CN108983248A - It is a kind of that vehicle localization method is joined based on the net of 3D laser radar and V2X - Google Patents
It is a kind of that vehicle localization method is joined based on the net of 3D laser radar and V2X Download PDFInfo
- Publication number
- CN108983248A CN108983248A CN201810667814.9A CN201810667814A CN108983248A CN 108983248 A CN108983248 A CN 108983248A CN 201810667814 A CN201810667814 A CN 201810667814A CN 108983248 A CN108983248 A CN 108983248A
- Authority
- CN
- China
- Prior art keywords
- radar
- barrier
- points
- net
- 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.)
- Pending
Links
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
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/06—Systems determining position data of a target
-
- 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
-
- 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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/495—Counter-measures or counter-counter-measures using electronic or electro-optical means
Abstract
Vehicle localization method is joined based on the net of 3D laser radar and V2X the invention discloses a kind of, this method is using the environment around the laser radar scanning net connection vehicle being mounted on net connection vehicle to obtain radar data；The radar data is pre-processed, pretreated laser point cloud data is obtained, point cloud data is transformed into net connection vehicle coordinate system, then according to the variable density feature of radar points in point cloud data, point cloud segmentation is carried out using grating map, to realize the separation on road surface and barrier；The cluster for carrying out barrier, obtains the corresponding radar point set of each barrier, then obtains the profile of barrier, then carries out linear fit, obtain the shape of barrier, and the identification of barrier is realized in conjunction with the size of barrier；Realize that obstacle information exchanges between other net connection vehicles finally by V2X module.The present invention solves the problems, such as that existing vehicle positioning method accuracy is low, poor reliability, while the problem of also alleviate, both expensive high to the rate request of mobile communications network.
Description
Technical field
The present invention relates to car networking vehicle location detection technique fields, more particularly to one kind to be based on 3D laser radar and V2X
Net join vehicle location detection method.
Background technique
Currently, the accuracy and reliability of net connection vehicle positioning method determines that user obtains net connection in real time and drives vehicle location
The accuracy rate of information.Accuracy is high and the vehicle location information of high reliablity can accurately determine net join between vehicle away from
From, avoid the accidents such as vehicle collision, thus realize intelligent network join traffic.
Current common vehicle positioning method is GPS (global positioning system), however, existing system is believed by sky satellite
It number influences, and many factors such as high-rise, overpass, electric wave influence whether satellite positioning signal, therefore, in high level
In the more urban environment of building, GNSS system is tended not to accurate reliable or is steadily worked, meanwhile, it can not also apply
In more closed indoor environment.
Summary of the invention
The present invention provide it is a kind of based on the net of 3D laser radar and V2X join vehicle location detection method and device, with solve
The problem of existing vehicle positioning method accuracy is low, poor reliability, while also alleviating and the speed of mobile communications network is wanted
The problem of asking height, both expensive.
In order to realize above-mentioned task, the invention adopts the following technical scheme:
It is a kind of that vehicle localization method is joined based on the net of 3D laser radar and V2X, comprising the following steps:
Using the environment around the laser radar scanning net connection vehicle being mounted on net connection vehicle to obtain radar data；
The radar data is pre-processed, radar point cloud data is obtained, radar points cloud data are transformed into net connection
In vehicle coordinate system, then according to the variable density feature of radar points in radar points cloud, point cloud segmentation is carried out using grating map, with
Realize the separation on road surface and barrier；
The cluster that barrier is carried out using the method for connected component labeling, obtains the corresponding radar points of each barrier
Collection obtains the profile of barrier using radar point set, then carries out linear fit to profile, obtains the shape of barrier, in conjunction with
The size of barrier realizes the identification of barrier；
Realize that obstacle information exchanges between other net connection vehicles using the V2X module on net connection vehicle.
Further, described that the radar data is pre-processed, obtain radar point cloud data, comprising:
In the radar data, filtering out with laser radar distance is zero or beyond the thunder within the scope of laser radar range
Median filtering is carried out up to point, and to the radar points in every scan line of laser radar, to filter out abnormal reflection point, to obtain
Radar point cloud data.
It is further, described that radar point cloud data is transformed into net connection vehicle coordinate system, comprising:
The radar point cloud data is made of radar points, and the coordinate representation of each radar points is (ρ, α, θ), wherein ρ
Radar points are indicated to the distance of laser radar, α is the azimuth of radar points, and θ indicates the elevation angle of radar points；
Error correction is carried out according to the internal reference file of laser radar, by the right angle of the coordinate transformation of radar points to laser radar
In coordinate system, the positional relationship of rectangular coordinate system and net connection vehicle coordinate system further according to laser radar converts the coordinate of radar points
Into net connection vehicle coordinate system.
Further, the variable density feature according to radar points in radar point cloud data, using grating map into
Row point cloud segmentation, to realize the separation on road surface and barrier, comprising:
Grating map is established, the radar points in net connection vehicle coordinate system are projected in grating map；
The radar points projected in grating map are filtered, if radar points in some grid number less than 2 if to this
Radar points in grid are filtered out；If the number of radar points is greater than 2 in some grid, calculating, which projects in the grid, owns
The grid tag is barrier grid if H is greater than 20cm by the difference H of the maxima and minima of the standoff height of radar points
Lattice；If within the scope of the 7*7 grid centered on barrier grid, without other barrier grids, then the barrier grid is isolated
Barrier grid filters out the radar points in the isolated danger grid；
The gradient value for calculating radar points, if the gradient value of radar points is more than preset threshold value, then by radar points labeled as just
Beginning obstacle object point, to realize the separation of road surface and barrier.
Further, the gradient value of the calculating radar points, comprising:
The height of the laser radar two radar points adjacent in adjacent two scan lines of same scanning direction
Difference is denoted as Δ z, and described two radar points are distance in the distance of horizontal plane, then is located at the gradient value of the radar points in outside
Calculation formula are as follows:
Further, the method using connected component labeling carries out the cluster of barrier, comprising:
The traversal grating map, finds first un-marked barrier grid, is marked for the first time, in the barrier
Hinder other barrier grids of searching in 8 neighborhoods of object grid similarly to be marked, the position of all obstacle grids is recorded, by this
The radar points of a little positions merge, and obtain the corresponding radar point set of connected domain；
The corresponding radar point set of other connected domains is obtained in the same manner.
