 1 to 6 when the controller calls the computer readable program.
In a third aspect, a vehicle according to the present invention employs a lane-line-free lane boundary detection system for parking according to the present invention.
The invention has the following advantages: the method can realize lane boundary detection under the scene without lane lines, so that the parking system can automatically search parking spaces according to the detected lane boundaries.
Drawings
FIG. 1 is a flow chart of the present embodiment;
FIG. 2 is a schematic drawing of gridmap with walls on one side;
FIG. 3 is a graph of the boundary line extraction effect for a wall on one side;
FIG. 4 is a schematic diagram of gridmap for curb parking a parallel vehicle;
FIG. 5 is a diagram showing the effect of boundary line extraction for a roadside parked parallel vehicle;
FIG. 6 is a schematic diagram of gridmap for curb parking a vertical vehicle;
fig. 7 is a boundary line extraction effect diagram of a roadside parked vertical vehicle.
Detailed Description
As shown in fig. 1, in the present embodiment, a lane boundary detection method without lane lines for parking includes the following steps:
a) And acquiring obstacle point information by using the ultrasonic sensor.
b) And acquiring the 3D visual point cloud information by using a visual sensor.
c) And fusing the ultrasonic information and the visual point cloud information to filter out dynamic obstacles.
d) And projecting the information to a two-dimensional plane to generate a Gridmap grid map. The Gridmap grid map is used for characterizing a local static environment, and has a size of 200 × 200 (i.e., 200 grids in length and 200 grids in width), and a grid resolution of 10cm × 10cm. When the vehicle is moving, the currently active raster image region should always cover the ROI (i.e. the region of interest) around the vehicle body, which in this embodiment refers to x e [ -10m, +10m ], y e [ -10m, +10m ] under the central coordinate system of the rear axle of the vehicle. The global coordinate system takes the center of the rear axle of the vehicle as the origin of coordinates, the right front of the vehicle is an X-axis, and the left of the vehicle is a Y-axis. The Gridmap raster map update output frequency is not less than 5fps. And the Gridmap grid graph only translates and does not rotate relative to the global coordinate system all the time. Each grid of the Gridmap grid map contains three probabilities of free, occupied, and unknown, with the sum of the three values being equal to 1.
e) And converting the Gridmap grid map into a binary image by using a threshold value.
f) The line extraction is carried out by a Hough line detection method provided by opencv.
g) And f, screening and filtering the line segments extracted in the step f by using the heading information of the vehicle, and outputting 1 boundary at most on the left and right.
h) And finally, filtering the boundary line.
Example 1 is a scene with a wall on one side (upper side), when a vehicle is driving forward, an ultrasonic sensor detects the wall surface and outputs a string of points, and then visual point cloud information is fused to generate a gridmap grid map, referring to fig. 2, the darker the color represents that the grid has a higher probability of being an obstacle, the pixel value of the corresponding image is 255 when the probability of being an obstacle is greater than 60%, and the pixel value of the corresponding image is 0 when the probability of being an obstacle is less than 60%; the distance accuracy is set to be 1 pixel (corresponding to the actual quantity to be 10 cm), the angle accuracy is set to be 1 degree, the threshold parameter of the accumulation plane is 30 degrees, the minimum length of the line segment is 30 pixels (corresponding to the actual minimum length to be 3 m), and the maximum allowable interval of two line segments in the same direction which are determined as one line segment is 8 pixels (corresponding to the actual maximum allowable interval to be 80 cm). The results obtained using the Hough Linear detection method (i.e., houghLinesP) provided by opencv are shown in FIG. 3.
