CN108151681A - A kind of vehicle-mounted road surface unevenness identifying system and method based on binocular camera - Google Patents
A kind of vehicle-mounted road surface unevenness identifying system and method based on binocular camera Download PDFInfo
- Publication number
- CN108151681A CN108151681A CN201711178413.9A CN201711178413A CN108151681A CN 108151681 A CN108151681 A CN 108151681A CN 201711178413 A CN201711178413 A CN 201711178413A CN 108151681 A CN108151681 A CN 108151681A
- Authority
- CN
- China
- Prior art keywords
- road roughness
- grade
- road
- image
- binocular
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/30—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces
- G01B11/303—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces using photoelectric detection means
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of vehicle-mounted road surface unevenness identifying system based on binocular camera, including：Image collection module obtains the binocular image on vehicle front road surface；The binocular image of reception is carried out three-dimensional reconstruction and transitable cloud is extracted by image processing module；Speed data collection module acquires speed signal；Steering wheel angle acquisition module acquires steering wheel angle signal；Unevenness computing module, based on reception can passage points cloud, speed signal and steering wheel angle signal calculate road roughness, and road roughness grade is determined based on the road roughness calculated and the instruction of corresponding road roughness grade is sent to reminding module；Reminding module instructs to export corresponding alerting signal according to the road roughness grade received.The present invention can obtain the unevenness information on road surface in advance, be reminded to driver, it is ensured that drive safety, and the signal can be supplied to other electronic control systems, improve vehicle road adaptability.
Description
Technical field
The present invention relates to a kind of vehicle-mounted road surface unevenness identifying system and method, and in particular to a kind of based on binocular camera
Vehicle-mounted road surface unevenness identifying system and method.
Background technology
Vehicle can be risen and fallen in the process of moving by road bump to be influenced, and causes the vertical vibration of wheel or vehicle body, directly
Connect the ride comfort for influencing vehicle and the comfort of driver；And since analysis of wheel vertical load and contact area are unstable, also can
The traction of each wheel is influenced, and then is had a significant impact to the braking and control stability of vehicle；Electronic air suspension
(ECAS), the systems such as anti-blocking brake system (ABS), electronic stability system (ESC) can also be directed to the variation of road roughness
Control model or parameter are adjusted, improves its adaptability in uneven road surface.
Existing system generally passes through the suspension height signal or electronic stability system of electronic air suspension system (ECAS)
(ESP) time domain/frequency domain character of wheel speed signal is identified.But these recognition methods can only be in vehicle on uneven road surface
Road surface variation can just be recognized after a distance by running over, and have certain hysteresis quality, and the uneven road surface in identification process
Detrimental effect is produced to driver and vehicle.
Therefore, the unevenness information on road surface is obtained in advance, is reminded to driver, and the signal is supplied to other electricity
Sub-control system is the key that promote vehicle driving experience and vehicle electric-control system adaptability.
Invention content
For above-mentioned technical problem, the present invention provides a kind of unevenness information that can obtain road surface in advance, to driver
The vehicle-mounted road surface unevenness identifying system and method based on binocular camera reminded.
The technical solution adopted by the present invention is：
One embodiment of the invention provides a kind of vehicle-mounted road surface unevenness identifying system based on binocular camera, is obtained including image
Modulus block, image processing module, speed data collection module, steering wheel angle acquisition module, unevenness computing module and reminding module；
Described image acquisition module is binocular camera, for the binocular image for obtaining vehicle front road surface and is sent out the binocular image of acquisition
Give described image processing module；Described image processing module is used for the binocular image received progress three-dimensional reconstruction and right
Transitable cloud extract and by extraction can passage points cloud be sent to the unevenness computing module；The speed
Acquisition module is used to acquire speed signal and is sent to the unevenness computing module；The steering wheel angle acquisition module is used for
Acquisition steering wheel angle signal is simultaneously sent to the unevenness computing module；The unevenness computing module is used for based on reception
Can passage points cloud, speed signal and steering wheel angle signal calculate road roughness, and true based on the road roughness calculated
Determine road roughness grade and the instruction of corresponding road roughness grade is sent to the reminding module；The prompting mould
Block is used to be instructed to export corresponding alerting signal according to the road roughness grade received.
Optionally, camera calibration module is further included, for being demarcated to obtain camera calibration ginseng to the binocular camera
It counts and is sent to described image processing module；The camera calibration parameter include parallax range between two cameras, lens focus,
Spin matrix, translation vector, distortion coefficients of camera lens and optical centre bias position.
Optionally, described image processing module is used to the binocular image received carrying out three-dimensional reconstruction and pair can pass through
Point cloud extract and specifically include：Using default bearing calibration and acquired camera calibration parameter to the binocular figure that is received
As being corrected, the binocular image after being corrected；Three-dimensional is carried out to the binocular image after correction using preset matching method
Match, obtain the correspondence between the image slices vegetarian refreshments of binocular image；Between image slices vegetarian refreshments based on the binocular image
Correspondence and camera calibration parameter calculate three-dimensional coordinate of the binocular image after Stereo matching in corresponding coordinate system；
Wherein, the corresponding coordinate system of binocular image is to correspond to the optical center of camera as origin using the image, and the vertical imaging plane of Z axis is outside, X
Axis along imaging plane level to the right, coordinate system of the Y-axis along imaging plane straight down；The three-dimensional of the binocular image calculated is sat
Mark carries out coordinate conversion so that the three-dimensional coordinate point cloud under camera coordinates system is converted to the point cloud under bodywork reference frame；Wherein, vehicle
Body coordinate system is using vehicle rear axle center as origin, and before X axis, Y-axis is left, the coordinate system of Z axis vertically upward；By car body coordinate
The grid of predetermined dimension is divided into the X-axis of system in X-Y plane less than the three-dimensional point cloud of pre-determined distance threshold value, by the grid
The difference of interior Z-direction height maxima and minimum value is extracted no more than the point cloud in the grid of predetermined threshold value, and obtaining can passage points
Cloud.
Optionally, the unevenness computing module for based on reception can passage points cloud, speed signal and steering wheel turn
Angle signal calculates road roughness, and determines road roughness grade based on the road roughness calculated, specifically includes：It utilizes
Default calculation formula, based on reception can passage points cloud, speed signal and steering wheel angle signal calculate the running region of vehicle,
Obtain running region point cloud；Based on the running region point cloud computing road roughness variance；Based on the Uneven road calculated
Degree variance and default decision procedure determine road roughness grade.
Optionally, included based on the running region point cloud computing road roughness variance：Based on the running region point
Cloud and preset algorithm determine the datum plane of fitting；Each sampled point in the running region point cloud is calculated to the benchmark
The distance of plane；It will sum after the distance progress square of each sampled point to the datum plane, obtain square distance
And value；By the total number of sample points in obtained square distance and value divided by running region point cloud, the road surface is obtained not
Pingdu variance.
Optionally, the road roughness grade includes the first estate, the second grade, the tertiary gradient and the fourth estate；Its
In, determine that road roughness grade includes based on the road roughness variance and default decision procedure calculated：Work as current road
When unevenness grade is the first estate, if the road roughness variance calculated is more than the first predetermined angle thresholding lower limiting value
When, then current road roughness grade is determined as the second grade；When current road unevenness grade is the second grade, such as
When the road roughness variance that fruit is calculated is more than the second predetermined angle thresholding lower limiting value, then by current road roughness grade
It is determined as the tertiary gradient；It, ought if the road roughness variance calculated is less than the first predetermined angle thresholding upper limit value
Preceding road roughness grade is determined as the first estate；When current road unevenness grade is the tertiary gradient, if calculated
Road roughness variance be more than third predetermined angle threshold value when, then current road roughness grade is determined as 4th etc.
Grade；If the road roughness variance calculated is less than the second predetermined angle thresholding upper limit value, by current Uneven road
Degree grade is determined as the second grade；When current road unevenness grade is the fourth estate, if the road roughness calculated
When variance is more than the 4th predetermined angle threshold value, then current road roughness grade is determined as the tertiary gradient.
Another embodiment of the present invention provides a kind of vehicle-mounted road surface unevenness recognition methods, including：Obtain vehicle front road surface
Binocular image；Receive the binocular image and the binocular image to being received carry out three-dimensional reconstruction and to transitable cloud into
Row extraction；Acquire speed signal and steering wheel angle signal；It can passage points cloud, speed signal and steering wheel angle letter described in reception
Number and based on received can passage points cloud, speed signal and steering wheel angle signal calculate road roughness, and based on being counted
The road roughness of calculation determines road roughness grade and sends corresponding road roughness grade instruction；Receive the road
Face unevenness grade instruction simultaneously instructs to export corresponding alerting signal based on the road roughness grade received.
Optionally, it further includes：The binocular camera is demarcated to obtain camera calibration parameter；The camera calibration ginseng
Number includes parallax range, lens focus, spin matrix, translation vector, distortion coefficients of camera lens and the optical centre bias between two cameras
Position.
Optionally, it is described to receive the binocular image and the binocular image to being received carries out three-dimensional reconstruction and pair pass through
Point cloud extract and specifically include：Using default bearing calibration and acquired camera calibration parameter to the binocular figure that is received
As being corrected, the binocular image after being corrected；Three-dimensional is carried out to the binocular image after correction using preset matching method
Match, obtain the correspondence between the image slices vegetarian refreshments of binocular image；Between image slices vegetarian refreshments based on the binocular image
Correspondence and camera calibration parameter calculate three-dimensional coordinate of the binocular image after Stereo matching in corresponding coordinate system；
Wherein, the corresponding coordinate system of binocular image is to correspond to the optical center of camera as origin using the image, and the vertical imaging plane of Z axis is outside, X
Axis along imaging plane level to the right, coordinate system of the Y-axis along imaging plane straight down；The three-dimensional of the binocular image calculated is sat
Mark carries out coordinate conversion so that the three-dimensional coordinate point cloud under camera coordinates system is converted to the point cloud under bodywork reference frame；Wherein, vehicle
Body coordinate system is using vehicle rear axle center as origin, and before X axis, Y-axis is left, the coordinate system of Z axis vertically upward；By car body coordinate
The grid of predetermined dimension is divided into the X-axis of system in X-Y plane less than the three-dimensional point cloud of pre-determined distance threshold value, by the grid
The difference of interior Z-direction height maxima and minimum value is extracted no more than the point cloud in the grid of predetermined threshold value, and obtaining can passage points
Cloud.
It optionally, can passage points cloud, speed signal and steering wheel angle signal and based on being received described in the reception
Can passage points cloud, speed signal and steering wheel angle signal calculate road roughness, and true based on the road roughness calculated
Determine road roughness grade, specifically include：Using default calculation formula, based on reception can passage points cloud, speed signal and side
The running region of vehicle is calculated to disk angular signal, obtains running region point cloud；Based on the running region point cloud computing road surface
Unevenness variance；Road roughness grade is determined based on the road roughness variance and default decision procedure calculated.
Vehicle-mounted road surface unevenness identifying system and method provided by the invention directly acquire road surface three-dimensional based on binocular camera
Point cloud data, and do not depend on Vibration of Vehicle Suspensions signal and carry out indirect analysis, not by noise jamming such as vehicle itself vibrations；This
Outside, road roughness in the running region of system and method concern vehicle front, compared to based on Vibration of Vehicle Suspensions signal point
The method for analysing road roughness, can obtain road bump information in advance, be suspension when characteristics of components adjust and provide more sufficient
Between, further improve vehicle ride comfort.
Description of the drawings
Fig. 1 is the structure diagram of vehicle-mounted road surface unevenness identifying system provided in an embodiment of the present invention；
Fig. 2 is camera coordinates system and bodywork reference frame relation schematic diagram；
Fig. 3 is vehicle front running region schematic diagram；
Fig. 4 is the flow diagram of vehicle-mounted road surface unevenness recognition methods provided in an embodiment of the present invention.
Specific embodiment
To make the technical problem to be solved in the present invention, technical solution and advantage clearer, below in conjunction with attached drawing and tool
Body embodiment is described in detail.
Fig. 1 is the structure diagram of vehicle-mounted road surface unevenness identifying system provided in an embodiment of the present invention.As shown in Figure 1,
The embodiment of the present invention provides a kind of vehicle-mounted road surface unevenness identifying system based on binocular camera, including image collection module 1, figure
As processing module 2, speed data collection module 3, steering wheel angle acquisition module 4, unevenness computing module 5 and reminding module 6.It is described
Image collection module 1 is binocular camera, for the binocular image for obtaining vehicle front road surface and is sent the binocular image of acquisition
To described image processing module 2；Described image processing module 2 is used for the binocular image received progress three-dimensional reconstruction and right
Transitable cloud extract and by extraction can passage points cloud be sent to the unevenness computing module 5；The speed
Acquisition module 3 is used to acquire speed signal and is sent to the unevenness computing module 5；The steering wheel angle acquisition module 4
For acquiring steering wheel angle signal and being sent to the unevenness computing module 5；The unevenness computing module 5 is used to be based on
Receive can passage points cloud, speed signal and steering wheel angle signal calculate road roughness, and based on the road surface calculated not
Pingdu determines road roughness grade and the instruction of corresponding road roughness grade is sent to the reminding module 6；It is described
Reminding module 6 is used to be instructed to export corresponding alerting signal according to the road roughness grade received.
Specifically, the binocular camera used in the present invention includes the camera of the two same model less parallel installations in left and right,
With the parallel fixation of the baseline distance of 20cm, which is mounted on front windshield of vehicle intermediate region, acquires road surface ahead
Binocular image.Image processing module 2 can be FPGA or the integrated circuit with image processing function, can be integrated in image acquisition
In module 1, vehicle speed sensor can be used in speed data collection module 3, and steering wheel angle sensing can be used in steering wheel angle acquisition module 4
Device, what unevenness determining module 5 can be encapsulated in vehicle control device or be separately provided has control and the control of processing function
Device processed or the integrated software realization with corresponding function in the control devices such as the original suspension controller of vehicle, are reminded
Module 6 is mounted on Vehicular instrument panel, warning light or sound output device including exporting alerting signal, according to the road surface
Unevenness grade instructs to export the alerting signal instructed corresponding to the road roughness grade.
In addition, the vehicle-mounted road surface unevenness identifying system based on binocular camera of the present invention further includes camera calibration module
(not shown), for being demarcated to obtain camera calibration parameter and be sent to described image processing mould to the binocular camera
Block；The camera calibration parameter includes parallax range, lens focus and the optical centre bias position between two cameras.Specifically, phase
Machine demarcating module is used to carry out off-line calibration to binocular camera.After camera is installed for the first time, need to obtain camera by camera calibration
Parameter calculates for the three-dimensional reconstruction of picture point.It can be by the gridiron pattern binocular camera calibration side that is provided in OpenCV function libraries
The parallax range that method obtains between two cameras is b, lens focus f, left camera photocentre deviation post cxAnd cy, right camera relative to
The spin matrix R of left cameral, translation vector TlAnd two camera distortion coefficients of camera lens matrix Dl, DrCamera parameters are waited, for the first time
After the completion of installation calibration, ensure without relative position to change then without repeating the calibration process between two cameras.
Further, described image processing module is used to the binocular image received carrying out three-dimensional reconstruction and pair can lead to
Capable point cloud, which extracts, specifically includes following steps：
S110, school is carried out to the binocular image received using default bearing calibration and acquired camera calibration parameter
Just, the binocular image after being corrected.
In this step, image processing module 2 obtains the left and right camera original graph of synchronous acquisition from image collection module 1
Then picture is corrected, the binocular image after being corrected using BOUGUET binoculars polar curve.The correction of BOUGUET binoculars polar curve belongs to this
The common bearing calibration in field, the present invention are omitted to this detailed description.
S111, Stereo matching is carried out to the binocular image after correction using preset matching method, obtains the figure of binocular image
As the correspondence between pixel.
In the step, half global Stereo Matching Algorithm can be used, Stereo matching is carried out to the binocular image after correction, so as to
The correspondence between binocular image is obtained, such as pixel P (x in left image after being corrected1,y1) right image after calibration
Middle corresponding points position P ' (x2,y1).Half global Stereo Matching Algorithm belongs to the common image matching algorithm in this field, and the present invention saves
Slightly to this detailed description.
Correspondence and camera calibration parameter between S112, the image slices vegetarian refreshments based on the binocular image calculate
Three-dimensional coordinate of the binocular image in corresponding coordinate system after Stereo matching；Wherein, the corresponding coordinate system of binocular image be with
The image corresponds to the optical center of camera as origin, and the vertical imaging plane of Z axis is outside, X-axis along imaging plane level to the right, Y-axis along into
The coordinate system of image plane straight down.
With pixel P (x in left image after correction1,y1) for, based on the correspondence between obtained binocular image with
And camera calibration parameter, for pixel P using left camera photocentre as origin, the vertical imaging plane of Z axis is outside, and X-axis is flat along imaging
To the right, Y-axis is along the three-dimensional coordinate (X in imaging plane left camera coordinates system (as shown in Figure 2) straight down for face levelc,Yc,
Zc) can be obtained by following formula (1)：
Similarly, three-dimensional can be carried out to all pixels point of the left camera coordinates system after correction and right camera coordinates system
Match.
S113, the three-dimensional coordinate of the binocular image calculated is subjected to coordinate conversion so that the three-dimensional under camera coordinates system
Coordinate points cloud is converted to the point cloud under bodywork reference frame；Wherein, bodywork reference frame is the X axis using vehicle rear axle center as origin
Before, Y-axis is left, the coordinate system (as shown in Figure 2) of Z axis vertically upward.
In this step, the three-dimensional coordinate obtained in step S103 is subjected to coordinate transform, three-dimensional coordinate is sat from camera
Mark system is transformed into using vehicle rear axle center as origin, and before X axis, Y-axis is left, the point under the bodywork reference frame of Z axis vertically upward
Cloud, such as shown in following formula (2)：
Wherein, R and T is respectively and can lead to camera installation site and the relevant spin matrix of angle and translation matrix, value
Static state acquires the camera coordinates [X that multigroup characteristic point not in the same plane is acquired in image after crossing camera installationc_s Yc_s
Zc_s] (wherein, Xc_s, Yc_s, Zc_sThe column vector that X-coordinate, Y coordinate and the Z coordinate of respectively multigroup collection point are formed) and car body
Coordinate data [Xv_s Yv_s Zv_s], parametric solution is carried out using least square method, coordinate system transformation can obtain car body seat later
Point cloud Ω under mark systemo.The method for solving of least square is as follows：
VecXv=inv ([Xv_s Yv_s Zv_s Iv_s]T*[Xv_s Yv_s Zv_s Iv_s])*[Xv_s Yv_s Zv_s Iv_s]T*
Xc_s
VecYv=inv ([Xv_s Yv_s Zv_s Iv_s]T*[Xv_s Yv_s Zv_s Iv_s])*[Xv_s Yv_s Zv_s Iv_s]T*
Yc_s
VecZv=inv ([Xv_s Yv_s Zv_s Iv_s]T*[Xv_s Yv_s Zv_s Iv_s])*[Xv_s Yv_s Zv_s Iv_s]T*
Zc_s
S114, by the X-axis of bodywork reference frame be less than pre-determined distance threshold value three-dimensional point cloud be divided into X-Y plane it is pre-
The grid of set pattern lattice, by the difference of Z-direction height maxima and minimum value in the grid no more than the point in the grid of predetermined threshold value
Cloud extracts, and obtaining can passage points cloud.
Due to including more periphery complex background from the original point cloud that binocular camera obtains, wherein there are non-ground points
Obstacle, need to filter out barrier point by image processing module 2, extraction effectively can passage points cloud.At one of the present invention
In example, by the X in bodywork reference framevLess than pre-determined distance threshold value ThrThree-dimensional point cloud in X-Y plane, be divided into 0.3 ×
The grid of 3.3m units is more than threshold value Th for the difference of Z-direction height maxima and minimum value in grid (the present embodiment takes 0.25m)
Grid, it is believed that the grid there are impassabitity obstacle, from bodywork reference frame original point cloud ΩoIn middle rejecting grid
Point finally obtains transitable cloud Ωv, i.e., by the difference of Z-direction height maxima and minimum value in the grid no more than default
Point cloud in the grid of threshold value extracts, by extraction can passage points cloud be sent to unevenness computing module 5.
Further, the unevenness computing module 5 for based on reception can passage points cloud, speed signal and steering wheel
Angular signal calculates road roughness, and determines road roughness grade based on the road roughness calculated, specifically include with
Lower step：
S120, using default calculation formula, based on reception can passage points cloud, speed signal and steering wheel angle signal meter
The running region of vehicle is calculated, obtains running region point cloud.
The acquisition of unevenness computing module 5 can passage points cloud ΩvAfterwards, vehicle is further estimated according to car status information
Running region, so as to calculate the road surface ahead point cloud for more accurately meeting driver intention.Specifically, unevenness computing module 5
Current vehicle speed u and steering wheel angle δ is obtained from speed data collection module 3 and steering wheel angle acquisition module 4swAfterwards, according to the following formula (3)
Calculate vehicle front running region：
Wherein, ThrFor the distance threshold of region-of-interest, 20m is taken in the present embodiment；K is the stabilization of vehicle understeer characteristics
Property factor takes 0.0006s in the present embodiment2/m2；L is vehicle wheelbase, and 2.97m is taken in the present embodiment；isteerFor turning rate,
13.75 are taken in the present embodiment；W is vehicle width, and 1.875m is taken in the present embodiment.For can passage points cloud ΩvMiddle X, Y value meet
The point of above formula (3) condition then puts it into final traveling region point cloud ΩrIn, it is calculated for road roughness, running region
Schematic diagram in Fig. 3 as shown in figure 3, define when vehicle turns left as positive direction.
S121, based on the running region point cloud computing road roughness variance.
The step specifically includes：
(1) based on the running region point cloud and preset algorithm, the datum plane of fitting is determined；
(2) each sampled point in the running region point cloud is calculated to the distance of the datum plane；
(3) it will sum after the progress square of the distance of each sampled point to the datum plane, obtain square distance
And value；
(4) it by the total number of sample points in obtained square distance and value divided by running region point cloud, obtains described
Road roughness variance.
Specifically, unevenness computing module 5 obtains running region point cloud Ω from image processing module 2r={ (xi,yi,zi)i
∈ [1, n] } after, the ground level being fitted, i.e. datum plane are obtained using least square fitting first, as shown in following formula (4)：
Wherein pinv () is represented through singular value decomposition (Singular Value Decomposition：SVD) method is asked
The pseudo inverse matrix of matrix, A, B, D are respectively plane fitting coefficient, and can then obtain space plane equation by above formula (4) calculating is
The datum plane of Ax+By-z+D=0.
Then according to formulaCalculate each sampled point (x in running region point cloudi,yi,zi) to institute
State the distance of datum plane；
Then it will sum after the progress square of the distance of each sampled point of calculating to datum plane, it is flat to obtain distance
Side and value, and by the total number of sample points in obtained square distance and value divided by running region point cloud, obtain the road
Face unevenness variance, you can calculate road roughness variance by following formula (5)：
Wherein, Inq is the mean value of square distance sum, and n represents the total number of sample points in running region point cloud.
S122, road roughness grade is determined based on the road roughness variance and default decision procedure calculated.
In embodiments of the present invention, road roughness grade may include the first estate, the second grade, the tertiary gradient and the 4th
Grade, such as it is divided into 0,1,2,3 four grade from low to high.Specifically, step S122 specifically may include：
When current road unevenness grade is the first estate, if the road roughness variance calculated is more than first in advance
If during angle threshold lower limiting value, then current road roughness grade is determined as the second grade；When current road unevenness etc.
It, ought if the road roughness variance calculated is more than the second predetermined angle thresholding lower limiting value when grade is the second grade
Preceding road roughness grade is determined as the tertiary gradient；If the road roughness variance calculated is less than the first predetermined angle door
When limiting upper limit value, then current road roughness grade is determined as the first estate；When current road unevenness grade is third
During grade, if the road roughness variance calculated is more than third predetermined angle threshold value, by current Uneven road
Degree grade is determined as the fourth estate；If the road roughness variance calculated is less than the second predetermined angle thresholding upper limit value,
Current road roughness grade is then determined as the second grade；When current road unevenness grade is the fourth estate, if
When the road roughness variance calculated is more than the 4th predetermined angle threshold value, then current road roughness grade is determined as
The tertiary gradient.
Specifically, road roughness grade R is determined in real time as follows：
As R=0, if Inq>Thrs01, then R=1；
As R=1, if Inq>Thrs12, R=2；If Inq<Thrs10, then R=0；
As R=2, if Inq>Thrs23, R=3；If Inq<Thrs21, then R=1；
As R=3, if Inq<Thrs32, then R=2；
Wherein, Thrs01, Thrs10, Thrs12, Thrs21, Thrs23, Thrs32 are the predetermined angle door that grade redirects
Limit value is corresponded under the first predetermined angle thresholding lower limiting value, the first predetermined angle thresholding upper limit value, the second predetermined angle thresholding respectively
Limit value, the second predetermined angle thresholding upper limit value, third predetermined angle threshold value and the 4th predetermined angle threshold value, between them
Relationship is：Thrs23>Thrs32>Thrs12>Thrs21>Thrs01>Thrs10.In the example of the present invention, each threshold value
Value be respectively：Thrs10=1.444 × 10-5, Thrs01=1.6 × 10-5, Thrs21=5.776 × 10-5, Thrs12=
6.084×10-5, Thrs32=2.310 × 10-4, Thrs23=2.372 × 10-4。
After road roughness grade is determined, unevenness computing module 5 can instruct corresponding road roughness grade
Reminding module 6 is sent to, to export corresponding alerting signal.In one exemplary embodiment of the invention, reminding module 6 can
Including 3 indicator lights.It is instructed corresponding to the different road roughness grades received, the corresponding operating of reminding module 6 can be such as
Shown in lower：
When reminding module 6 receives the road roughness grade instruction corresponding to R=0,3 indicator lights is controlled to be in and are closed
Closed state, i.e. 3 lamps are gone out entirely；
When reminding module 6 receives the road roughness grade instruction corresponding to R=1, one in 3 indicator lights is controlled
It is a to be in opening, i.e., bright 1 lamp；
When reminding module 6 receives the road roughness grade instruction corresponding to R=2,2 in 3 indicator lights are controlled
It is a to be in opening, i.e., bright 2 lamps；
When reminding module 6 receives the road roughness grade instruction corresponding to R=3, control at 3 indicator lights wholes
In opening, i.e. 3 lamp all lights.
Based on same inventive concept, the embodiment of the present invention additionally provides a kind of vehicle-mounted road surface unevenness recognition methods, due to
The principle that this method solves the problems, such as is similar to aforementioned identification system, therefore the implementation of this method may refer to the reality of aforementioned system
It applies, overlaps will not be repeated.
As shown in figure 4, vehicle-mounted road surface unevenness recognition methods provided in an embodiment of the present invention, includes the following steps：
S210, the binocular image for obtaining vehicle front road surface.
S220, the reception binocular image and the binocular image to being received carry out three-dimensional reconstruction and to transitable cloud
It extracts.
S230, acquisition speed signal and steering wheel angle signal.
S240, receive described in passage points cloud, speed signal and steering wheel angle signal and can be passed through based on what is received
Point cloud, speed signal and steering wheel angle signal calculate road roughness, and determine road surface based on the road roughness calculated
Unevenness grade and the corresponding road roughness grade instruction of transmission.
S250, it receives the road roughness grade instruction and instructs to export based on the road roughness grade received
Corresponding alerting signal.
The no stringent sequence of execution of step S210 and S230.
Further, the method for the present embodiment further includes：The binocular camera is demarcated to obtain camera calibration ginseng
Number；Parallax range, lens focus, spin matrix, translation vector, the camera lens that the camera calibration parameter is included between two cameras are abnormal
Variable coefficient and optical centre bias position.
Further, step S220 specifically includes following steps：
S221, school is carried out to the binocular image received using default bearing calibration and acquired camera calibration parameter
Just, the binocular image after being corrected.
In the step, corrected using BOUGUET binoculars polar curve, the binocular image after being corrected.
S222, Stereo matching is carried out to the binocular image after correction using preset matching method, obtains the figure of binocular image
As the correspondence between pixel.
In the step, half global Stereo Matching Algorithm can be used, Stereo matching is carried out to the binocular image after correction, so as to
The correspondence between binocular image is obtained, such as pixel P (x in left image after being corrected1,y1) right image after calibration
Middle corresponding points position P ' (x2,y1)。
Correspondence and camera calibration parameter between S223, the image slices vegetarian refreshments based on the binocular image calculate
Three-dimensional coordinate of the binocular image in corresponding coordinate system after Stereo matching；Wherein, the corresponding coordinate system of binocular image be with
The image corresponds to the optical center of camera as origin, and the vertical imaging plane of Z axis is outside, X-axis along imaging plane level to the right, Y-axis along into
The coordinate system of image plane straight down.
The particular content of the step can be found in the step S112 in previous embodiment.
S224, the three-dimensional coordinate of the binocular image calculated is subjected to coordinate conversion so that the three-dimensional under camera coordinates system
Coordinate points cloud is converted to the point cloud under bodywork reference frame；Wherein, bodywork reference frame is the X axis using vehicle rear axle center as origin
Before, Y-axis is left, the coordinate system of Z axis vertically upward.
The particular content of the step can be found in the step S113 in previous embodiment.
S225, by the X-axis of bodywork reference frame be less than pre-determined distance threshold value three-dimensional point cloud be divided into X-Y plane it is pre-
The grid of set pattern lattice, by the difference of Z-direction height maxima and minimum value in the grid no more than the point in the grid of predetermined threshold value
Cloud extracts, and obtaining can passage points cloud.
The particular content of the step can be found in the step S114 in previous embodiment.
Further, step S240 specifically includes following steps：
S241, using default calculation formula, based on reception can passage points cloud, speed signal and steering wheel angle signal meter
The running region of vehicle is calculated, obtains running region point cloud.
S242, based on the running region point cloud computing road roughness variance.
The step specifically includes：
(1) based on the running region point cloud and preset algorithm, the datum plane of fitting is determined；
(2) each sampled point in the running region point cloud is calculated to the distance of the datum plane；
(3) it will sum after the progress square of the distance of each sampled point to the datum plane, obtain square distance
And value；
(4) it by the total number of sample points in obtained square distance and value divided by running region point cloud, obtains described
Road roughness variance.
S243, road roughness grade is determined based on the road roughness variance and default decision procedure calculated.
In the present invention, the road roughness grade may include the first estate, the second grade, the tertiary gradient and 4th etc.
Grade；Wherein：
When current road unevenness grade is the first estate, if the road roughness variance calculated is more than first in advance
If during angle threshold lower limiting value, then current road roughness grade is determined as the second grade；
When current road unevenness grade is the second grade, if the road roughness variance calculated is more than second in advance
If during angle threshold lower limiting value, then current road roughness grade is determined as the tertiary gradient；If the road surface calculated is not
When Pingdu variance is less than the first predetermined angle thresholding upper limit value, then current road roughness grade is determined as the first estate；
When current road unevenness grade is the tertiary gradient, if the road roughness variance calculated is pre- more than third
If during angle threshold value, then current road roughness grade is determined as the fourth estate；If the road roughness calculated
When variance is less than the second predetermined angle thresholding upper limit value, then current road roughness grade is determined as the second grade；
When current road unevenness grade is the fourth estate, if the road roughness variance calculated is more than the 4th in advance
If during angle threshold value, then current road roughness grade is determined as the tertiary gradient.
The particular content of step S241 to S243 can be found in the step S120 to S122 of previous embodiment.
Above steps can realize that details are not described herein by aforementioned each module.And in the absence of conflict,
The feature in embodiment and embodiment in the application can be combined with each other.
Embodiment described above, specific embodiment only of the invention, to illustrate technical scheme of the present invention rather than
It is limited, protection scope of the present invention is not limited thereto, although having been carried out in detail to the present invention with reference to the foregoing embodiments
Illustrate, it will be understood by those of ordinary skill in the art that：Any one skilled in the art the invention discloses
In technical scope, still can modify to the technical solution recorded in previous embodiment or can readily occur in variation or
Person carries out equivalent replacement to which part technical characteristic；And these modifications, variation or replacement, do not make corresponding technical solution
Essence is detached from the spirit and scope of technical solution of the embodiment of the present invention, should be covered by the protection scope of the present invention.Therefore,
Protection scope of the present invention described should be subject to the protection scope in claims.
Claims (10)
1. a kind of vehicle-mounted road surface unevenness identifying system based on binocular camera, which is characterized in that including image collection module, figure
As processing module, speed data collection module, steering wheel angle acquisition module, unevenness computing module and reminding module；
Described image acquisition module is binocular camera, for the binocular image that obtains vehicle front road surface and by the binocular figure of acquisition
As being sent to described image processing module；
Described image processing module is used to the binocular image received carrying out three-dimensional reconstruction and transitable cloud is carried out
Extraction and by extraction can passage points cloud be sent to the unevenness computing module；
The speed data collection module is used to acquire speed signal and is sent to the unevenness computing module；
The steering wheel angle acquisition module is used to acquire steering wheel angle signal and is sent to the unevenness computing module；
The unevenness computing module for based on reception can passage points cloud, speed signal and steering wheel angle signal calculate road
Face unevenness, and road roughness grade is determined and by corresponding road roughness etc. based on the road roughness calculated
Grade instruction is sent to the reminding module；
The reminding module is used to be instructed to export corresponding alerting signal according to the road roughness grade received.
2. system according to claim 1, which is characterized in that camera calibration module is further included, for the binocular phase
Machine is demarcated to obtain camera calibration parameter and be sent to described image processing module；The camera calibration parameter includes two
Parallax range, lens focus, spin matrix, translation vector, distortion coefficients of camera lens and optical centre bias position between camera.
3. system according to claim 2, which is characterized in that described image processing module is used for the binocular received
Image carries out three-dimensional reconstruction and extracts transitable cloud to specifically include：
The binocular image received is corrected using default bearing calibration and acquired camera calibration parameter, is corrected
Binocular image afterwards；
Using preset matching method to after correction binocular image carry out Stereo matching, obtain binocular image image slices vegetarian refreshments it
Between correspondence；
Correspondence and camera calibration parameter between image slices vegetarian refreshments based on the binocular image, calculate through Stereo matching
Three-dimensional coordinate of the binocular image afterwards in corresponding coordinate system；Wherein, the corresponding coordinate system of binocular image is corresponded to the image
The optical center of camera is origin, and the vertical imaging plane of Z axis is outside, and along imaging plane level to the right, Y-axis is vertical along imaging plane for X-axis
Downward coordinate system；
The three-dimensional coordinate of the binocular image calculated is subjected to coordinate conversion so that the three-dimensional coordinate point cloud under camera coordinates system turns
The point cloud being changed under bodywork reference frame；Wherein, bodywork reference frame is using vehicle rear axle center as origin, and before X axis, Y-axis is left,
The coordinate system of Z axis vertically upward；
The three-dimensional point cloud for being less than pre-determined distance threshold value in the X-axis of bodywork reference frame is divided into predetermined dimension in X-Y plane
Grid carries the difference of Z-direction height maxima and minimum value in the grid no more than the point cloud in the grid of predetermined threshold value
It takes, obtaining can passage points cloud.
4. system according to claim 3, which is characterized in that the unevenness computing module is used for leading to based on reception
Row point cloud, speed signal and steering wheel angle signal calculate road roughness, and determine road based on the road roughness calculated
Face unevenness grade, specifically includes：
Using default calculation formula, based on reception can passage points cloud, speed signal and steering wheel angle signal calculate vehicle
Running region obtains running region point cloud；
Based on the running region point cloud computing road roughness variance；
Road roughness grade is determined based on the road roughness variance and default decision procedure calculated.
5. system according to claim 4, which is characterized in that based on the running region point cloud computing road roughness side
Difference includes：
Based on the running region point cloud and preset algorithm, the datum plane of fitting is determined；
Each sampled point in the running region point cloud is calculated to the distance of the datum plane；
It will sum after the distance progress square of each sampled point to the datum plane, obtain square distance and value；
By the total number of sample points in obtained square distance and value divided by running region point cloud, the Uneven road is obtained
Spend variance.
6. system according to claim 4, which is characterized in that the road roughness grade includes the first estate, second
Grade, the tertiary gradient and the fourth estate；Wherein, road surface is determined based on the road roughness variance and default decision procedure that are calculated
Unevenness grade includes：
When current road unevenness grade is the first estate, if the road roughness variance calculated is more than the first preset angle
When spending thresholding lower limiting value, then current road roughness grade is determined as the second grade；
When current road unevenness grade is the second grade, if the road roughness variance calculated is more than the second preset angle
When spending thresholding lower limiting value, then current road roughness grade is determined as the tertiary gradient；If the road roughness calculated
When variance is less than the first predetermined angle thresholding upper limit value, then current road roughness grade is determined as the first estate；
When current road unevenness grade is the tertiary gradient, if the road roughness variance calculated is more than third preset angle
When spending threshold value, then current road roughness grade is determined as the fourth estate；If the road roughness variance calculated
During less than the second predetermined angle thresholding upper limit value, then current road roughness grade is determined as the second grade；
When current road unevenness grade is the fourth estate, if the road roughness variance calculated is more than the 4th preset angle
When spending threshold value, then current road roughness grade is determined as the tertiary gradient.
7. a kind of vehicle-mounted road surface unevenness recognition methods, which is characterized in that including：
Obtain the binocular image on vehicle front road surface；
It receives the binocular image and the binocular image to being received carries out three-dimensional reconstruction and transitable cloud is extracted；
Acquire speed signal and steering wheel angle signal；
It can passage points cloud, speed signal and steering wheel angle signal and can passage points cloud, speed based on what is received described in reception
Signal and steering wheel angle signal calculate road roughness, and determine road roughness etc. based on the road roughness calculated
Grade and the corresponding road roughness grade instruction of transmission；
It receives the road roughness grade instruction and is instructed based on the road roughness grade received and carried accordingly to export
Awake signal.
8. it the method according to the description of claim 7 is characterized in that further includes：The binocular camera is demarcated to obtain
Camera calibration parameter；The camera calibration parameter includes parallax range between two cameras, lens focus, spin matrix, translation
Vector, distortion coefficients of camera lens and optical centre bias position.
9. according to the method described in claim 8, it is characterized in that, described receive the binocular image and the binocular to being received
Image carries out three-dimensional reconstruction and extracts transitable cloud to specifically include：
The binocular image received is corrected using default bearing calibration and acquired camera calibration parameter, is corrected
Binocular image afterwards；
Using preset matching method to after correction binocular image carry out Stereo matching, obtain binocular image image slices vegetarian refreshments it
Between correspondence；
Correspondence and camera calibration parameter between image slices vegetarian refreshments based on the binocular image, calculate through Stereo matching
Three-dimensional coordinate of the binocular image afterwards in corresponding coordinate system；Wherein, the corresponding coordinate system of binocular image is corresponded to the image
The optical center of camera is origin, and the vertical imaging plane of Z axis is outside, and along imaging plane level to the right, Y-axis is vertical along imaging plane for X-axis
Downward coordinate system；
The three-dimensional coordinate of the binocular image calculated is subjected to coordinate conversion so that the three-dimensional coordinate point cloud under camera coordinates system turns
The point cloud being changed under bodywork reference frame；Wherein, bodywork reference frame is using vehicle rear axle center as origin, and before X axis, Y-axis is left,
The coordinate system of Z axis vertically upward；
The three-dimensional point cloud for being less than pre-determined distance threshold value in the X-axis of bodywork reference frame is divided into predetermined dimension in X-Y plane
Grid carries the difference of Z-direction height maxima and minimum value in the grid no more than the point cloud in the grid of predetermined threshold value
It takes, obtaining can passage points cloud.
It 10. the method according to the description of claim 7 is characterized in that can passage points cloud, speed signal and side described in the reception
To disk angular signal and based on received can passage points cloud, speed signal and steering wheel angle signal calculate road roughness,
And road roughness grade is determined based on the road roughness calculated, it specifically includes：
Using default calculation formula, based on reception can passage points cloud, speed signal and steering wheel angle signal calculate vehicle
Running region obtains running region point cloud；
Based on the running region point cloud computing road roughness variance；
Road roughness grade is determined based on the road roughness variance and default decision procedure calculated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711178413.9A CN108151681B (en) | 2017-11-23 | 2017-11-23 | Binocular camera-based vehicle-mounted road surface unevenness recognition system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711178413.9A CN108151681B (en) | 2017-11-23 | 2017-11-23 | Binocular camera-based vehicle-mounted road surface unevenness recognition system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108151681A true CN108151681A (en) | 2018-06-12 |
| CN108151681B CN108151681B (en) | 2019-12-31 |
Family
ID=62469036
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711178413.9A Active CN108151681B (en) | 2017-11-23 | 2017-11-23 | Binocular camera-based vehicle-mounted road surface unevenness recognition system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108151681B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108973574A (en) * | 2018-08-12 | 2018-12-11 | 苏州青科艾莉电子科技有限公司 | A kind of ride height control method and apparatus based on pavement grade |
| CN109855574A (en) * | 2019-02-01 | 2019-06-07 | 广东工业大学 | A kind of weld seam side surface roughness detecting method, device, equipment and storage medium |
| CN110942010A (en) * | 2019-11-22 | 2020-03-31 | 斑马网络技术有限公司 | Detection method and device for determining road surface undulation state based on vehicle-mounted camera |
| CN111536981A (en) * | 2020-04-23 | 2020-08-14 | 中国科学院上海技术物理研究所 | Embedded binocular non-cooperative target relative pose measuring method |
| CN111596090A (en) * | 2020-06-17 | 2020-08-28 | 中国第一汽车股份有限公司 | Method and device for measuring vehicle running speed, vehicle and medium |
| CN113569958A (en) * | 2021-07-29 | 2021-10-29 | 清华大学苏州汽车研究院（吴江） | Laser point cloud data clustering method, device, equipment and medium |
| CN113674275A (en) * | 2021-10-21 | 2021-11-19 | 北京中科慧眼科技有限公司 | Dense disparity map-based road surface unevenness detection method and system and intelligent terminal |
| CN113696896A (en) * | 2021-08-24 | 2021-11-26 | 中国第一汽车股份有限公司 | Road surface information processing method, device, electronic equipment and storage medium |
| CN114919659A (en) * | 2022-04-22 | 2022-08-19 | 一汽奔腾轿车有限公司 | Method and system for inhibiting wheel shimmy based on EPS |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0948298A (en) * | 1995-08-09 | 1997-02-18 | Mitsubishi Motors Corp | Object position measuring method in running passage |
| CN1218355A (en) * | 1998-11-24 | 1999-06-02 | 杨更新 | Automatic driving system of vehicle |
| EP1079241A2 (en) * | 1999-08-27 | 2001-02-28 | Honda Giken Kogyo Kabushiki Kaisha | Inter-vehicle distance measuring apparatus measuring time difference between each detection time of same road surface condition |
| CN1785723A (en) * | 2005-09-09 | 2006-06-14 | 中国科学院自动化研究所 | Vehicle imbedding type system |
| CN101319485A (en) * | 2008-07-16 | 2008-12-10 | 张小虎 | Photographing measurement method for detecting highroad pavement planeness |
| JP4415704B2 (en) * | 2004-03-03 | 2010-02-17 | 株式会社アドヴィックス | Road surface roughness estimation device and road surface roughness estimation method |
| CN202134106U (en) * | 2011-07-29 | 2012-02-01 | 四川中唯交通科技有限公司 | Intelligent vehicle-mounted road condition collector |
| GB2510672A (en) * | 2013-12-02 | 2014-08-13 | Daimler Ag | Maintaining ride comfort by adapting vehicle speed according to road roughness |
| CN104276174A (en) * | 2013-07-03 | 2015-01-14 | 沃尔沃汽车公司 | A vehicle system, a vehicle and a method for autonomous road irregularity avoidance |
| KR20150098987A (en) * | 2014-02-21 | 2015-08-31 | 주식회사 만도 | Apparatus and method for deciding hand-off or hand-on of steering wheel |
| EP3178674A1 (en) * | 2015-12-09 | 2017-06-14 | The Goodyear Tire & Rubber Company | Tire sensor-based robust road surface roughness classification system and method |
-
2017
- 2017-11-23 CN CN201711178413.9A patent/CN108151681B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0948298A (en) * | 1995-08-09 | 1997-02-18 | Mitsubishi Motors Corp | Object position measuring method in running passage |
| CN1218355A (en) * | 1998-11-24 | 1999-06-02 | 杨更新 | Automatic driving system of vehicle |
| EP1079241A2 (en) * | 1999-08-27 | 2001-02-28 | Honda Giken Kogyo Kabushiki Kaisha | Inter-vehicle distance measuring apparatus measuring time difference between each detection time of same road surface condition |
| JP4415704B2 (en) * | 2004-03-03 | 2010-02-17 | 株式会社アドヴィックス | Road surface roughness estimation device and road surface roughness estimation method |
| CN1785723A (en) * | 2005-09-09 | 2006-06-14 | 中国科学院自动化研究所 | Vehicle imbedding type system |
| CN101319485A (en) * | 2008-07-16 | 2008-12-10 | 张小虎 | Photographing measurement method for detecting highroad pavement planeness |
| CN202134106U (en) * | 2011-07-29 | 2012-02-01 | 四川中唯交通科技有限公司 | Intelligent vehicle-mounted road condition collector |
| CN104276174A (en) * | 2013-07-03 | 2015-01-14 | 沃尔沃汽车公司 | A vehicle system, a vehicle and a method for autonomous road irregularity avoidance |
| GB2510672A (en) * | 2013-12-02 | 2014-08-13 | Daimler Ag | Maintaining ride comfort by adapting vehicle speed according to road roughness |
| KR20150098987A (en) * | 2014-02-21 | 2015-08-31 | 주식회사 만도 | Apparatus and method for deciding hand-off or hand-on of steering wheel |
| EP3178674A1 (en) * | 2015-12-09 | 2017-06-14 | The Goodyear Tire & Rubber Company | Tire sensor-based robust road surface roughness classification system and method |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108973574A (en) * | 2018-08-12 | 2018-12-11 | 苏州青科艾莉电子科技有限公司 | A kind of ride height control method and apparatus based on pavement grade |
| CN109855574A (en) * | 2019-02-01 | 2019-06-07 | 广东工业大学 | A kind of weld seam side surface roughness detecting method, device, equipment and storage medium |
| CN110942010A (en) * | 2019-11-22 | 2020-03-31 | 斑马网络技术有限公司 | Detection method and device for determining road surface undulation state based on vehicle-mounted camera |
| CN110942010B (en) * | 2019-11-22 | 2023-06-02 | 斑马网络技术有限公司 | Detection method and device for determining road surface fluctuation state based on vehicle-mounted camera |
| CN111536981A (en) * | 2020-04-23 | 2020-08-14 | 中国科学院上海技术物理研究所 | Embedded binocular non-cooperative target relative pose measuring method |
| CN111536981B (en) * | 2020-04-23 | 2023-09-12 | 中国科学院上海技术物理研究所 | Embedded binocular non-cooperative target relative pose measurement method |
| CN111596090A (en) * | 2020-06-17 | 2020-08-28 | 中国第一汽车股份有限公司 | Method and device for measuring vehicle running speed, vehicle and medium |
| CN113569958A (en) * | 2021-07-29 | 2021-10-29 | 清华大学苏州汽车研究院（吴江） | Laser point cloud data clustering method, device, equipment and medium |
| CN113569958B (en) * | 2021-07-29 | 2024-04-12 | 清华大学苏州汽车研究院（吴江） | Laser point cloud data clustering method, device, equipment and medium |
| CN113696896B (en) * | 2021-08-24 | 2023-02-28 | 中国第一汽车股份有限公司 | Road surface information processing method, device, electronic equipment and storage medium |
| CN113696896A (en) * | 2021-08-24 | 2021-11-26 | 中国第一汽车股份有限公司 | Road surface information processing method, device, electronic equipment and storage medium |
| CN113674275A (en) * | 2021-10-21 | 2021-11-19 | 北京中科慧眼科技有限公司 | Dense disparity map-based road surface unevenness detection method and system and intelligent terminal |
| CN114919659A (en) * | 2022-04-22 | 2022-08-19 | 一汽奔腾轿车有限公司 | Method and system for inhibiting wheel shimmy based on EPS |
| CN114919659B (en) * | 2022-04-22 | 2023-11-21 | 一汽奔腾轿车有限公司 | EPS-based method and system for inhibiting wheel shimmy |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108151681B (en) | 2019-12-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108151681A (en) | A kind of vehicle-mounted road surface unevenness identifying system and method based on binocular camera | |
| CN106256606B (en) | A kind of lane departure warning method based on vehicle-mounted binocular camera | |
| CN103996053B (en) | Lane departure alarm method based on machine vision | |
| CN104859563B (en) | lane departure warning method and system | |
| CN111829549B (en) | Snow pavement virtual lane line projection method based on high-precision map | |
| CN110207650A (en) | Automobile-used highway height-limiting frame height measurement method and device based on binocular vision | |
| CN105206109B (en) | A kind of vehicle greasy weather identification early warning system and method based on infrared CCD | |
| WO2011039989A1 (en) | Vehicle-surroundings monitoring device | |
| CN105550665A (en) | Method for detecting pilotless automobile through area based on binocular vision | |
| CN107093180A (en) | The wet road surface condition detection of the view-based access control model splashed backward using tire | |
| CN107697154A (en) | Steering angle correcting method, electronic power assist steering system and vehicle | |
| GB2494528A (en) | A vehicle suspension control including a vehicle mounted time of flight camera | |
| WO2017089057A1 (en) | Method and device for determining a type of the road on which a vehicle is driving | |
| CN106864458A (en) | It is a kind of automatic around barrier system and method, intelligent automobile | |
| CN110210339B (en) | Method for identifying road jolt by fusing multiple sensors for ECAS system | |
| CN105599765A (en) | Lane departure judging and early warning method | |
| CN105160356A (en) | Method and system for fusing sensor data of vehicle active safety system | |
| CN111559308B (en) | Vehicle adaptive curve lighting system and control method thereof | |
| CN110176151A (en) | A kind of method, apparatus, medium and the equipment of determining parking behavior | |
| WO2016020718A1 (en) | Method and apparatus for determining the dynamic state of a vehicle | |
| CN104108391B (en) | The method and apparatus of the drive propulsion of vehicle of the control with rut compensation | |
| Leng et al. | Tire-road peak adhesion coefficient estimation method based on fusion of vehicle dynamics and machine vision | |
| WO2023131253A1 (en) | Autonomous driving method and system capable of recognizing road conditions, and vehicle | |
| CN110888441A (en) | Gyroscope-based wheelchair control system | |
| CN113320518B (en) | Method and system for preventing vehicle from sliding after parking on ramp |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |