CN114089376A - Single laser radar-based negative obstacle detection method - Google Patents
Single laser radar-based negative obstacle detection method Download PDFInfo
- Publication number
- CN114089376A CN114089376A CN202111161457.7A CN202111161457A CN114089376A CN 114089376 A CN114089376 A CN 114089376A CN 202111161457 A CN202111161457 A CN 202111161457A CN 114089376 A CN114089376 A CN 114089376A
- Authority
- CN
- China
- Prior art keywords
- points
- point
- point cloud
- plane
- laser radar
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 claims abstract description 24
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000007613 environmental effect Effects 0.000 claims description 6
- 230000004888 barrier function Effects 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 abstract description 10
- 238000000605 extraction Methods 0.000 abstract description 3
- 238000005286 illumination Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
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)
Abstract
The invention discloses a negative obstacle detection method based on point cloud characteristics of a single laser radar, which is suitable for a structured road environment. The method comprises the following steps: firstly, superposing multi-frame point cloud data of a single radar according to the motion pose of the laser radar to obtain point clouds fusing different time and space positions; and calculating the curvature of the fused point cloud, and dividing the point cloud into plane points and angular points according to the curvature. And then, carrying out ground extraction on the plane points to obtain a fitted ground plane equation. And extracting the negative obstacle point cloud according to the geometrical characteristics of the point cloud, such as curvature, height, distance jump and the like. And filtering and clustering on the basis of the obstacle point cloud to obtain the final outline and position of the obstacle. The invention has good real-time performance and higher detection success rate for middle and small-sized obstacles.
Description
Technical Field
The invention belongs to the field of unmanned environment perception, and particularly relates to a negative obstacle detection method used in a road environment.
Background
The unmanned vehicle may face various environmental obstacles during driving, and may be generally classified into a positive obstacle and a negative obstacle according to a relationship with respect to the ground. The positive barrier protrudes from the ground and is an actually existing object, such as a pedestrian, a vehicle, a traffic cone, and the like; negative obstacles refer to parts of the ground that are partially missing, such as waterways, ditches, and the like. The negative direction barrier has no less threat to the unmanned vehicle than the positive direction barrier, so the negative direction barrier detection has important significance to the safe driving of the unmanned vehicle.
The research aiming at the positive obstacle detection is relatively deep, a mature algorithm frame and a detection flow are formed, however, the attention obtained by the negative obstacle detection is less, and the minimum detection size, the detection precision and the robustness of the existing method need to be improved.
The detection method based on vision cannot be applied to scenes with poor illumination such as night and foggy days, and the detection position precision is not high. The laser radar has the advantages of long detection distance, high measurement precision and the like, does not depend on environmental illumination, and has higher reliability, so that the laser radar is more applied to negative-direction obstacle detection.
Due to the fact that shielding and the vertical resolution of the laser radar line beam are low, the coverage rate of the laser radar point cloud to the negative obstacle is low. The existing negative obstacle detection method based on the laser radar generally depends on two or more groups of obliquely installed laser radars to cover a target area, then calculates the geometric and statistical characteristics of local point cloud, and finally extracts the area with obstacles through clustering. The disadvantages of this type of approach are the high cost, the small detection area and the problems of sensor calibration and data synchronization.
Disclosure of Invention
The invention provides a negative obstacle detection method based on a single laser radar aiming at the defects of the existing method.
The technical scheme provided by the invention is as follows: a negative obstacle detection method based on a single laser radar comprises the following steps:
step 1, collecting environmental point cloud data by using a laser radar, and simultaneously acquiring a linear velocity and an angular velocity of a motion from an inertial sensor of a vehicle; and superposing the single-frame laser point clouds at a plurality of adjacent moments according to the pose state of the vehicle carrier to obtain fused high-density point clouds.
Further, in step 1, a multi-line laser radar of a rescan type is used. The repeated scanning type laser radar has a constant scanning angle and period, and can form a continuous, regularly arranged beam.
Further, in the step 1, the movement speed of the laser radar can be obtained from an inertial sensor installed on the vehicle, the poses of the laser radar at different moments can be obtained through integration, and the coordinates of point clouds in different frames are transformed relative to the poses of the laser radar to obtain point cloud coordinates in the same coordinate system, so that the point clouds in one frame can be superposed.
Further, in step 1, if the vehicle has no inertial sensor for measuring speed and cannot provide radar pose, the pose of the laser radar at the adjacent time can be calculated by interframe matching by using a laser odometer method, for example, a method proposed in the article LOAM (Lidar Odometry and Mapping in Real-time).
And 2, calculating the curvature of each point according to the fused point cloud obtained in the step 1, wherein the curvature reflects the position information of the point. Points in the point cloud are divided into two types according to the curvature: planar points and angular points.
Further, in step 2, the curvature of each point is calculated using the following formula.
In step 2, the curvature of the point is calculated by using the adjacent points on the single scanning line, and the specific calculation method is shown in the following formula:
in the formula, ciIs the curvature of the ith point in the point cloud, n is the number of adjacent points, piAnd pkRespectively the coordinates of the ith point and the kth point in the point cloud, wherein k is a traversal subscript;
setting a plane curvature threshold cplaneAnd corner curvature threshold cedgeC is mixingi≤cplaneIs stored as a point cloud, which is marked as a plane point, ci≥cedgeThe points of (a) are stored as a point cloud, which is denoted as an angular point.
Further, in step 2, two threshold values are set to divide the points, and a plane curvature threshold value c is setplaneAnd corner curvature threshold cedge。ci≤cplaneIs located on the ground or inside the building surface, stored as a point cloud, noted as a plane point, ci≥cedgeThe points of (a) are stored as a point cloud, which is denoted as an angular point.
And 3, assuming that the road on which the vehicle runs is a flat road surface, fitting a plane equation of the ground according to the plane points by adopting an RANSAC method.
Further, in step 3, only plane points are used and no angular points are used for road plane fitting;
further, in step 3, the filtering in the height direction is performed first. The vertical coordinate of the ith laser reflection point in the laser radar coordinate system is assumed to be ziAccording to the installation height H of the laser radar, the condition of-H-H is extracted1≤zi≤-H+h2A point of (a); wherein the threshold value range is not less than 0 and not more than h1≤H，0≤h2H is less than or equal to H. Threshold value h1The method is used for filtering out points which are higher than the ground, do not belong to the ground, and are removed before ground extraction; threshold value h2For filtering out specular reflection points due to mishandling of the lidar, the coordinates of these points lying below ground, beingTo interference.
Further, in step 3, plane fitting is performed on the plane points remaining after filtering by using RANSAC. The equation for the fitted plane takes the form ax + by + cz + d as 0, and the plane normal vector is (a, b, c).
And 4, calculating the distance jump of the adjacent points, and extracting the points meeting the design threshold.
Further, in step 4, distance jump of adjacent corner points is calculated line by line, and the calculation method is as follows: firstly, ordering the angular points according to line numbers and angle sequences, namely dividing the angular points into different scanning lines according to the wire harness number information carried by each angular point, calculating the horizontal angle of the angular point on each scanning line, and ordering the angular points on each scanning line from small to large according to the angle; then, sequentially calculating the distance from each point to the origin of the laser radar coordinate system on the sequenced angular points; finally, the distance difference d between adjacent corner points is calculatedsIf d issGreater than a designed threshold dthAnd the horizontal angle difference of adjacent corner points is less than an angle threshold value thetathThen the corner point is likely to be a boundary point of the obstacle and is therefore stored as a boundary point.
Further, in step 4, distance clustering is performed according to the calculated boundary points, and an area where a negative obstacle may exist is obtained. Extracting all points in the region, calculating the horizontal angle of the points, dividing the points into different groups according to the preset horizontal angle resolution, and sequencing the points in each group from small to large according to the line number. Calculating the height difference of adjacent points for each group of points, and projecting the obtained plane normal vector (a, b, c) along the ground to obtain delta h, if delta h is satisfied>hthrIs more than a preset ratio, the area is classified as a negative obstacle area, hthrIs a designed threshold.
And 5, clustering the extracted negative obstacle points, and solving the minimum enclosing polygon of each cluster.
Further, in step 5, clusters are formed using a euclidean distance based clustering method.
Further, in step 5, a minimum bounding polygon of the corner points in each cluster is calculated, and the geometric center coordinates of the vertices of the polygons represent the spatial position of the obstacle.
Compared with the prior art, the technical scheme of the invention has the following advantages:
in the existing method, the density of point clouds is used as a characteristic, but the point cloud density in the actual point cloud is greatly influenced by distance, so that the characteristics of far sparsity and near density are generated, the density of the point clouds obtained by superposing multiple frames is not uniform, and the density of the side surface of a negative obstacle is not obvious. The method adopts the geometrical characteristics of point curvature, distance jump, height difference and the like, and is not influenced by the distance and density change; secondly, the invention can realize the detection of the negative obstacle by only adopting a single multi-line laser radar without the data synchronization and calibration work of a plurality of sensors.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a single lidar-based negative obstacle detection method according to the present invention;
FIG. 2 is a schematic diagram illustrating a detection result of a negative obstacle at a sewer opening in an embodiment;
FIG. 3 is a time-consuming statistical chart of the detection procedure based on the detection method of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the following explains the technical solutions of the present invention by taking a sewer opening, which is commonly found in urban roads, as an example.
The environmental data collection used in this embodiment is from 128 lines of rotation type machinery lidar, installs in the roof of the motor vehicle, and the rotation axis is perpendicular with ground. The point cloud output by the lidar is shown in fig. 2.
The embodiment of the invention provides a negative obstacle detection method based on a single laser radar, which comprises the following steps:
step 1, collecting environmental point cloud data by using a laser radar, and simultaneously acquiring a linear velocity and an angular velocity of a motion from an inertial sensor of a vehicle; and superposing the single-frame laser point clouds at a plurality of adjacent moments according to the pose state of the vehicle carrier to obtain fused high-density point clouds.
The present embodiment uses a rescan-type multiline lidar having a constant scanning angle and period, capable of forming a continuous, regularly arranged line beam. And estimating the pose of the laser radar at the adjacent moment by adopting a three-dimensional laser odometer, carrying out coordinate transformation on the multi-frame original point cloud in the queue, and uniformly converting the multi-frame original point cloud into a coordinate system of the first frame of laser radar in the queue. Designing an annular queue to store a plurality of frames of point clouds, adding the point clouds into the queue after receiving a new point cloud frame, and deleting the earliest added frame after the queue exceeds a preset length. The present embodiment sets the queue length to 4, i.e., superimposes the adjacent four frames of laser point clouds.
Further, in the step 1, the movement speed of the laser radar can be obtained from an inertial sensor installed on the vehicle, the poses of the laser radar at different moments can be obtained through integration, and the coordinates of point clouds in different frames are transformed relative to the poses of the laser radar to obtain point cloud coordinates in the same coordinate system, so that the point clouds in one frame can be superposed.
Further, in step 1, if the vehicle has no inertial sensor for measuring speed and cannot provide radar pose, the pose of the laser radar at the adjacent time can be calculated by interframe matching by using a laser odometer method, for example, a method proposed in the article LOAM (Lidar Odometry and Mapping in Real-time).
And 2, calculating the curvature of each point according to the fused point cloud obtained in the step 1, wherein the curvature reflects the position information of the point. Points in the point cloud are divided into two types according to the curvature: planar points and angular points.
Further, in step 2, the curvature of the point is calculated by using the adjacent points on the single scan line, and the specific calculation method is shown in the following formula:
in the formula, ciIs the curvature of the ith point in the point cloud, n is the number of adjacent points, piAnd pkRespectively the coordinates of the ith point and the kth point in the point cloud, wherein k is a traversal subscript, and the traversal range is from i-n/2 to i + n/2;
setting a plane curvature threshold cplaneAnd corner curvature threshold cedgeC is mixingi≤cplaneIs stored as a point cloud, which is marked as a plane point, ci≥cedgeThe points of (a) are stored as a point cloud, which is denoted as an angular point.
Further, in step 2, two threshold values are set to divide the points, and a plane curvature threshold value c is setplaneAnd corner curvature threshold cedge。ci≤cplaneIs located on the ground or inside the building surface, stored as a point cloud, noted as a plane point, ci≥cedgeThe points of (a) are stored as a point cloud, which is denoted as an angular point.
Calculating the curvature of each point according to the fused point cloud, and the plane curvature threshold c of the embodimentplaneSet to 0.001, corner curvature threshold cedgeSet to 0.01.
And 3, assuming that the road on which the vehicle runs is a flat road surface, fitting a plane equation of the ground according to the plane points by adopting an RANSAC method.
Further, in step 3, only plane points are used and no angular points are used for road plane fitting;
further, in step 3, the filtering in the height direction is performed first. The vertical coordinate of the ith laser reflection point in the laser radar coordinate system is assumed to be ziAccording to the installation height H of the laser radar, the condition of-H-H is extracted1≤zi≤-H+h2A point of (a); wherein the threshold value range is not less than 0 and not more than h1≤H，0≤h2H is less than or equal to H. Threshold value h1For filtering out points at excessive height above ground, which do not belong to the ground and should be rejected before ground extraction(ii) a Threshold value h2The method is used for filtering mirror reflection points caused by error processing of the laser radar, and coordinates of the points are located below the ground and belong to interference.
Further, in step 3, plane fitting is performed on the plane points remaining after filtering by using RANSAC. The equation of the fitting plane takes the form of ax + by + cz + d being 0, and the plane normal vector is (a, b, c); filtering threshold h1Set to 0.4, threshold h2Set to 0.5.
And 4, calculating the distance jump of the adjacent points, and extracting the points meeting the design threshold.
Further, in step 4, distance jump of adjacent corner points is calculated line by line, and the calculation method is as follows: firstly, ordering the angular points according to line numbers and angle sequences, namely dividing the angular points into different scanning lines according to the wire harness number information carried by each angular point, calculating the horizontal angle of the angular point on each scanning line, and ordering the angular points on each scanning line from small to large according to the angle; then, sequentially calculating the distance from each point to the origin of the laser radar coordinate system on the sequenced angular points; finally, the distance difference d between adjacent corner points is calculatedsE.g. thetathThen the corner point is likely to be a boundary point of the obstacle and is therefore stored as a boundary point. Distance threshold dthIs set to dth＝1.2·riΔ θ, where riIs the distance from the current point to the lidar origin, and Δ θ is the vertical angular resolution of the lidar.
Further, in step 4, distance clustering is performed according to the calculated boundary points, and an area where a negative obstacle may exist is obtained. Extracting all points in the region, calculating the horizontal angle of the points, dividing the points into different groups according to the preset horizontal angle resolution, and sequencing the points in each group from small to large according to the line number. Calculating the height difference of adjacent points for each group of points, and projecting the obtained plane normal vector (a, b, c) along the ground to obtain delta h, if delta h is satisfied>hthrIs more than a preset ratio, the area is classified as a negative obstacle area, hthrIs a designed threshold.
Calculating the distance jump of adjacent angular points line by line according to the calculationAnd performing distance clustering on the boundary points to obtain an area where negative obstacles possibly exist. Extracting all points in the region, calculating the horizontal angle of the points, dividing the points into different groups according to the preset horizontal angle resolution, and sequencing the points in each group from small to large according to the line number. Calculating the height difference of adjacent points for each group of points, and projecting the obtained plane normal vector (a, b, c) along the ground to obtain delta h, if delta h is satisfied>hthrIs more than a preset ratio, the area is classified as a negative obstacle area, and a threshold hthrSet to 0.05 and the preset ratio to 10%.
And 5, clustering the extracted negative obstacle points, and solving the minimum enclosing polygon of each cluster, namely calculating the enclosing polygon of the obstacle point cloud, as shown in fig. 2. In this embodiment, clusters are formed by using a clustering method based on euclidean distance, a minimum bounding polygon of a corner point in each cluster is calculated, and geometric center coordinates of vertices of the polygons represent spatial positions of obstacles.
A continuous detection test is performed on a single negative obstacle (sewer opening), wherein the diameter of the obstacle is 0.5m, the ground level of the environment is horizontal, the height of the laser radar from the ground is 1.8m, the moving speed of the unmanned vehicle carrying the laser radar is about 0.5m/s, the consumed time of a negative obstacle detection procedure (without a laser odometer part) is shown in fig. 3, and the average consumed time is 60 ms.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and adjustments can be made without departing from the principle of the present invention, and these modifications and adjustments should also be regarded as the protection scope of the present invention.
Claims (7)
1. A negative obstacle detection method based on a single laser radar is characterized by comprising the following steps:
step 1, collecting point cloud data of a road environment by using a laser radar, and acquiring the linear velocity and the angular velocity of the movement of a vehicle carrier; superposing the single-frame laser point cloud data of a plurality of adjacent moments with the pose state of the vehicle carrier to obtain fused high-density point cloud data;
step 2, calculating the curvature of each point based on the high-density point cloud data, and dividing the points in the point cloud into plane points and angular points according to the curvature;
step 3, performing plane fitting on the plane points to obtain a plane equation of a plane where the road is located;
step 4, calculating distance jump of adjacent points by diagonal points and plane points, and extracting points meeting a design threshold;
and 5, clustering the extracted points meeting the design threshold, and calculating a surrounding polygon on the basis of the clustering points to obtain the shape and the position of the negative barrier.
2. The method for detecting the negative obstacle based on the single laser radar is characterized in that in the step 1, a single repeated scanning type multi-line laser radar is used for acquiring environmental data, and the laser radar is arranged on a vehicle carrier; the vehicle carrying system is characterized in that an IMU inertial sensor is mounted on the vehicle carrying system and used for collecting the linear motion velocity and the rotation angular velocity of the vehicle carrying system in real time.
3. The method for detecting the negative obstacle based on the single laser radar as claimed in claim 1, wherein in the step 2, the curvature of the point is calculated by using the adjacent points on the single scanning line, and the specific calculation method is shown in the following formula:
in the formula, ciIs the curvature of the ith point in the point cloud, n is the number of adjacent points, piAnd pkRespectively the coordinates of the ith point and the kth point in the point cloud, wherein k is a traversal subscript;
setting a plane curvature threshold cplaneAnd corner curvature threshold cedgeC is mixingi≤cplaneIs stored as a point cloud, which is marked as a plane point, ci≥cedgeIs stored as a point cloud, denoted as an angleAnd (4) point.
4. The method of claim 1, wherein in step 3, a plane equation of the ground is fitted according to the plane points by using RANSAC method.
5. The method for detecting negative obstacle based on single laser radar as claimed in claim 3, wherein in step 4, the distance jump of the adjacent corner points is calculated line by line, and the calculation method is as follows:
firstly, ordering the angular points according to line numbers and angle sequences, namely dividing the angular points into different scanning lines according to the wire harness number information carried by each angular point, calculating the horizontal angle of the angular point on each scanning line, and ordering the angular points on each scanning line from small to large according to the angle;
then, sequentially calculating the distance from each angular point to the origin of the laser radar coordinate system at the sequenced angular points;
finally, the distance difference d between adjacent corner points is calculatedsIf d issGreater than a designed threshold dthAnd the horizontal angle difference of adjacent corner points is less than an angle threshold value thetathThen the corner point may be a boundary point of the obstacle and stored as a boundary point.
6. The method for detecting the negative obstacle based on the single laser radar as claimed in claim 5, wherein in the step 4, distance clustering is performed according to the calculated boundary points to obtain an area where the negative obstacle may exist;
extracting all points in the region, calculating horizontal angles of the points, dividing the points into different groups according to a preset horizontal angle resolution, and sequencing the points in each group from small to large according to line numbers;
calculating the height difference of adjacent points and projecting the obtained plane normal vector (a, b, c) along the ground to obtain delta h, if delta h is satisfied>hthrIs more than a preset ratio, the area is classified as a negative obstacle area, hthrIs a set threshold.
7. The method according to claim 1, wherein in step 5, a minimum bounding polygon is calculated for the corner points in the negative obstacle region, and the center coordinates of the polygon represent the spatial position of the obstacle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111161457.7A CN114089376A (en) | 2021-09-30 | 2021-09-30 | Single laser radar-based negative obstacle detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111161457.7A CN114089376A (en) | 2021-09-30 | 2021-09-30 | Single laser radar-based negative obstacle detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114089376A true CN114089376A (en) | 2022-02-25 |
Family
ID=80296463
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111161457.7A Pending CN114089376A (en) | 2021-09-30 | 2021-09-30 | Single laser radar-based negative obstacle detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114089376A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115876098A (en) * | 2022-12-12 | 2023-03-31 | 苏州思卡信息系统有限公司 | Vehicle size measuring method of multi-beam laser radar |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650640A (en) * | 2016-12-05 | 2017-05-10 | 浙江大学 | Negative obstacle detection method based on local structure feature of laser radar point cloud |
| CN110109145A (en) * | 2019-04-30 | 2019-08-09 | 北京云迹科技有限公司 | Running region detection method and device based on multi-line laser radar |
| CN110221603A (en) * | 2019-05-13 | 2019-09-10 | 浙江大学 | A kind of long-distance barrier object detecting method based on the fusion of laser radar multiframe point cloud |
| CN110244321A (en) * | 2019-04-22 | 2019-09-17 | 武汉理工大学 | A kind of road based on three-dimensional laser radar can traffic areas detection method |
| CN110441791A (en) * | 2019-08-14 | 2019-11-12 | 深圳无境智能机器人有限公司 | A kind of ground obstacle detection method based on the 2D laser radar that leans forward |
| WO2019242174A1 (en) * | 2018-06-21 | 2019-12-26 | 华南理工大学 | Method for automatically detecting building structure and generating 3d model based on laser radar |
| CN113176585A (en) * | 2021-04-14 | 2021-07-27 | 浙江工业大学 | Three-dimensional laser radar-based road surface anomaly detection method |
-
2021
- 2021-09-30 CN CN202111161457.7A patent/CN114089376A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650640A (en) * | 2016-12-05 | 2017-05-10 | 浙江大学 | Negative obstacle detection method based on local structure feature of laser radar point cloud |
| WO2019242174A1 (en) * | 2018-06-21 | 2019-12-26 | 华南理工大学 | Method for automatically detecting building structure and generating 3d model based on laser radar |
| CN110244321A (en) * | 2019-04-22 | 2019-09-17 | 武汉理工大学 | A kind of road based on three-dimensional laser radar can traffic areas detection method |
| CN110109145A (en) * | 2019-04-30 | 2019-08-09 | 北京云迹科技有限公司 | Running region detection method and device based on multi-line laser radar |
| CN110221603A (en) * | 2019-05-13 | 2019-09-10 | 浙江大学 | A kind of long-distance barrier object detecting method based on the fusion of laser radar multiframe point cloud |
| CN110441791A (en) * | 2019-08-14 | 2019-11-12 | 深圳无境智能机器人有限公司 | A kind of ground obstacle detection method based on the 2D laser radar that leans forward |
| CN113176585A (en) * | 2021-04-14 | 2021-07-27 | 浙江工业大学 | Three-dimensional laser radar-based road surface anomaly detection method |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115876098A (en) * | 2022-12-12 | 2023-03-31 | 苏州思卡信息系统有限公司 | Vehicle size measuring method of multi-beam laser radar |
| CN115876098B (en) * | 2022-12-12 | 2023-10-24 | 苏州思卡信息系统有限公司 | Vehicle size measurement method of multi-beam laser radar |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111352112B (en) | Target detection method based on vision, laser radar and millimeter wave radar | |
| CN109961440B (en) | Three-dimensional laser radar point cloud target segmentation method based on depth map | |
| CN110320504B (en) | Unstructured road detection method based on laser radar point cloud statistical geometric model | |
| CN108647646B (en) | Low-beam radar-based short obstacle optimized detection method and device | |
| CN106780524B (en) | Automatic extraction method for three-dimensional point cloud road boundary | |
| CN111046776B (en) | Method for detecting obstacle of path of mobile robot based on depth camera | |
| CN105404844B (en) | A kind of Method for Road Boundary Detection based on multi-line laser radar | |
| Zhao et al. | Reconstructing a textured CAD model of an urban environment using vehicle-borne laser range scanners and line cameras | |
| JP5820774B2 (en) | Road boundary estimation apparatus and program | |
| CN109752701A (en) | A kind of road edge detection method based on laser point cloud | |
| CN113177593B (en) | Fusion method of radar point cloud and image data in water traffic environment | |
| CN114821526A (en) | Obstacle three-dimensional frame detection method based on 4D millimeter wave radar point cloud | |
| CN111640323A (en) | Road condition information acquisition method | |
| CN114325634A (en) | Method for extracting passable area in high-robustness field environment based on laser radar | |
| CN116109601A (en) | Real-time target detection method based on three-dimensional laser radar point cloud | |
| CN115797640A (en) | Road boundary extraction method for strip mine area | |
| CN113030997B (en) | Method for detecting travelable area of open-pit mine area based on laser radar | |
| CN115151954A (en) | Method and device for detecting a drivable region | |
| CN114089376A (en) | Single laser radar-based negative obstacle detection method | |
| CN115267827A (en) | Laser radar harbor area obstacle sensing method based on height density screening | |
| CN112348950B (en) | Topological map node generation method based on laser point cloud distribution characteristics | |
| CN111239761B (en) | Method for indoor real-time establishment of two-dimensional map | |
| CN113554705A (en) | Robust positioning method for laser radar in changing scene | |
| CN112197773A (en) | Visual and laser positioning mapping method based on plane information | |
| CN111414848B (en) | Full-class 3D obstacle detection method, system and medium |
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 |