Example 2 is a scene of parking parallel vehicles on the roadside, when a vehicle travels forwards, an ultrasonic sensor detects a vehicle body parked on the roadside vehicle and outputs a string of points, visual point cloud information is fused to generate a gridmap grid map, see fig. 4, wherein a rectangular frame is a parking space detected by a sensing module, the deeper the grid color represents that the probability that the grid is an obstacle is higher, the pixel value of a corresponding image is 255 when the probability that the grid is an obstacle is greater than 60%, and the pixel value of the corresponding image is 0 when the probability that the grid is an obstacle is less than 60%; the distance accuracy is set to be 1 pixel (corresponding to the actual quantity to be 10 cm), the angle accuracy is set to be 1 degree, the threshold parameter of the accumulation plane is 30 degrees, the minimum length of the line segment is 30 pixels (corresponding to the actual minimum length to be 3 m), and the maximum allowable interval of two line segments in the same direction which are determined as one line segment is 8 pixels (corresponding to the actual maximum allowable interval to be 80 cm). The results obtained using the HoughLinesP method supplied by opencv are shown in FIG. 5.
Example 3 is a scene of parking a vertical vehicle on the roadside, in the process of forward driving of the vehicle, the ultrasonic sensor detects the head of the vertical vehicle parked on the roadside and outputs a string of points, then visual point cloud information is fused to generate a gridmap, see fig. 6, wherein a rectangular frame is a parking space detected by the sensing module, the darker the color of a grid represents that the grid has a higher probability of being an obstacle, it is set that when the probability of being an obstacle is greater than 60%, the pixel value of a corresponding image is 255, and when the probability of being an obstacle is less than 60%, the pixel value of the corresponding image is 0; the distance accuracy is set to be 1 pixel (corresponding to the actual quantity to be 10 cm), the angle accuracy is set to be 1 degree, the threshold parameter of the accumulation plane is 30 degrees, the minimum length of the line segment is 30 pixels (corresponding to the actual minimum length to be 3 m), and the maximum allowable interval of two line segments in the same direction which are determined as one line segment is 8 pixels (corresponding to the actual maximum allowable interval to be 80 cm). The results obtained using the HoughLinesP method supplied by opencv are shown in FIG. 7.
Claims (8)
1. A lane boundary detection method for parking without a lane line, characterized by comprising the steps of:
a) Acquiring obstacle point information by using an ultrasonic sensor;
b) Acquiring 3D visual point cloud information by using a visual sensor;
c) Fusing the obstacle point information and the visual point cloud information, and filtering out dynamic obstacles;
d) Projecting the information to a two-dimensional plane to generate a Gridmap grid map, wherein the Gridmap grid map is used for representing a local static environment;
e) Converting the Gridmap grid map into a binary image by using a threshold value;
f) Performing linear extraction by a Hough linear detection method provided by opencv;
g) F, screening and filtering the line segments extracted in the step f by using the heading information of the vehicle, and outputting 1 boundary at most on the left and right;
h) And finally, filtering the boundary line.
2. The lane boundary detection method for parking according to claim 1, characterized in that: when the vehicle is moving, the currently active raster image region should always cover the ROI around the vehicle body.
3. The lane boundary detection method without a lane line for parking according to claim 2, characterized in that: the Gridmap grid graph only translates and does not rotate relative to a global coordinate system all the time, wherein the global coordinate system takes the center of a rear axle of a vehicle as a coordinate origin, the right front of the vehicle is an X axle, and the left of the vehicle is a Y axle.
4. The lane boundary detection method without a lane line for parking according to claim 3, characterized in that: each grid of the Gridmap grid map contains three probabilities of free, occupied, and unknown, and the sum of the three values is equal to 1.
5. The lane boundary detection method without a lane line for parking according to claim 4, characterized in that: and the output frequency of Gridmap raster map updating is not less than 5fps.
6. The lane boundary detection method without a lane line for parking according to claim 5, characterized in that: the Gridmap grid map is 200 × 200 in size, with a grid resolution of 10cm × 10cm.
7. A lane boundary detection system for parking without a lane line, comprising:
the ultrasonic sensor is used for acquiring obstacle point information;
the visual sensor is used for acquiring 3D visual point cloud information;
a memory having a computer readable program stored therein;
a controller, characterized by: the controller, when invoking the computer readable program, is capable of performing the steps of the lane boundary detection method for parking without a lane line according to any of claims 1 to 6.
8. A vehicle, characterized in that: the lane boundary detection system for parking a vehicle using a lane-less according to claim 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110742580.1A CN113479191B (en) | 2021-06-30 | 2021-06-30 | Lane-line-free lane boundary detection system and method for parking and vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110742580.1A CN113479191B (en) | 2021-06-30 | 2021-06-30 | Lane-line-free lane boundary detection system and method for parking and vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113479191A CN113479191A (en) | 2021-10-08 |
| CN113479191B true CN113479191B (en) | 2023-04-07 |
Family
ID=77936998
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110742580.1A Active CN113479191B (en) | 2021-06-30 | 2021-06-30 | Lane-line-free lane boundary detection system and method for parking and vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113479191B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113942458B (en) * | 2021-10-29 | 2022-07-29 | 禾多科技（北京）有限公司 | Control method, device, equipment and medium for vehicle-mounted camera adjusting system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010140170A (en) * | 2008-12-10 | 2010-06-24 | Toyota Motor Corp | Lane boundary deriving device |
| DE102013200387A1 (en) * | 2013-01-14 | 2014-07-17 | Robert Bosch Gmbh | Creation of a obstacle map |
| CN106627582A (en) * | 2016-12-09 | 2017-05-10 | 重庆长安汽车股份有限公司 | Path planning system and method for overtaking vehicle on adjacent lane in single-lane automatic drive mode |
| CN108805065A (en) * | 2018-05-31 | 2018-11-13 | 华南理工大学 | One kind being based on the improved method for detecting lane lines of geometric properties |
| CN110147109A (en) * | 2019-05-21 | 2019-08-20 | 重庆长安汽车股份有限公司 | A kind of archetype development system of automated driving system |
| CN110838233A (en) * | 2019-10-12 | 2020-02-25 | 中国平安人寿保险股份有限公司 | Vehicle behavior analysis method and device and computer readable storage medium |
| CN111273305A (en) * | 2020-02-18 | 2020-06-12 | 中国科学院合肥物质科学研究院 | Multi-sensor fusion road extraction and indexing method based on global and local grid maps |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9090263B2 (en) * | 2010-07-20 | 2015-07-28 | GM Global Technology Operations LLC | Lane fusion system using forward-view and rear-view cameras |
| KR102267562B1 (en) * | 2015-04-16 | 2021-06-22 | 한국전자통신연구원 | Device and method for recognition of obstacles and parking slots for unmanned autonomous parking |
| CN104850834A (en) * | 2015-05-11 | 2015-08-19 | 中国科学院合肥物质科学研究院 | Road boundary detection method based on three-dimensional laser radar |
| US10832061B2 (en) * | 2016-07-22 | 2020-11-10 | Hitachi Automotive Systems, Ltd. | Traveling road boundary estimation apparatus and traveling assistance system using same |
| CN110044371A (en) * | 2018-01-16 | 2019-07-23 | 华为技术有限公司 | A kind of method and vehicle locating device of vehicle location |
| CN112313663A (en) * | 2018-06-18 | 2021-02-02 | 祖克斯有限公司 | Occlusion aware planning |
| CN109532821A (en) * | 2018-11-09 | 2019-03-29 | 重庆长安汽车股份有限公司 | Merge parking system |
| CN110287779B (en) * | 2019-05-17 | 2021-10-15 | 百度在线网络技术（北京）有限公司 | Method, device and equipment for detecting lane line |
| CN110861639B (en) * | 2019-11-28 | 2021-06-15 | 安徽江淮汽车集团股份有限公司 | Parking information fusion method and device, electronic equipment and storage medium |
| CN111319615B (en) * | 2020-03-16 | 2021-02-26 | 湖北亿咖通科技有限公司 | Intelligent passenger-replacing parking method, computer-readable storage medium and electronic device |
| CN112776797A (en) * | 2021-02-27 | 2021-05-11 | 重庆长安汽车股份有限公司 | Original parking space parking establishment method and system, vehicle and storage medium |
-
2021