Further, the profile that barrier is obtained using radar point set, comprising:
A bianry image is established, the length of bianry image and wide pixel number correspond respectively to length and the width institute of grating map
The grid number for including；All radar point sets are mapped in bianry image, sector scanning is carried out to bianry image and retrieves profile,
To obtain the profile of each barrier.
Further, described that linear fit is carried out to profile, obtain the shape of barrier, comprising:
Remember that the collection of the corresponding radar points of the profile of a barrier is combined into P=(p1,p2... pn), by the starting point of the set
p1With terminal pnIt is connected with straight line, the distance d of the every bit in set of computations to the straight linei, by comparing acquisition maximum distance
dj, judge the distance and former given threshold DthSize, if more than the threshold value, then with point pjFor breakpoint, it is two that set P, which is divided to,
A subset P1=(p1,p2... pj) and P2=(pj,pj+1... pn)；Above step is repeated to subset, until the point in set P
Distance value to straight line is respectively less than threshold value, completes segmentation；
For the set after segmentation, the minimum area-encasing rectangle side of the corresponding barrier of the set is acquired using Minimum Convex Closure method
Frame, to obtain the central point of barrier, length, width.
The present invention has following technical characterstic:
The present invention utilizes advantage of the 3D laser radar sensor in terms of vehicle location under traffic scene, fixed instead of tradition
The method and theory of position, meanwhile, complicated occlusion issue, complex illumination problem for tradition positioning provide solution.Together
The dedicated short-range communication technology of Shi Liyong V2X carries out the two-way communication between net connection vehicle, the continuous vehicle obtained in preset range
Velocity information, location information between make driver's traveling make full use of path space, avoid that crowded and blocking occurs.
Detailed description of the invention
Fig. 1 is the flow chart of the method for the present invention；
Fig. 2 a is the schematic diagram for the rectangular coordinate system that radar data is transformed into laser radar；
Fig. 2 b is the schematic diagram that radar points are transformed into net connection vehicle coordinate system from the rectangular coordinate system of laser radar；
Fig. 3 a is the schematic diagram for the original radar points cloud that laser radar obtains；
Fig. 3 b is the schematic diagram of the radar points cloud after over-segmentation；
Fig. 4 is the schematic diagram for carrying out barrier cluster；
Fig. 5 is the schematic diagram for getting barrier profile；
Fig. 6 is the schematic diagram of detection of obstacles result；
Fig. 7 is the flow diagram that net connection vehicle interacts.
Specific embodiment
Particular content of the invention is described in detail below in conjunction with attached drawing.As shown in Figure 1, the invention discloses one kind
Join vehicle localization method based on the net of 3D laser radar and V2X, comprising the following steps:
Step 1, using the environment around the laser radar scanning net connection vehicle being mounted on net connection vehicle to obtain radar data；
In the present invention, laser radar is installed on net connection vehicle, which can be used such as Velodyne HDL-
32E laser radar, the radar is small by external environmental interference, effective detection range 100m, vertical field of view be -30.47 °~+
10.47 °, laser radar can obtain 360 ° of rings by rotation by 32 single laser constitutions on a column, 32 scan lines
Border information.Net joins vehicle in the process of moving, by laser radar scanning ambient enviroment to obtain radar data.As shown in Fig. 2, O
For coordinate origin, i.e. radar site；P is target position.Radar data is made of radar points, the coordinate of each radar points
Pass through polar coordinate representation: (ρ, α, θ), wherein ρ indicates the distance that radar detection is arrived, i.e. the distance between radar points and radar, α table
Show azimuth of the radar points relative to radar, θ indicates the elevation angle of the radar points relative to radar.
Step 2, the radar data is pre-processed, pretreated radar point cloud data is obtained, by radar points
Cloud data are transformed into net connection vehicle coordinate system, then according to the variable density feature of radar points in radar point cloud data, utilize grid
Lattice map carries out point cloud segmentation, to realize the separation on road surface and barrier；Specifically includes the following steps:
Step 2.1, described that the radar data is pre-processed, obtain radar point cloud data
The emissive porwer and object that the range measurement of laser radar depends primarily on laser are therefore right to the reflectivity of laser
There are measurement errors for unlike material and the object of color, and there may be invalid to swash in the laser data of return
Light data.Specific manifestation are as follows: there are some distances to be zero or have exceeded the maximum measure distance range of laser radar, there are also it is some due to
Physical cause causes the point of abnormal reflection.
In the radar data, filtering out with laser radar distance ρ is zero or beyond within the scope of laser radar range
Radar points, and median filtering is carried out to the radar points in every scan line of laser radar, to filter out abnormal reflection point, thus
To pretreated radar point cloud data.
By being pre-processed to radar data, corrects due to laser radar installation, calibration etc., ground is caused to throw
Shadow angle and the skimble-scamble problem of initial data scanning angle.
Step 2.2, pretreated radar points cloud data are transformed into net connection vehicle coordinate system
The coordinate that the three-dimensional laser radar that the present invention uses returns to radar points is to be with polar form (ρ, α, θ) storage
It need to first be transformed into the rectangular coordinate system of laser radar by subsequent algorithm, and the rectangular coordinate system and net further according to laser radar join
The coordinate of radar points is transformed into net connection vehicle coordinate system by the positional relationship of vehicle coordinate system.
Since there are certain offsets with radar fix system for radar ray receiving surface, therefore error correction is carried out first:
(1) range correction parameter Dcorr
The distance value D that laser returnsret, it is ρ in polar coordinates, in addition the range correction factor of the laser beam is that this swashs
The real measurement distance of light beam.
(2) vertical offset Voffset
In X-Z plane, the distance of straight line where origin O (radar site center) to laser scanning line.
(3) horizontal offset Hoffset
In X-Y plane, the distance of straight line where origin O (radar site center) to laser scanning line.
Formula used by error correction are as follows:
D=Dret+Dcorr=ρ+Dcorr
X '=ρ sin α cos θ+Hoffset*cosθ
Y '=ρ cos α cos θ-voffset*sinθ
Z '=ρ sin θ+D*sin θ
By error correction, coordinate transformation of the coordinate (ρ, α, θ) of radar points in the rectangular coordinate system of laser radar is
(x ', y ', z '), as shown in Figure 2 a.
The coordinate for the radar points that will transition in the rectangular coordinate system of laser radar is further transformed into net connection vehicle coordinate
In system, conversion formula is as follows:
Since laser radar is mounted on net connection roof, the rectangular coordinate system and net connection vehicle coordinate system of laser radar exist
There are deviations on pitch angle.As shown in Figure 2 b, the origin O of net connection vehicle coordinate system is the rectangular coordinate system origin O ' of laser radar
The point on ground is projected to along vertical direction, the X-axis of net connection vehicle coordinate system is along vehicle heading, and Y-axis is along X-axis in horizontal plane
The direction being inside rotated by 90 ° counterclockwise, Z axis are to face upward perpendicular to level.In above formula, α, beta, gamma is respectively to net connection vehicle coordinate system
Rectangular coordinate system relative to laser radar is along X, Y, the rotation angular deviation of Z-direction.It, can be by radar points by above-mentioned formula
Coordinate (x ', y ', z ') is further converted in net connection vehicle coordinate system, and the coordinate representation after conversion is (x, y, z).
Step 2.3, according to the variable density feature of radar points in radar point cloud data, a cloud minute is carried out using grating map
It cuts, to realize the separation on road surface and barrier
Although having carried out primary filtration to the data of acquisition, very due to the data volume that is obtained by laser radar
Greatly (about 700,000 point/seconds), if directly being operated in initial data, many memory headrooms and time need to be consumed, it is inconvenient
In processing, therefore grating map is used to carry out detection of obstacles, three-dimensional information is dropped into two-dimensional signal, is largely reduced
The complexity and calculation amount of sensor data analysis, this method is simple and quick, and has good stability and real-time；The step
It is rapid specific as follows:
Step 2.3.1, establishes grating map, and the radar points in net connection vehicle coordinate system are projected in grating map；
The grating map is using face where netting the XY axis of connection vehicle coordinate system as plane, and Z-direction is that height establishes 2.5
The grating map of dimension, the size of grid is G in grating map, the radar points in net connection vehicle coordinate system that then will be obtained after conversion
It projects in grating map, the radar points coordinate representation after projection is (Row, Col, z), and Row, Col respectively indicate radar points and exist
Row row after projection in grating map, Col are arranged in a grid, and z is the height of radar points, value and net connection vehicle coordinate
The z value of coordinate (x, y, z) in system is identical.
In order to ensure the coordinate value after projection, that is, Row, Col are positive value, therefore increase offset on coordinate value
(mapx,mapy), grid conversion formula are as follows:
Row=(y+mapy)/G
Col=(x+mapx)/G
Radar detection area in this programme: preceding 50m, rear 20m, each 20m in left and right.Using grating map 500*200, grid is big
Small is G=20cm*20cm, therefore offset (mapx,mapy) it is (250,200).
Step 2.3.2, obstacle filtering
Due to three-dimensional laser radar data be it is fine and close, radar points are typically all appearance in heaps, are needed in this step pair
Radar points are filtered.
The radar points projected in grating map are filtered, if radar points in some grid number less than 2 if judge
Radar points in grid are isolated point, and are filtered out to the radar points in the grid.
If the number of radar points is greater than 2 in some grid, the projection height for projecting to all radar points in the grid is calculated
The difference H of the maxima and minima of degree:
H=Max_z-Min_z
In above formula, Max_z, Min_z are respectively the height value z of all radar points in grid.If H is greater than 20cm, should
Grid tag is barrier grid.
If within the scope of the 7*7 grid centered on barrier grid, without other barrier grids, then the barrier grid is
Isolated danger grid filters out the radar points in the isolated danger grid.
Step 2.3.3 calculates the gradient value of radar points, if the gradient value of radar points is more than preset threshold value, then by radar
Point is labeled as initial obstacle object-point, to realize the separation of road surface and barrier.
Laser radar is to detect barrier by more scan line rotary scannings, and scan line shows more on smooth ground
Layer annulus shape, as shown in Figure 3a.Radar has 32 scan lines in the present embodiment, in the angle of each scanning, according to vertical
Angle sequence, it is adjacent (perpendicular to thunder in adjacent scanning lines from closely to 32 remote scan lines on the same scanning direction
Distance is nearest on up to the direction of rotary shaft) two radar points are used to calculate gradient value:
The difference in height of note two radar points is denoted as Δ z, and described two radar points are in the distance of horizontal plane
Distance is then located at the calculation formula of the gradient value of the radar points of outside (i.e. farther out with laser radar) are as follows:
By calculating gradient value, the gradient value of the radar points of available different location to radar points in adjacent scanning lines.
The distance of each radar points to radar rotary shaft is determining, if encountering convex obstacle, laser can not pass through barrier, leads to radar
The distance of point to radar rotary shaft is less than its theoretical distance, and distance is made to become smaller；If encountering recessed obstacle, actual scanning distance will
Greater than theoretical distance, distance is made to become larger.Therefore it can use the variation characteristic of barrier distance in scanning element cloud
To detect barrier.
It is more than default by certain radar points gradient values in step 2.3.2 treated barrier grid if be computed
Threshold value th=20cm, then by the radar points be labeled as initial obstacle object-point, to realize the separation of road surface and barrier.
Radar points cloud is split according to above method, Fig. 3 a is the original radar points cloud signal that laser radar obtains
Figure, Fig. 3 b are the result schematic diagram after the step is split radar points cloud.
Step 3, the cluster that barrier is carried out using the method for connected component labeling, obtains the corresponding thunder of each barrier
Up to point set, the profile of barrier is obtained using radar point set, and linear fit then is carried out to profile, obtains the shape of barrier,
The identification of barrier is realized in conjunction with the size of barrier；
Step 3.1, the cluster of barrier is carried out using the method for connected component labeling
Since the initial data that laser radar obtains is sparse discrete, ordinary circumstance lower barrier (such as vehicle, road
Mark, pedestrian etc.) it can be split as many discrete fritters, Direct Classification can not be carried out to barrier.Generally fall into an object
The all the points of body are to be centered around central point to be distributed around, therefore first have to poly- to barrier point obtained after detection progress correlation
The barrier point for belonging to a barrier is brought together by class.
By step 2.3, radar points have been projected in the corresponding grid of grating map.Carry out the process of barrier cluster
Are as follows:
The traversal grating map, finds first un-marked barrier grid, is marked for the first time, in the barrier
Other barrier grids of searching in 8 neighborhoods of object grid are hindered similarly to be marked；
It selects a barrier grid as mark point first, other barrier grids in its 8 neighborhood is marked
Then note carries out 8 neighbourhood signatures respectively as mark point to other barrier grids in 8 neighborhood；Until all obstacles
Without barrier grid that other are not labeled in 8 neighborhoods of object grid；
The position of all obstacle grids is recorded, these barrier grids constitute connected domain, by the radar points of these positions
Merge, obtains the corresponding radar point set of connected domain；
The corresponding radar point set of other connected domains is obtained according to the identical method of step 2.3, one of connected domain is pair
A barrier is answered, as shown in Figure 4.
Step 3.2, the profile of barrier is obtained using radar point set
When the barrier of detection is vehicle, profile is generally straight line or the approximate dual slope in " L ", or is the three of " U "
Broken line.A bianry image is established, the length of bianry image and wide pixel number correspond respectively to the length of grating map and width is wrapped
The grid number contained；All radar point sets that step 3.1 is obtained are mapped in bianry image (if the point is barrier grid
Labeled as 0, otherwise labeled as 255), sector scanning is carried out to bianry image and retrieves profile, to obtain the wheel of each barrier
Exterior feature, as a result as shown in Figure 5.
Step 3.3, linear fit is carried out to profile, obtains the shape of barrier
In field of image processing, the general method using pattern match carries out identification classification to vehicle, but this programme
The object of processing is three-dimensional point cloud, is not suitable for the matched method of direct use pattern.In the present solution, the method used are as follows:
Step 3.3.1 remembers that the collection of the corresponding radar points of the profile of a barrier is combined into P=(p1,p2... pn), n is thunder
Up to the quantity of point；By the starting point p of the set1With terminal pnConnected with straight line, the every bit in set of computations to the straight line away from
From di, by comparing acquisition maximum distance dj, judge the distance and former given threshold DthSize, if more than the threshold value, then with
Point pjFor breakpoint, by set P from breakpoint pjPunishment is two subset P1=(p1,p2... pj) and P2=(pj,pj+1... pn)；
Above step is repeated to subset, until the distance value of point to straight line in set P is respectively less than threshold value, completes segmentation；
Step 3.3.2 acquires the minimum of the corresponding barrier of the set using Minimum Convex Closure method for the set after segmentation
Area-encasing rectangle frame, to obtain the central point of barrier, length, width, as shown in Figure 6.The four of rectangle is used in this programme
Vertex scheme is described, and establishes vehicle identification table, storage form on net connection vehicle are as follows:
Box=[center, width, height, D]
Central point center, width width, high heigh and the central point for separately including rectangle frame join vehicle (laser thunder to net
Up to) distance D.
By step 3.3, the shape and size of barrier have been obtained；Common impairments on road surface are stored in net connection vehicle
Object, such as large car, compact car, pedestrian, building shape (or form range) and size (or size range), then pass through by
Shape, the size of barrier are compared with the shape of common impairments object, size, and the identification of barrier can be realized.
Step 4, using net connection vehicle on V2X (vehicle to everything) module realize with other net connection vehicles it
Between obstacle information exchange.
The shared of connection workshop information is netted to realize, carries out the conversion of net connection vehicle coordinate system to earth coordinates.Net Lian Chean
Data acquired in the laser radar of dress are with the vehicle sheet as coordinate system.However, when net connection workshop information sharing, if making
With respective coordinate system, then location information can not be measured, therefore also need to carry out corresponding coordinate conversion, sat from the net at place connection vehicle
Mark system is transformed into earth coordinates, and the data of such laser radar scanning could really reflect in entire environmental model,
And it is positioned and is navigated.
Earth coordinates described in this programme select BJ54 Coordinate System, and conversion formula is as follows:
x1'=xn+dn·cos(α+β)
y1'=yn+dn·sin(α+β)
Wherein, (x1',y1') indicate coordinate (i.e. step 3.3.2 central point of the barrier identified in world coordinate system
Coordinate in world coordinate system), (xn,yn) indicate the origin for carrying the net connection vehicle coordinate system of laser radar in world coordinate system
In coordinate, α is that Airborne Lidar measures the azimuth of barrier (central point)；dnFor laser radar to barrier (central point)
Distance, β is laser radar to the elevation angle of barrier (central point).We carry out table using two-dimensional earth coordinates herein
To show, has cast out range information, when carrying out information sharing due to net connection workshop, there are certain scope limitations, in the range, institute
Some nets join vehicle and are believed that in same plane, therefore only with (x1',y1') two-dimensional coordinate positioned.
Net connection vehicle in the process of moving, identification, the positioning of barrier is carried out using the above method, then by net connection vehicle
The V2X short distance private communication module (DSRC, Dedicated Short Range Communications) of installation and nearby its
His net connection vehicle interacts, and receives the location information that preset range Intranet connection vehicle is sent, comprising: net connection vehicle itself and laser thunder
The information such as the position of barrier, speed, direction in up to investigative range, and send the vehicle identification table of net connection vehicle itself storage to
Other net connection vehicles nearby, the information sharing being achieved between net connection vehicle, as shown in Figure 7.Utilize V2X Dedicated Short Range Communications
DSRC technology carries out the information sharing between vehicle, using the network the Ad-Hoc model based on multi-hop, carry out end to end in real time, can
It leans on, accurate bidirectional mobile communication.
Claims (8)
1. a kind of join vehicle localization method based on the net of 3D laser radar and V2X, which comprises the following steps:
Using the environment around the laser radar scanning net connection vehicle being mounted on net connection vehicle to obtain radar data；
The radar data is pre-processed, radar point cloud data is obtained, radar points cloud data are transformed into net connection vehicle and are sat
In mark system, then according to the variable density feature of radar points in radar point cloud data, point cloud segmentation is carried out using grating map, with
Realize the separation on road surface and barrier；
The cluster that barrier is carried out using the method for connected component labeling, obtains the corresponding radar point set of each barrier, benefit
The profile of barrier is obtained with radar point set, and linear fit then is carried out to profile, the shape of barrier is obtained, in conjunction with barrier
Size realize barrier identification；
Realize that obstacle information exchanges between other net connection vehicles using the V2X module on net connection vehicle.
2. joining vehicle localization method based on the net of 3D laser radar and V2X as described in claim 1, which is characterized in that described
The radar data is pre-processed, radar point cloud data is obtained, comprising:
In the radar data, filtering out with laser radar distance is zero or beyond the radar within the scope of laser radar range
Point, and median filtering is carried out to the radar points in every scan line of laser radar, to filter out abnormal reflection point, to obtain thunder
Up to point cloud data.
3. joining vehicle localization method based on the net of 3D laser radar and V2X as described in claim 1, which is characterized in that described
Pretreated radar point cloud data is transformed into net connection vehicle coordinate system, comprising:
The radar point cloud data is made of radar points, and the coordinate representation of each radar points is (ρ, α, θ), and wherein ρ is indicated
Radar points are to the distance of laser radar, and α is the azimuth of radar points, and θ indicates the elevation angle of radar points；
Error correction is carried out according to the internal reference file of laser radar, by the rectangular co-ordinate of the coordinate transformation of radar points to laser radar
In system, the coordinate of radar points is transformed into net by the positional relationship of rectangular coordinate system and net connection vehicle coordinate system further according to laser radar
Join in vehicle coordinate system.
4. joining vehicle localization method based on the net of 3D laser radar and V2X as described in claim 1, which is characterized in that described
According to the variable density feature of radar points in radar point cloud data, carry out point cloud segmentation using grating map, with realize road surface with
The separation of barrier, comprising:
Grating map is established, the radar points in net connection vehicle coordinate system are projected in grating map；
The radar points projected in grating map are filtered, if radar points in some grid number less than 2 if to the grid
In radar points filtered out；If the number of radar points is greater than 2 in some grid, calculating projects to all radars in the grid
The grid tag is barrier grid if H is greater than 20cm by the difference H of the maxima and minima of the standoff height of point；If
Within the scope of 7*7 grid centered on barrier grid, without other barrier grids, then the barrier grid is isolated danger
Grid filters out the radar points in the isolated danger grid；
The gradient value for calculating radar points, if the gradient value of radar points is more than preset threshold value, then by radar points labeled as initial barrier
Hinder object-point, to realize the separation of road surface and barrier.
5. joining vehicle localization method based on the net of 3D laser radar and V2X as claimed in claim 4, which is characterized in that described
Calculate the gradient value of radar points, comprising:
The difference in height of the laser radar two radar points adjacent in adjacent two scan lines of same scanning direction is remembered
For Δ z, described two radar points are distance in the distance of horizontal plane, then are located at the calculating of the gradient value of the radar points in outside
Formula are as follows:
6. joining vehicle localization method based on the net of 3D laser radar and V2X as described in claim 1, which is characterized in that described
The cluster of barrier is carried out using the method for connected component labeling, comprising:
The traversal grating map, finds first un-marked barrier grid, is marked for the first time, in the barrier
Other barrier grids are found in 8 neighborhoods of grid similarly to be marked, and the position of all obstacle grids are recorded, by these positions
The radar points set merge, and obtain the corresponding radar point set of connected domain；
The corresponding radar point set of other connected domains is obtained in the same manner.
7. joining vehicle localization method based on the net of 3D laser radar and V2X as described in claim 1, which is characterized in that described
The profile of barrier is obtained using radar point set, comprising:
A bianry image is established, the length of bianry image and wide pixel number correspond respectively to the length of grating map and width is included
Grid number；All radar point sets are mapped in bianry image, sector scanning is carried out to bianry image and retrieves profile, to obtain
Obtain the profile of each barrier.
8. joining vehicle localization method based on the net of 3D laser radar and V2X as described in claim 1, which is characterized in that described
Linear fit is carried out to profile, obtains the shape of barrier, comprising:
Remember that the collection of the corresponding radar points of the profile of a barrier is combined into P=(p1,p2... pn), by the starting point p of the set1And end
Point pnIt is connected with straight line, the distance d of the every bit in set of computations to the straight linei, by comparing acquisition maximum distance dj, sentence
The distance of breaking and former given threshold DthSize, if more than the threshold value, then with point pjFor breakpoint, set P is divided to for two sons
Collect P1=(p1,p2... pj) and P2=(pj,pj+1... pn)；Above step is repeated to subset, until the point in set P is to directly
The distance value of line is respectively less than threshold value, completes segmentation；
For the set after segmentation, the minimum area-encasing rectangle frame of the corresponding barrier of the set is acquired using Minimum Convex Closure method,
To obtain central point, length, the width of barrier.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810667814.9A CN108983248A (en) | 2018-06-26 | 2018-06-26 | It is a kind of that vehicle localization method is joined based on the net of 3D laser radar and V2X |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810667814.9A CN108983248A (en) | 2018-06-26 | 2018-06-26 | It is a kind of that vehicle localization method is joined based on the net of 3D laser radar and V2X |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108983248A true CN108983248A (en) | 2018-12-11 |
Family
ID=64538329
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810667814.9A Pending CN108983248A (en) | 2018-06-26 | 2018-06-26 | It is a kind of that vehicle localization method is joined based on the net of 3D laser radar and V2X |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108983248A (en) |
Cited By (40)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110068819A (en) * | 2019-03-27 | 2019-07-30 | 东软睿驰汽车技术（沈阳）有限公司 | A kind of method and device for extracting obstacle position information |
| CN110208819A (en) * | 2019-05-14 | 2019-09-06 | 江苏大学 | A kind of processing method of multiple barrier three-dimensional laser radar data |
| CN110208842A (en) * | 2019-05-28 | 2019-09-06 | 长安大学 | Vehicle high-precision locating method under a kind of car networking environment |
| CN110221615A (en) * | 2019-06-18 | 2019-09-10 | 长春理工大学 | A kind of auxiliary vehicle drive method based on road conditions identification |
| CN110221616A (en) * | 2019-06-25 | 2019-09-10 | 清华大学苏州汽车研究院（吴江） | A kind of method, apparatus, equipment and medium that map generates |
| CN110286387A (en) * | 2019-06-25 | 2019-09-27 | 深兰科技（上海）有限公司 | Obstacle detection method, device and storage medium applied to automated driving system |
| CN110296713A (en) * | 2019-06-17 | 2019-10-01 | 深圳数翔科技有限公司 | Trackside automatic driving vehicle Position Fixing Navigation System and single, multiple vehicle positioning and navigation methods |
| CN110378919A (en) * | 2019-06-14 | 2019-10-25 | 江苏裕兰信息科技有限公司 | A kind of current obstacle detection method of the arrow path based on SLAM |
| CN110503040A (en) * | 2019-08-23 | 2019-11-26 | 斯坦德机器人（深圳）有限公司 | Obstacle detection method and device |
| CN110728210A (en) * | 2019-09-25 | 2020-01-24 | 上海交通大学 | Semi-supervised target labeling method and system for three-dimensional point cloud data |
| CN110726993A (en) * | 2019-09-06 | 2020-01-24 | 武汉光庭科技有限公司 | Obstacle detection method using single line laser radar and millimeter wave radar |
| CN110794392A (en) * | 2019-10-15 | 2020-02-14 | 上海创昂智能技术有限公司 | Vehicle positioning method and device, vehicle and storage medium |
| CN111144228A (en) * | 2019-12-05 | 2020-05-12 | 山东超越数控电子股份有限公司 | Obstacle identification method based on 3D point cloud data and computer equipment |
| CN111308500A (en) * | 2020-04-07 | 2020-06-19 | 三一机器人科技有限公司 | Obstacle sensing method and device based on single-line laser radar and computer terminal |
| CN111308499A (en) * | 2020-03-09 | 2020-06-19 | 中振同辂(江苏)机器人有限公司 | Obstacle detection method based on multi-line laser radar |
| CN111337941A (en) * | 2020-03-18 | 2020-06-26 | 中国科学技术大学 | Dynamic obstacle tracking method based on sparse laser radar data |
| CN111474560A (en) * | 2020-04-16 | 2020-07-31 | 苏州大学 | Obstacle positioning method, device and equipment |
| CN111508272A (en) * | 2019-01-31 | 2020-08-07 | 斯特拉德视觉公司 | Method and apparatus for providing robust camera-based object distance prediction |
| CN111537994A (en) * | 2020-03-24 | 2020-08-14 | 江苏徐工工程机械研究院有限公司 | Unmanned mine card obstacle detection method |
| CN111538009A (en) * | 2019-01-21 | 2020-08-14 | 杭州海康威视数字技术股份有限公司 | Radar point marking method and device |
| CN111590595A (en) * | 2020-06-30 | 2020-08-28 | 深圳市银星智能科技股份有限公司 | Positioning method and device, mobile robot and storage medium |
| CN111990929A (en) * | 2020-08-26 | 2020-11-27 | 北京石头世纪科技股份有限公司 | Obstacle detection method and device, self-walking robot and storage medium |
| CN112034476A (en) * | 2020-08-24 | 2020-12-04 | 北京首汽智行科技有限公司 | Point cloud data generation method based on laser radar |
| CN112149695A (en) * | 2019-06-26 | 2020-12-29 | 北京海益同展信息科技有限公司 | Boundary detection method, device, equipment and computer readable storage medium |
| CN112199991A (en) * | 2020-08-27 | 2021-01-08 | 广州中国科学院软件应用技术研究所 | Simulation point cloud filtering method and system applied to vehicle-road cooperative roadside sensing |
| WO2021012254A1 (en) * | 2019-07-25 | 2021-01-28 | 深圳市大疆创新科技有限公司 | Target detection method, system, and mobile platform |
| CN112464812A (en) * | 2020-11-27 | 2021-03-09 | 西北工业大学 | Vehicle-based sunken obstacle detection method |
| CN112505704A (en) * | 2020-11-10 | 2021-03-16 | 北京埃福瑞科技有限公司 | Method for improving safety of train autonomous intelligent sensing system and train |
| CN112543877A (en) * | 2019-04-03 | 2021-03-23 | 华为技术有限公司 | Positioning method and positioning device |
| CN113077473A (en) * | 2020-01-03 | 2021-07-06 | 广州汽车集团股份有限公司 | Three-dimensional laser point cloud pavement segmentation method, system, computer equipment and medium |
| CN113343835A (en) * | 2021-06-02 | 2021-09-03 | 合肥泰瑞数创科技有限公司 | Object identification method and system suitable for emergency rescue and storage medium |
| CN113496491A (en) * | 2020-03-19 | 2021-10-12 | 广州汽车集团股份有限公司 | Road surface segmentation method and device based on multi-line laser radar |
| CN113661755A (en) * | 2019-03-18 | 2021-11-16 | 苹果公司 | System and method for balancing time resource allocation for cellular vehicle to all (C-V2X) and intelligent transportation system (ITS-G5) or Dedicated Short Range Communication (DSRC) depending on user configuration |
| CN114255252A (en) * | 2022-02-28 | 2022-03-29 | 新石器慧通（北京）科技有限公司 | Obstacle contour acquisition method, device, equipment and computer-readable storage medium |
| CN114419605A (en) * | 2022-03-29 | 2022-04-29 | 之江实验室 | Visual enhancement method and system based on multi-network vehicle-connected space alignment feature fusion |
| CN114660568A (en) * | 2022-02-21 | 2022-06-24 | 广西柳工机械股份有限公司 | Laser radar obstacle detection method and device |
| CN114758504A (en) * | 2022-06-13 | 2022-07-15 | 之江实验室 | Online vehicle overspeed early warning method and system based on filtering correction |
| CN115236696A (en) * | 2022-09-22 | 2022-10-25 | 毫末智行科技有限公司 | Method and device for determining obstacle, electronic equipment and storage medium |
| CN116359932A (en) * | 2023-03-27 | 2023-06-30 | 珠海创智科技有限公司 | Barrier distance feedback method and device based on laser radar sensor |
| CN114660568B (en) * | 2022-02-21 | 2024-04-30 | 广西柳工机械股份有限公司 | Laser radar obstacle detection method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105774729A (en) * | 2016-03-25 | 2016-07-20 | 奇瑞汽车股份有限公司 | V2V active safety system and control method thereof |
| CN106199558A (en) * | 2016-08-18 | 2016-12-07 | 宁波傲视智绘光电科技有限公司 | Barrier method for quick |
| CN106997049A (en) * | 2017-03-14 | 2017-08-01 | 奇瑞汽车股份有限公司 | A kind of method and apparatus of the detection barrier based on laser point cloud data |
-
2018
- 2018-06-26 CN CN201810667814.9A patent/CN108983248A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105774729A (en) * | 2016-03-25 | 2016-07-20 | 奇瑞汽车股份有限公司 | V2V active safety system and control method thereof |
| CN106199558A (en) * | 2016-08-18 | 2016-12-07 | 宁波傲视智绘光电科技有限公司 | Barrier method for quick |
| CN106997049A (en) * | 2017-03-14 | 2017-08-01 | 奇瑞汽车股份有限公司 | A kind of method and apparatus of the detection barrier based on laser point cloud data |
Non-Patent Citations (4)
| Title |
|---|
| 李小毛 等: ""基于3D激光雷达的无人水面艇海上目标检测"", 《上海大学学报（自然科学版）》 * |
| 杨飞: ""基于三维激光雷达的运动目标实时检测与跟踪"", 《中国优秀硕士学位论文全文数据库（电子期刊）信息科技辑》 * |
| 王盛: ""复杂野外环境下机器人障碍物检测方法研究"", 《中国优秀硕士学位论文全文数据库（电子期刊）信息科技辑》 * |
| 苏致远 等: ""基于三维激光雷达的车辆目标检测方法"", 《军事交通学院学报》 * |
Cited By (57)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111538009B (en) * | 2019-01-21 | 2022-09-16 | 杭州海康威视数字技术股份有限公司 | Radar point marking method and device |
| CN111538009A (en) * | 2019-01-21 | 2020-08-14 | 杭州海康威视数字技术股份有限公司 | Radar point marking method and device |
| CN111508272A (en) * | 2019-01-31 | 2020-08-07 | 斯特拉德视觉公司 | Method and apparatus for providing robust camera-based object distance prediction |
| CN113661755A (en) * | 2019-03-18 | 2021-11-16 | 苹果公司 | System and method for balancing time resource allocation for cellular vehicle to all (C-V2X) and intelligent transportation system (ITS-G5) or Dedicated Short Range Communication (DSRC) depending on user configuration |
| CN110068819A (en) * | 2019-03-27 | 2019-07-30 | 东软睿驰汽车技术（沈阳）有限公司 | A kind of method and device for extracting obstacle position information |
| CN110068819B (en) * | 2019-03-27 | 2021-08-31 | 东软睿驰汽车技术（沈阳）有限公司 | Method and device for extracting position information of obstacle |
| CN112543877B (en) * | 2019-04-03 | 2022-01-11 | 华为技术有限公司 | Positioning method and positioning device |
| CN112543877A (en) * | 2019-04-03 | 2021-03-23 | 华为技术有限公司 | Positioning method and positioning device |
| CN110208819A (en) * | 2019-05-14 | 2019-09-06 | 江苏大学 | A kind of processing method of multiple barrier three-dimensional laser radar data |
| CN110208842A (en) * | 2019-05-28 | 2019-09-06 | 长安大学 | Vehicle high-precision locating method under a kind of car networking environment |
| CN110378919A (en) * | 2019-06-14 | 2019-10-25 | 江苏裕兰信息科技有限公司 | A kind of current obstacle detection method of the arrow path based on SLAM |
| CN110378919B (en) * | 2019-06-14 | 2023-05-23 | 珠海大轩信息科技有限公司 | Narrow-road passing obstacle detection method based on SLAM |
| CN110296713A (en) * | 2019-06-17 | 2019-10-01 | 深圳数翔科技有限公司 | Trackside automatic driving vehicle Position Fixing Navigation System and single, multiple vehicle positioning and navigation methods |
| CN110221615A (en) * | 2019-06-18 | 2019-09-10 | 长春理工大学 | A kind of auxiliary vehicle drive method based on road conditions identification |
| CN110286387B (en) * | 2019-06-25 | 2021-09-24 | 深兰科技（上海）有限公司 | Obstacle detection method and device applied to automatic driving system and storage medium |
| CN110221616A (en) * | 2019-06-25 | 2019-09-10 | 清华大学苏州汽车研究院（吴江） | A kind of method, apparatus, equipment and medium that map generates |
| CN110286387A (en) * | 2019-06-25 | 2019-09-27 | 深兰科技（上海）有限公司 | Obstacle detection method, device and storage medium applied to automated driving system |
| CN112149695A (en) * | 2019-06-26 | 2020-12-29 | 北京海益同展信息科技有限公司 | Boundary detection method, device, equipment and computer readable storage medium |
| WO2021012254A1 (en) * | 2019-07-25 | 2021-01-28 | 深圳市大疆创新科技有限公司 | Target detection method, system, and mobile platform |
| CN110503040A (en) * | 2019-08-23 | 2019-11-26 | 斯坦德机器人（深圳）有限公司 | Obstacle detection method and device |
| CN110726993A (en) * | 2019-09-06 | 2020-01-24 | 武汉光庭科技有限公司 | Obstacle detection method using single line laser radar and millimeter wave radar |
| CN110728210A (en) * | 2019-09-25 | 2020-01-24 | 上海交通大学 | Semi-supervised target labeling method and system for three-dimensional point cloud data |
| CN110794392A (en) * | 2019-10-15 | 2020-02-14 | 上海创昂智能技术有限公司 | Vehicle positioning method and device, vehicle and storage medium |
| CN110794392B (en) * | 2019-10-15 | 2024-03-19 | 上海创昂智能技术有限公司 | Vehicle positioning method and device, vehicle and storage medium |
| CN111144228B (en) * | 2019-12-05 | 2023-09-12 | 超越科技股份有限公司 | Obstacle identification method based on 3D point cloud data and computer equipment |
| CN111144228A (en) * | 2019-12-05 | 2020-05-12 | 山东超越数控电子股份有限公司 | Obstacle identification method based on 3D point cloud data and computer equipment |
| CN113077473A (en) * | 2020-01-03 | 2021-07-06 | 广州汽车集团股份有限公司 | Three-dimensional laser point cloud pavement segmentation method, system, computer equipment and medium |
| CN111308499A (en) * | 2020-03-09 | 2020-06-19 | 中振同辂(江苏)机器人有限公司 | Obstacle detection method based on multi-line laser radar |
| CN111337941A (en) * | 2020-03-18 | 2020-06-26 | 中国科学技术大学 | Dynamic obstacle tracking method based on sparse laser radar data |
| CN111337941B (en) * | 2020-03-18 | 2022-03-04 | 中国科学技术大学 | Dynamic obstacle tracking method based on sparse laser radar data |
| CN113496491B (en) * | 2020-03-19 | 2023-12-15 | 广州汽车集团股份有限公司 | Road surface segmentation method and device based on multi-line laser radar |
| CN113496491A (en) * | 2020-03-19 | 2021-10-12 | 广州汽车集团股份有限公司 | Road surface segmentation method and device based on multi-line laser radar |
| CN111537994A (en) * | 2020-03-24 | 2020-08-14 | 江苏徐工工程机械研究院有限公司 | Unmanned mine card obstacle detection method |
| CN111308500A (en) * | 2020-04-07 | 2020-06-19 | 三一机器人科技有限公司 | Obstacle sensing method and device based on single-line laser radar and computer terminal |
| CN111308500B (en) * | 2020-04-07 | 2022-02-11 | 三一机器人科技有限公司 | Obstacle sensing method and device based on single-line laser radar and computer terminal |
| CN111474560B (en) * | 2020-04-16 | 2023-11-24 | 苏州大学 | Obstacle positioning method, device and equipment |
| CN111474560A (en) * | 2020-04-16 | 2020-07-31 | 苏州大学 | Obstacle positioning method, device and equipment |
| CN111590595A (en) * | 2020-06-30 | 2020-08-28 | 深圳市银星智能科技股份有限公司 | Positioning method and device, mobile robot and storage medium |
| CN111590595B (en) * | 2020-06-30 | 2021-09-28 | 深圳市银星智能科技股份有限公司 | Positioning method and device, mobile robot and storage medium |
| CN112034476A (en) * | 2020-08-24 | 2020-12-04 | 北京首汽智行科技有限公司 | Point cloud data generation method based on laser radar |
| CN111990929A (en) * | 2020-08-26 | 2020-11-27 | 北京石头世纪科技股份有限公司 | Obstacle detection method and device, self-walking robot and storage medium |
| WO2022041740A1 (en) * | 2020-08-26 | 2022-03-03 | 北京石头世纪科技股份有限公司 | Method and apparatus for detecting obstacle, self-propelled robot, and storage medium |
| CN112199991A (en) * | 2020-08-27 | 2021-01-08 | 广州中国科学院软件应用技术研究所 | Simulation point cloud filtering method and system applied to vehicle-road cooperative roadside sensing |
| CN112199991B (en) * | 2020-08-27 | 2024-04-30 | 广州中国科学院软件应用技术研究所 | Simulation point cloud filtering method and system applied to vehicle-road cooperation road side perception |
| CN112505704A (en) * | 2020-11-10 | 2021-03-16 | 北京埃福瑞科技有限公司 | Method for improving safety of train autonomous intelligent sensing system and train |
| CN112464812A (en) * | 2020-11-27 | 2021-03-09 | 西北工业大学 | Vehicle-based sunken obstacle detection method |
| CN112464812B (en) * | 2020-11-27 | 2023-11-24 | 西北工业大学 | Vehicle-based concave obstacle detection method |
| CN113343835B (en) * | 2021-06-02 | 2022-04-15 | 合肥泰瑞数创科技有限公司 | Object identification method and system suitable for emergency rescue and storage medium |
| CN113343835A (en) * | 2021-06-02 | 2021-09-03 | 合肥泰瑞数创科技有限公司 | Object identification method and system suitable for emergency rescue and storage medium |
| CN114660568A (en) * | 2022-02-21 | 2022-06-24 | 广西柳工机械股份有限公司 | Laser radar obstacle detection method and device |
| CN114660568B (en) * | 2022-02-21 | 2024-04-30 | 广西柳工机械股份有限公司 | Laser radar obstacle detection method and device |
| CN114255252A (en) * | 2022-02-28 | 2022-03-29 | 新石器慧通（北京）科技有限公司 | Obstacle contour acquisition method, device, equipment and computer-readable storage medium |
| CN114419605A (en) * | 2022-03-29 | 2022-04-29 | 之江实验室 | Visual enhancement method and system based on multi-network vehicle-connected space alignment feature fusion |
| CN114758504A (en) * | 2022-06-13 | 2022-07-15 | 之江实验室 | Online vehicle overspeed early warning method and system based on filtering correction |
| CN115236696B (en) * | 2022-09-22 | 2022-12-09 | 毫末智行科技有限公司 | Method and device for determining obstacle, electronic equipment and storage medium |
| CN115236696A (en) * | 2022-09-22 | 2022-10-25 | 毫末智行科技有限公司 | Method and device for determining obstacle, electronic equipment and storage medium |
| CN116359932A (en) * | 2023-03-27 | 2023-06-30 | 珠海创智科技有限公司 | Barrier distance feedback method and device based on laser radar sensor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108983248A (en) | It is a kind of that vehicle localization method is joined based on the net of 3D laser radar and V2X | |
| JP7398506B2 (en) | Methods and systems for generating and using localization reference data | |
| US10254395B2 (en) | System and methods for scanning with integrated radar detection and image capture | |
| US9664784B2 (en) | System and methods for data point detection and spatial modeling | |
| US9898821B2 (en) | Determination of object data by template-based UAV control | |
| US20240053477A1 (en) | System and method for measuring image distance of power transmission lines with unmanned aerial vehicle (uav) | |
| Nouri et al. | Cloud height and tracking accuracy of three all sky imager systems for individual clouds | |
| WO2022206978A1 (en) | Roadside millimeter-wave radar calibration method based on vehicle-mounted positioning apparatus | |
| CN103065323B (en) | Subsection space aligning method based on homography transformational matrix | |
| CN104200086A (en) | Wide-baseline visible light camera pose estimation method | |
| CN109031346A (en) | A kind of periphery parking position aided detection method based on 3D laser radar | |
| CN110837080A (en) | Rapid calibration method of laser radar mobile measurement system | |
| Wen et al. | Toward efficient 3-d colored mapping in gps-/gnss-denied environments | |
| CN104932001A (en) | Real-time 3D nuclear radiation environment reconstruction monitoring system | |
| CN105203023A (en) | One-stop calibration method for arrangement parameters of vehicle-mounted three-dimensional laser scanning system | |
| CN106370160A (en) | Robot indoor positioning system and method | |
| US20220383755A1 (en) | Unmanned aerial vehicle positioning method based on millimeter-wave radar | |
| CN208027170U (en) | A kind of power-line patrolling unmanned plane and system | |
| El Natour et al. | Radar and vision sensors calibration for outdoor 3D reconstruction | |
| CN115027482A (en) | Fusion positioning method in intelligent driving | |
| CN113419235A (en) | Unmanned aerial vehicle positioning method based on millimeter wave radar | |
| CN206096937U (en) | Unmanned distribution vehicle | |
| CN115166721B (en) | Radar and GNSS information calibration fusion method and device in roadside sensing equipment | |
| CN112254722A (en) | Vehicle positioning method based on QR code and inertial navigation fusion | |
| Iannucci et al. | Cross-Modal Localization: Using automotive radar for absolute geolocation within a map produced with visible-light imagery |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181211 |