CN109188459A - A kind of small obstacle recognition method in ramp based on multi-line laser radar - Google Patents
A kind of small obstacle recognition method in ramp based on multi-line laser radar Download PDFInfo
- Publication number
- CN109188459A CN109188459A CN201811000910.4A CN201811000910A CN109188459A CN 109188459 A CN109188459 A CN 109188459A CN 201811000910 A CN201811000910 A CN 201811000910A CN 109188459 A CN109188459 A CN 109188459A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- road
- road surface
- points
- point
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 230000009466 transformation Effects 0.000 claims description 19
- 230000009467 reduction Effects 0.000 claims description 9
- 238000009826 distribution Methods 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000003708 edge detection Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 abstract description 11
- 230000004888 barrier function Effects 0.000 description 17
- 230000000694 effects Effects 0.000 description 10
- 238000005457 optimization Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 7
- 238000001914 filtration Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000019771 cognition Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000010408 sweeping Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
- G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Traffic Control Systems (AREA)
- Optical Radar Systems And Details Thereof (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
本发明涉及一种基于多线激光雷达的坡道小障碍物识别方法，实现了坡道路面的小障碍物识别，快速准确，节省运算资源，保证了实时性；有效避免了传统障碍物识别方法在即将下坡路段障碍物的漏检以及上坡路段把路面识别为障碍物的弊端，提高了智能驾驶汽车的行车安全性和对复杂路况的适应性。
The invention relates to a method for identifying small obstacles on a slope based on a multi-line laser radar, which realizes the identification of small obstacles on a slope road surface, is fast and accurate, saves computing resources, and ensures real-time performance, and effectively avoids the traditional obstacle identification method. The missed detection of obstacles in the upcoming downhill section and the disadvantages of identifying the road surface as an obstacle in the uphill section improve the driving safety of intelligent driving vehicles and the adaptability to complex road conditions.
Description
Technical field
The present invention relates to a kind of small obstacle recognition methods in the ramp based on multi-line laser radar, belong to unmanned obstacle
Object identifies field.
Background technique
Detection of obstacles on vehicle driving road is in pilotless automobile surrounding enviroment cognition technology research field
The effect of important component, detection of obstacles is directly related to the traffic safety of pilotless automobile.Tradition is based on multi-thread thunder
The obstacle detection method reached is all the point cloud that filters out all z coordinates be less than-h rough according to the mounting height h of radar, this
Method can be identified as ramp large obstacle in the section that will be gone up a slope, and can filter out downhill path section on ramp
Small obstacle.So a kind of obstacle recognition method that can be suitably used for ramp road conditions is particularly important.
Summary of the invention
The present invention provides a kind of small obstacle recognition method in the ramp based on multi-line laser radar, is applicable not only to horizontal road
Face, and it is suitable for ramp road surface, accurately road surface and barrier can be distinguished, there is stronger practicability and wide
Application prospect.
The technical solution adopted by the present invention to solve the technical problems is:
A kind of small obstacle recognition method in ramp based on multi-line laser radar, comprising the following steps:
The first step is carried out road-edge detection to original point cloud, filters out the region on the outside of road based on Hough transformation, obtain
In to road with the reference point cloud of obstacle recognition；
Second step is projected to the reference point cloud on the road surface of ramp with obstacle recognition and road based on projection dimension reduction method
On the vertical perpendicular XOZ in face, ramp road identify and filter out ramp road surface, Hough straight line is recycled to become
Swap-in row detection, so that all reference point clouds for belonging to road surface scanning element are accurately found out, then in the phase for belonging to road surface scanning element
It closes and all road surface points is filtered out again in point cloud；
Third step carries out outlier to the range within ten meters on the road surface of ramp and filters out, filters out road surface zero according to outlier
Scattered point cloud makes the point cloud after filtering out belong to barrier scanning element；
4th step, the barrier scanning element after being filtered out using the DBCSAN clustering algorithm based on density to outlier carry out into
One step filters out noise spot, realizes barrier cluster；
As present invention further optimization, by the region segmentation inside and outside road, the point cloud in region forms image coordinate
Space, using Hough transformation by point Cloud transform conllinear in image coordinate space into parameter space, these clouds parameter sky
Between in intersect at same point, the straight line that intersects with ramp pavement edge is excluded by Hough transformation, with this to road on the outside of
Region is filtered out, and the reference point cloud in road with obstacle recognition is obtained；
As present invention further optimization, if not finding road edge, i.e. selection two lateral extent of automobile body is each
Five meters of inner regions are as cut zone；
As present invention further optimization, based on projection dimension reduction method, by the phase on the road surface of ramp with obstacle recognition
It closes point cloud to project on the perpendicular XOZ vertical with road surface, the projection plane of reference point cloud is formed, on a projection plane from thunder
A continuous straight line is formed up to nearest stretch face, intermediate one section of ramp is the straight line of segmentation, and remaining stretch face exists
It is not shown on projection plane；Hough transformation is carried out for the point cloud number after aforementioned all projections, the length to be formed is found out and is greater than
0.5 meter of straight line, so that the straight line for road surface is found, to accurately find out all reference point clouds for belonging to road surface scanning element；
As present invention further optimization, in the range of within ten meters on the road surface of ramp, in input data to point
The range distribution of cloud to point of proximity cloud is calculated, and show that a cloud to the average distance of its point of proximity cloud, is in this average departure
From average value ranges in point cloud be barrier scanning element, incongruent cloud is rejected；
As present invention further optimization, it is circular conical surface by the region that every line of radar scans, uses
DBCSAN clustering algorithm judges this by calculating the density value in the Euclidean distance and its place sweep radius between each point
A little points are to belong to core point, boundary point either noise spot.
By above technical scheme, compared with the existing technology, the invention has the following advantages:
The present invention realizes the small obstacle recognition on ramp road surface, quick and precisely, saves calculation resources, ensure that in real time
Property；Traditional obstacle recognition method is effectively prevented road surface to be known the missing inspection of downhill path section barrier and uphill way
Not Wei barrier the drawbacks of, improve the travel safety of intelligent driving automobile and the adaptability to complex road condition.
Detailed description of the invention
Present invention will be further explained below with reference to the attached drawings and examples.
It is the small obstacle recognition flow chart in ramp of the preferred embodiment of the present invention shown in Fig. 1；
It is the confirmatory experiment scene figure of the preferred embodiment of the present invention shown in Fig. 2；
In Fig. 3,3a is that Hough transformation detects the straight line schematic diagram in image coordinate space, and 3b is that straight line schematic diagram is mapped to
Sine curve in parameter space；
In Fig. 4,4a is road-edge detection effect picture, and 4b filters out the related point cloud chart after Independent Point cloud；
It is the xoz projecting method schematic diagram of the preferred embodiment of the present invention shown in Fig. 5；
It is the reference point cloud xoz drop shadow effect figure of the preferred embodiment of the present invention shown in Fig. 6；
It is the road surface recognition effect figure of the preferred embodiment of the present invention shown in Fig. 7；
It is the effect picture filtered out behind road surface of the preferred embodiment of the present invention shown in Fig. 8；
It is the effect picture rejected after outlier of the preferred embodiment of the present invention shown in Fig. 9；
It is the BSDCAN algorithm principle figure of the preferred embodiment of the present invention shown in Figure 10；
Be shown in Figure 11 the preferred embodiment of the present invention cluster after point cloud effect and each class mass center position.
In figure: 1-10 indicates the multiple cone buckets being distributed on the ground as small obstacle.
Specific embodiment
In conjunction with the accompanying drawings, the present invention is further explained in detail.These attached drawings are simplified schematic diagram, only with
Illustration illustrates basic structure of the invention, therefore it only shows the composition relevant to the invention.
As shown in Figure 1, a kind of small obstacle recognition method in ramp based on multi-line laser radar of the invention, including it is following
Step:
The first step is carried out road-edge detection to original point cloud, filters out the region on the outside of road based on Hough transformation, obtain
In to road with the reference point cloud of obstacle recognition；
Second step is projected to the reference point cloud on the road surface of ramp with obstacle recognition and road based on projection dimension reduction method
On the vertical perpendicular XOZ in face, ramp road identify and filter out ramp road surface, Hough straight line is recycled to become
Swap-in row detection, so that all reference point clouds for belonging to road surface scanning element are accurately found out, then in the phase for belonging to road surface scanning element
It closes and all road surface points is filtered out again in point cloud；
Third step carries out outlier to the range within ten meters on the road surface of ramp and filters out, filters out road surface zero according to outlier
Scattered point cloud makes the point cloud after filtering out belong to barrier scanning element；
4th step, the barrier scanning element after being filtered out using the DBCSAN clustering algorithm based on density to outlier carry out into
One step filters out noise spot, realizes barrier cluster；
As present invention further optimization, by the region segmentation inside and outside road, the point cloud in region forms image coordinate
Space, using Hough transformation by point Cloud transform conllinear in image coordinate space into parameter space, these clouds parameter sky
Between in intersect at same point, the straight line that intersects with ramp pavement edge is excluded by Hough transformation, with this to road on the outside of
Region is filtered out, and the reference point cloud in road with obstacle recognition is obtained；
As present invention further optimization, if not finding road edge, i.e. selection two lateral extent of automobile body is each
Five meters of inner regions are as cut zone；
As present invention further optimization, based on projection dimension reduction method, by the phase on the road surface of ramp with obstacle recognition
It closes point cloud to project on the perpendicular XOZ vertical with road surface, the projection plane of reference point cloud is formed, on a projection plane from thunder
A continuous straight line is formed up to nearest stretch face, intermediate one section of ramp is the straight line of segmentation, and remaining stretch face exists
It is not shown on projection plane；Hough transformation is carried out for the point cloud number after aforementioned all projections, the length to be formed is found out and is greater than
0.5 meter of straight line, so that the straight line for road surface is found, to accurately find out all reference point clouds for belonging to road surface scanning element；
As present invention further optimization, in the range of within ten meters on the road surface of ramp, in input data to point
The range distribution of cloud to point of proximity cloud is calculated, and show that a cloud to the average distance of its point of proximity cloud, is in this average departure
From average value ranges in point cloud be barrier scanning element, incongruent cloud is rejected；
As present invention further optimization, it is circular conical surface by the region that every line of radar scans, uses
DBCSAN clustering algorithm judges this by calculating the density value in the Euclidean distance and its place sweep radius between each point
A little points are to belong to core point, boundary point either noise spot.
Specific operating process is as follows:
The first step shown in Fig. 2, is distributed multiple cone buckets on the ground, will be unrelated with small obstacle detection as small obstacle
Point cloud in region is first filtered out, to reduce the calculation resources consumption of points cloud processing below；Extraneous areas referred to herein
For the region on the outside of road, this system only detects the barrier on road；
Straight line is detected first with Hough transformation, searches road edge, Hough transformation detects straight line schematic diagram, in Fig. 3,3a
With shown in 3b: Hough transformation by the point transformation in image coordinate space into parameter space, conllinear point in image coordinate space
After transforming in parameter space, same point is all intersected in parameter space, at this time obtained ρ, θ, ρ, θ are required straight line
Pole coordinate parameter, the conllinear two o'clock (x of image coordinate spacei,yi) and (xj,yj), (xi,yi) and (xj,yj) it is mapped to parameter
Space is two sine curves, intersects at point (ρ0,θ0), as shown in 3b in Fig. 3；Conversely, intersecting at same point in parameter space
All sine curves have conllinear point to be corresponding to it in image coordinate space；According to this characteristic, given image coordinate space
Some marginal points, so that it may pass through Hough transformation determine connection these point linear equation；When using Hough transformation here,
Here we limit θ ∈ (- 45 °, 45 °), to exclude the straight line intersected with road edge as far as possible.
If finding road edge, as shown in 4a in Fig. 4, two straight lines in figure are the road edge detected；
Then the point cloud data on the outside of road is filtered out, as shown in 4b in Fig. 4；Wherein, the method for filtering out a cloud is, is point with road edge
Boundary line is the point cloud on the inside of road close to middle line, such as straight line " x=0 " in figure, otherwise is the point cloud on the outside of road, will put cloud
X, y, z coordinate value is set as INF, and (preset infinitely large quantity is herein 10000.0, i.e., 10000.0 meters are considered as nothing in systems
It is poor big), coordinate is that can all be ignored in operation of the point of INF below, that is, is filtered out, side point cloud is not dealt in road；
If not finding road edge, using each 5 meters of inner regions of distance at left and right sides of automobile body as relevant range (with two
Body width is safe distance)；In addition, passing through detection road edge, it is known that, there is 4.7 ° of drift angle immediately ahead of road and radar,
Radar center point is deviated to the right 0.2 meter of lane center.
Second step, general detection of obstacles is will not specially to go to filter out road surface point cloud, because of general detection of obstacles
Object be mostly large obstacle, they included point cloud signal it is more much more than the point cloud signal on the ground near them,
So road surface point cloud is very small on large obstacle detection influence；And in the process of multi-line laser radar identification small obstacle
In, road surface filtering is extremely important, it is unobvious with road surface differentiation because the point cloud quantity that small barrier itself includes is considerably less,
By 3b in Fig. 3 as it can be seen that small barrier almost combines together with ground, road surface point cloud is maximum noise in entire point cloud signal,
In order to preferably carry out further work, first have to filter out in road surface point cloud, to improve signal-to-noise ratio；But accurately filter out ground ratio
More difficult, 3b can be seen that in a top view from Fig. 3, and the point cloud shape on road surface is circular arc or ellipse, be unfavorable for unified
Feature screened, need to go out ground point cloud from new angle extraction thus, and this paper presents one kind in cloud signal
Definition goes out the new method on road surface, this process employs the method for projection dimensionality reduction and Hough transformation, projection drop herein
Dimension is different from the projection dimensionality reduction in most papers, and the projection dimensionality reduction of most papers is all that 3D point cloud is projected to 2D water
Plane forms 2D map, and herein then from new angle, all the points cloud is projected into the perpendicular perpendicular to road surface
On, it is specific as shown in Figure 4；
In Fig. 5, using the central point of multi-line laser radar as coordinate origin, cartesian coordinate system is established, wherein o ' (x0,y0,
z0) it is corresponding points of the radar center point on lane center, o ' o " is the tangent line at the place lane center o ', and θ is o ' o " in y-axis
Angle；The plane perpendicular to road surface where o ' o " is the projection plane of all the points cloud；Herein with matrix A storage region point
The three-dimensional coordinate information for cutting rear all the points cloud, then have
Enable the point cloud coordinate matrix of projectionIt indicates, then has
For the point cloud in 4b in Fig. 4, θ=4.7 °, x0=-0.2, y0=0, projection result, as shown in Figure 6；
For figure after Binding experiment projection it is found that shown in Fig. 7, road surface is approximately straight line, and nearest from radar one
Road section surface forms a continuous straight line, intermediate one section of ramp, because sweeping to road surface since radar point cloud data is than comparatively dense
Harness it is less, so be segmentation straight line, and last stretch face inherently without point cloud distribution, so from projection
It is displayed without road surface.After projection, Hough transformation is carried out herein for the 2D point cloud data, the inside length is found out and is greater than 0.5 meter
Straight line because the diameter of small barrier circumsphere be no more than 0.5 meter, find straight line herein and be only possible to be road surface；Hough
After transformation, the point cloud below the point cloud and the direction straight line z that all straight lines for meeting length include is filtered out；Effect after filtering out road surface
For fruit as shown in figure 8, after the filtering of road surface, ground point cloud has filtered out the position that completely can clearly be observed that cone bucket completely.Third
In step, how outlier filters out scattered point；
After the point cloud for having filtered out road surface and extraneous areas, still have as shown in Figure 8, in point cloud chart some very scattered
Point cloud noise, at this point, the application will filter out scattered point cloud according to outlier, to guarantee that remaining cloud is barrier institute
The point cloud for including；Specific algorithm used in this application is that statistics outlier rejects algorithm, and the analysis of this algorithm execution point cloud is simultaneously
And the point cloud for being unsatisfactory for designated statistics feature can be rejected；Statistical nature in the application is in being averaged for distance between a cloud
Within the scope of one near value, and reject the too many point of those deviation averages.One statistics is carried out to the neighborhood of each cloud
Analysis, and trim those point clouds for not meeting certain standard；
Specifically, being calculated herein each point the calculating of the range distribution of cloud to point of proximity cloud in input data
It arrives the average distance of its all point of proximity clouds.Assuming that obtain the result is that a Gaussian Profile, shape is by mean value and mark
Quasi- difference determines, point of the average distance except critical field (being defined by global distance average and variance), can be defined as from
Group puts and can get rid of from data set；After this algorithm, the point isolated in point cloud chart is also filtered out, to prevent from filtering
The cone bucket of distant place is removed, only the carry out outlier in 10 meters is filtered out, as shown in Figure 9:
In 4th step, the region of the every line scanning of multi-line laser radar is a circular conical surface, therefore even if radar scanning is arrived
The object of rule, an even plane, point cloud distribution on this plane is also non-uniform.Complicated in face of motor vehicle environment
When environment, the point cloud distribution scanned is that extremely unevenly, shape also calculate by irregular, classical Kmeans algorithm and hierarchical clustering
Method will face huge failure risk, use the BDSCAN algorithm based on density herein to adapt to such case, BDSCAN algorithm
Barrier can be effectively clustered out, and noise spot can be further filtered out, accomplishes quick and precisely, calculation resources to be saved, to protect
Demonstrate,prove real-time.The schematic diagram of BDSCAN algorithm, as shown in Figure 10:
Judge that these points belong to by calculating the density value in the Euclidean distance and its place sweep radius between each point
In core point, boundary point or noise spot.It is 1.5 that sweep radius is set in Figure 10, density threshold 3, so:
(1) P0 point is boundary point, because only there are two point P0 and P1 in the sweep radius centered on it；
(2) P1 point is core point, because there are four point P0, P1, P2, P4 in the sweep radius centered on it；
(3) P8 is noise spot, because it is neither core point nor boundary point；
(4) other are put.
After point cloud tracking BDSCAN in Fig. 9 is clustered, obtained effect is as shown in figure 11, the effect of comparative diagram 11 with
And the position in kind of Fig. 2, it is known that all cone buckets are detected.
After BDSCAN algorithm according to third step step by further cancelling noise point cloud (i.e. by the XYZ coordinate assignment of cloud
For INF).
Those skilled in the art of the present technique are appreciated that unless otherwise defined, all terms used herein (including technology art
Language and scientific term) there is meaning identical with the general understanding of those of ordinary skill in the application fields.Should also
Understand, those terms such as defined in the general dictionary, which should be understood that, to be had and the meaning in the context of the prior art
The consistent meaning of justice, and unless defined as here, it will not be explained in an idealized or overly formal meaning.
The meaning of "and/or" described herein refers to that the case where respective individualism or both exists simultaneously wraps
Including including.
The meaning of " connection " described herein can be between component be directly connected to be also possible to pass through between component
Other components are indirectly connected with.
Taking the above-mentioned ideal embodiment according to the present invention as inspiration, through the above description, relevant staff is complete
Various changes and amendments can be carried out without departing from the scope of the technological thought of the present invention' entirely.The technology of this invention
Property range is not limited to the contents of the specification, it is necessary to which the technical scope thereof is determined according to the scope of the claim.
Claims (6)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811000910.4A CN109188459B (en) | 2018-08-29 | 2018-08-29 | Ramp small obstacle identification method based on multi-line laser radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811000910.4A CN109188459B (en) | 2018-08-29 | 2018-08-29 | Ramp small obstacle identification method based on multi-line laser radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109188459A true CN109188459A (en) | 2019-01-11 |
| CN109188459B CN109188459B (en) | 2022-04-15 |
Family
ID=64917334
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811000910.4A Active CN109188459B (en) | 2018-08-29 | 2018-08-29 | Ramp small obstacle identification method based on multi-line laser radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109188459B (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109947109A (en) * | 2019-04-02 | 2019-06-28 | 北京石头世纪科技股份有限公司 | Robot working area map construction method, device, robot and medium |
| CN109959910A (en) * | 2019-03-18 | 2019-07-02 | 哈尔滨理工大学 | Ground Filtering Algorithm Based on Gaussian Distribution |
| CN109961440A (en) * | 2019-03-11 | 2019-07-02 | 重庆邮电大学 | A 3D LiDAR Point Cloud Target Segmentation Method Based on Depth Map |
| CN110109142A (en) * | 2019-04-04 | 2019-08-09 | 深圳市速腾聚创科技有限公司 | Point cloud filtering method, device, computer equipment and storage medium |
| CN110135599A (en) * | 2019-05-15 | 2019-08-16 | 南京林业大学 | UAV power inspection point cloud intelligent processing and analysis service platform |
| CN110471086A (en) * | 2019-09-06 | 2019-11-19 | 北京云迹科技有限公司 | A kind of radar survey barrier system and method |
| CN110598541A (en) * | 2019-08-05 | 2019-12-20 | 香港理工大学深圳研究院 | Method and equipment for extracting road edge information |
| CN110780305A (en) * | 2019-10-18 | 2020-02-11 | 华南理工大学 | A racetrack cone detection and target tracking method based on multi-line lidar |
| CN111583257A (en) * | 2020-05-28 | 2020-08-25 | 中国铁道科学研究院集团有限公司 | Railway boundary foreign body intrusion detection method, device and system |
| CN112313538A (en) * | 2019-11-05 | 2021-02-02 | 深圳市大疆创新科技有限公司 | Target detection method, radar, device and storage medium |
| CN112634310A (en) * | 2020-12-14 | 2021-04-09 | 深兰人工智能(深圳)有限公司 | Road boundary extraction method, device, electronic equipment and storage medium |
| CN112904306A (en) * | 2021-01-18 | 2021-06-04 | 深圳市普渡科技有限公司 | Slope sensing method and device, robot and storage medium |
| CN113589324A (en) * | 2021-07-26 | 2021-11-02 | 北京理工大学 | Unmanned vehicle gradient identification method and system based on laser radar and storage medium |
| CN114545916A (en) * | 2020-11-27 | 2022-05-27 | 苏州宝时得电动工具有限公司 | Object detection system and method |
| CN115249270A (en) * | 2022-09-22 | 2022-10-28 | 广州市德赛西威智慧交通技术有限公司 | Automatic re-labeling method and system for radar-vision all-in-one machine |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005258941A (en) * | 2004-03-12 | 2005-09-22 | Toyota Central Res & Dev Lab Inc | Obstacle detection device |
| CN104950313A (en) * | 2015-06-11 | 2015-09-30 | 同济大学 | Road-surface abstraction and road gradient recognition method |
| CN105184852A (en) * | 2015-08-04 | 2015-12-23 | 百度在线网络技术（北京）有限公司 | Laser-point-cloud-based urban road identification method and apparatus |
| US20170185089A1 (en) * | 2015-12-27 | 2017-06-29 | Toyota Motor Engineering & Manufacturing North America, Inc. | Detection of overhanging objects |
| CN108226895A (en) * | 2017-12-27 | 2018-06-29 | 吉林大学 | Static-obstacle thing identifying system and recognition methods based on laser radar |
| CN108256577A (en) * | 2018-01-18 | 2018-07-06 | 东南大学 | A kind of barrier clustering method based on multi-line laser radar |
| CN108280840A (en) * | 2018-01-11 | 2018-07-13 | 武汉理工大学 | A kind of road real time method for segmenting based on three-dimensional laser radar |
| CN108280401A (en) * | 2017-12-27 | 2018-07-13 | 达闼科技（北京）有限公司 | A kind of pavement detection method, apparatus, cloud server and computer program product |
-
2018
- 2018-08-29 CN CN201811000910.4A patent/CN109188459B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005258941A (en) * | 2004-03-12 | 2005-09-22 | Toyota Central Res & Dev Lab Inc | Obstacle detection device |
| CN104950313A (en) * | 2015-06-11 | 2015-09-30 | 同济大学 | Road-surface abstraction and road gradient recognition method |
| CN105184852A (en) * | 2015-08-04 | 2015-12-23 | 百度在线网络技术（北京）有限公司 | Laser-point-cloud-based urban road identification method and apparatus |
| US20170185089A1 (en) * | 2015-12-27 | 2017-06-29 | Toyota Motor Engineering & Manufacturing North America, Inc. | Detection of overhanging objects |
| CN108226895A (en) * | 2017-12-27 | 2018-06-29 | 吉林大学 | Static-obstacle thing identifying system and recognition methods based on laser radar |
| CN108280401A (en) * | 2017-12-27 | 2018-07-13 | 达闼科技（北京）有限公司 | A kind of pavement detection method, apparatus, cloud server and computer program product |
| CN108280840A (en) * | 2018-01-11 | 2018-07-13 | 武汉理工大学 | A kind of road real time method for segmenting based on three-dimensional laser radar |
| CN108256577A (en) * | 2018-01-18 | 2018-07-06 | 东南大学 | A kind of barrier clustering method based on multi-line laser radar |
Non-Patent Citations (1)
| Title |
|---|
| 蔡云飞 等: "基于双多线激光雷达的非结构化环境负障碍感知技术", 《自动化学报》 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109961440A (en) * | 2019-03-11 | 2019-07-02 | 重庆邮电大学 | A 3D LiDAR Point Cloud Target Segmentation Method Based on Depth Map |
| CN109959910A (en) * | 2019-03-18 | 2019-07-02 | 哈尔滨理工大学 | Ground Filtering Algorithm Based on Gaussian Distribution |
| CN109947109A (en) * | 2019-04-02 | 2019-06-28 | 北京石头世纪科技股份有限公司 | Robot working area map construction method, device, robot and medium |
| US12201250B2 (en) | 2019-04-02 | 2025-01-21 | Beijing Roborock Innovation Technology Co., Ltd. | Method and apparatus for constructing map of working region for robot, robot, and medium |
| CN109947109B (en) * | 2019-04-02 | 2022-06-21 | 北京石头创新科技有限公司 | Robot working area map construction method and device, robot and medium |
| CN110109142B (en) * | 2019-04-04 | 2021-04-02 | 深圳市速腾聚创科技有限公司 | Point cloud filtering method and device, computer equipment and storage medium |
| CN110109142A (en) * | 2019-04-04 | 2019-08-09 | 深圳市速腾聚创科技有限公司 | Point cloud filtering method, device, computer equipment and storage medium |
| CN110135599A (en) * | 2019-05-15 | 2019-08-16 | 南京林业大学 | UAV power inspection point cloud intelligent processing and analysis service platform |
| CN110598541A (en) * | 2019-08-05 | 2019-12-20 | 香港理工大学深圳研究院 | Method and equipment for extracting road edge information |
| CN110598541B (en) * | 2019-08-05 | 2021-07-23 | 香港理工大学深圳研究院 | A method and device for extracting road edge information |
| CN110471086A (en) * | 2019-09-06 | 2019-11-19 | 北京云迹科技有限公司 | A kind of radar survey barrier system and method |
| CN110780305B (en) * | 2019-10-18 | 2023-04-21 | 华南理工大学 | Track cone detection and target point tracking method based on multi-line laser radar |
| CN110780305A (en) * | 2019-10-18 | 2020-02-11 | 华南理工大学 | A racetrack cone detection and target tracking method based on multi-line lidar |
| CN112313538A (en) * | 2019-11-05 | 2021-02-02 | 深圳市大疆创新科技有限公司 | Target detection method, radar, device and storage medium |
| CN111583257A (en) * | 2020-05-28 | 2020-08-25 | 中国铁道科学研究院集团有限公司 | Railway boundary foreign body intrusion detection method, device and system |
| CN114545916A (en) * | 2020-11-27 | 2022-05-27 | 苏州宝时得电动工具有限公司 | Object detection system and method |
| CN112634310A (en) * | 2020-12-14 | 2021-04-09 | 深兰人工智能(深圳)有限公司 | Road boundary extraction method, device, electronic equipment and storage medium |
| CN112904306A (en) * | 2021-01-18 | 2021-06-04 | 深圳市普渡科技有限公司 | Slope sensing method and device, robot and storage medium |
| WO2022152263A1 (en) * | 2021-01-18 | 2022-07-21 | 深圳市普渡科技有限公司 | Slope sensing method and device, robot, and storage medium |
| CN113589324A (en) * | 2021-07-26 | 2021-11-02 | 北京理工大学 | Unmanned vehicle gradient identification method and system based on laser radar and storage medium |
| CN115249270A (en) * | 2022-09-22 | 2022-10-28 | 广州市德赛西威智慧交通技术有限公司 | Automatic re-labeling method and system for radar-vision all-in-one machine |
| CN115249270B (en) * | 2022-09-22 | 2022-12-30 | 广州市德赛西威智慧交通技术有限公司 | Automatic re-labeling method and system for radar-vision all-in-one machine |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109188459B (en) | 2022-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109188459A (en) | A kind of small obstacle recognition method in ramp based on multi-line laser radar | |
| CN109858460B (en) | Lane line detection method based on three-dimensional laser radar | |
| CN108509918B (en) | Target detection and tracking method fusing laser point cloud and image | |
| CN110781827B (en) | A road edge detection system and method based on lidar and fan-shaped space segmentation | |
| CN108363065B (en) | Object detection system | |
| WO2022188663A1 (en) | Target detection method and apparatus | |
| CN111337941A (en) | Dynamic obstacle tracking method based on sparse laser radar data | |
| CN113030997B (en) | Method for detecting travelable area of open-pit mine area based on laser radar | |
| CN110208819A (en) | A kind of processing method of multiple barrier three-dimensional laser radar data | |
| CN106338989B (en) | A kind of field robot binocular vision navigation methods and systems | |
| JP5145585B2 (en) | Target detection device | |
| Lee et al. | Stereo vision–based vehicle detection using a road feature and disparity histogram | |
| CN111079611A (en) | Automatic extraction method for road surface and marking line thereof | |
| CN109001757B (en) | Parking space intelligent detection method based on 2D laser radar | |
| CN108280840B (en) | Road real-time segmentation method based on three-dimensional laser radar | |
| CN107192994A (en) | Multi-line laser radar mass cloud data is quickly effectively extracted and vehicle, lane line characteristic recognition method | |
| CN108535727B (en) | Method for tracking multiple objects in the vicinity of the host vehicle | |
| CN116524219A (en) | A Method of Obstacle Detection Based on LiDAR Point Cloud Clustering | |
| KR20100034409A (en) | Method for recognizing lane and lane departure with single lane extraction | |
| JP2022089828A (en) | Information processing apparatus | |
| CN108828608A (en) | Laser radar background data filtering method in vehicle checking method | |
| Rasmussen et al. | On-vehicle and aerial texture analysis for vision-based desert road following | |
| Poggenhans et al. | A universal approach to detect and classify road surface markings | |
| CN115359453A (en) | A method and system for road obstacle detection based on 3D point cloud | |
| CN110119751B (en) | Laser radar point cloud target segmentation method, target matching method, device and vehicle |
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 |