- 2021-06-30 CN CN202110742580.1A patent/CN113479191B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010140170A (en) * | 2008-12-10 | 2010-06-24 | Toyota Motor Corp | Lane boundary deriving device |
| DE102013200387A1 (en) * | 2013-01-14 | 2014-07-17 | Robert Bosch Gmbh | Creation of a obstacle map |
| CN106627582A (en) * | 2016-12-09 | 2017-05-10 | 重庆长安汽车股份有限公司 | Path planning system and method for overtaking vehicle on adjacent lane in single-lane automatic drive mode |
| CN108805065A (en) * | 2018-05-31 | 2018-11-13 | 华南理工大学 | One kind being based on the improved method for detecting lane lines of geometric properties |
| CN110147109A (en) * | 2019-05-21 | 2019-08-20 | 重庆长安汽车股份有限公司 | A kind of archetype development system of automated driving system |
| CN110838233A (en) * | 2019-10-12 | 2020-02-25 | 中国平安人寿保险股份有限公司 | Vehicle behavior analysis method and device and computer readable storage medium |
| CN111273305A (en) * | 2020-02-18 | 2020-06-12 | 中国科学院合肥物质科学研究院 | Multi-sensor fusion road extraction and indexing method based on global and local grid maps |
Non-Patent Citations (2)
| Title |
|---|
| 一种自适应动态窗口车道线高速检测方法;龚建伟等;《北京理工大学学报》(第06期);20-24 * |
| 基于多传感融合的车道线检测与跟踪方法的研究;吴彦文等;《计算机应用研究》(第02期);286-289、293 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113479191A (en) | 2021-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220107651A1 (en) | Predicting three-dimensional features for autonomous driving | |
| CN109117709B (en) | Collision avoidance system for autonomous vehicles | |
| CN108959321B (en) | Parking lot map construction method, system, mobile terminal and storage medium | |
| US20210342637A1 (en) | Generating ground truth for machine learning from time series elements | |
| US20190347498A1 (en) | Systems and methods for automated detection of trailer properties | |
| CN111874006B (en) | Route planning processing method and device | |
| CN110765894B (en) | Target detection method, device, equipment and computer readable storage medium | |
| CN106647776B (en) | Method and device for judging lane changing trend of vehicle and computer storage medium | |
| US9360332B2 (en) | Method for determining a course of a traffic lane for a vehicle | |
| US10325163B2 (en) | Vehicle vision | |
| US20200082182A1 (en) | Training data generating method for image processing, image processing method, and devices thereof | |
| US11003920B2 (en) | Detection and planar representation of three dimensional lanes in a road scene | |
| CN110320531B (en) | Obstacle identification method based on laser radar, map creation method and device | |
| US20230184560A1 (en) | Visual interface display method and apparatus, electronic device, and storage medium | |
| US10286901B2 (en) | Map of the surroundings for driving areas with random altitude profile | |
| US11662221B2 (en) | Change point detection device and map information distribution system | |
| CN108960035A (en) | The method and system of usage scenario detection pattern object of classification in perception scene figure | |
| CN111192311A (en) | Automatic extraction method and device for longitudinal deceleration marked line in high-precision map making | |
| CN108725318B (en) | Automobile safety early warning method and device and computer readable storage medium | |
| CN113479191B (en) | Lane-line-free lane boundary detection system and method for parking and vehicle | |
| CN114537447A (en) | Safe passing method and device, electronic equipment and storage medium | |
| CN114694108A (en) | Image processing method, device, equipment and storage medium | |
| CN113838060A (en) | Perception system for autonomous vehicle | |
| CN114954442A (en) | Vehicle control method and system and vehicle | |
| CN114779276A (en) | Obstacle detection method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |