CN112991732A - Real-time curve rollover early warning system and method based on binocular camera - Google Patents

Real-time curve rollover early warning system and method based on binocular camera Download PDF

Info

Publication number
CN112991732A
CN112991732A CN202110225599.9A CN202110225599A CN112991732A CN 112991732 A CN112991732 A CN 112991732A CN 202110225599 A CN202110225599 A CN 202110225599A CN 112991732 A CN112991732 A CN 112991732A
Authority
CN
China
Prior art keywords
curve
early warning
binocular camera
processing module
module
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
Application number
CN202110225599.9A
Other languages
Chinese (zh)
Inventor
李稳
关威
毕秋实
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangsu XCMG Construction Machinery Institute Co Ltd
Original Assignee
Jiangsu XCMG Construction Machinery Institute Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jiangsu XCMG Construction Machinery Institute Co Ltd filed Critical Jiangsu XCMG Construction Machinery Institute Co Ltd
Priority to CN202110225599.9A priority Critical patent/CN112991732A/en
Publication of CN112991732A publication Critical patent/CN112991732A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B3/00—Audible signalling systems; Audible personal calling systems
    • G08B3/10—Audible signalling systems; Audible personal calling systems using electric transmission; using electromagnetic transmission
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Electromagnetism (AREA)
  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Emergency Management (AREA)
  • Image Processing (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention relates to a binocular camera-based real-time curve rollover early warning system and method, belongs to the technical field of automobile active safety control, and comprises an image acquisition module for acquiring images of a straight road and a curve in real time. And the image processing module is used for processing the image in real time, identifying the road, judging whether the front is a curve or not, calculating the curvature of the curve, and finding out the characteristic point of the curve through a binocular camera to measure the distance of the curve. And the detection module is used for measuring the speed, the roll angle and the roll angle speed of the detected vehicle. And the information processing module is used for integrating various information and processing the various information. And the display module is used for displaying the front road information transmitted by the information processing module in real time and carrying out voice reminding on the driver. The system can detect the front road information in real time, and the critical speed of the vehicle passing through the curve is obtained by combining the detection module, so that the early warning can be given to the driver about to enter the curve, and the driving safety of the road is improved.

Description

Real-time curve rollover early warning system and method based on binocular camera
Technical Field
The invention relates to the technical field of automobile active safety control, in particular to a binocular camera-based real-time curve rollover early warning system and method.
Background
With the rapid development of economy, the automobile becomes a preferred travel tool for modern people, and the automobile holding capacity is rapidly increased. The increasing number of vehicles leads to an increasing trend of traffic accidents. Traffic accidents cause great loss to people's lives and properties, so the traffic safety problem gradually becomes a focus of attention of people. The curve is taken as a road section where traffic accidents are easy to happen, has high accident occurrence frequency and high harm degree, and is one of key research objects of road traffic safety at home and abroad. In a curve accident, rollover of the vehicle caused by too high driving speed occupies a considerable proportion, and is more attractive. Moreover, the hazards caused by rollover are more unlimited for large vehicles such as all-terrain cranes.
At present, the common road curve early warning device is provided with a curve set deceleration strip, a curve speed limit sign, a curve warning sign and the like. The curve early warning device has an unsatisfactory early warning effect, a plurality of rollover prevention systems are developed at present, and the practical application in rollover prevention control is limited because the dynamic performance of a suspension is not considered in some systems. Some of them lack consideration of dynamic motion of the vehicle, and thus their predictability is insufficient. Still other applications are costly.
Disclosure of Invention
The invention provides a binocular camera-based real-time curve rollover early warning system and method for overcoming the defects in the prior art, which have an early warning effect on a vehicle, ensure the safety of the vehicle from passing a curve and prevent the vehicle from rolling over.
In order to solve the technical problems, the invention is realized by the following technical scheme:
the utility model provides a real-time bend early warning system of turning on one's side based on binocular camera, includes: the device comprises an image acquisition module, an image processing module, a detection module, an information processing module and a display module;
the image acquisition module transmits a road image in front of a vehicle acquired in real time to the image processing module;
the image processing module judges whether the front is a curve or not, and transmits the obtained road surface edge of the front road, curve curvature information and distance information from the curve to the information processing module;
the detection module comprises an Inertial Measurement Unit (IMU), a roll angle sensor and an angular velocity sensor, and transmits the measured information to the information processing module;
the information processing module receives the vehicle parameters, the information transmitted by the image processing module and the detection module, calculates and generates rollover early warning information and transmits the rollover early warning information to the display module;
the display module is used for displaying the rollover early warning information and carrying out voice reminding on the driver.
Furthermore, the image acquisition module is composed of a binocular camera and is installed below a windshield at the front end of the vehicle.
Further, the detection module is installed at the position of the mass center of the vehicle inertial reference frame.
Further, the display module comprises an LED display and an alarm.
Furthermore, the information processing module consists of a data acquisition card and a computer.
Furthermore, the image processing module is a data acquisition card, and the information processing module is a computer.
A real-time curve rollover early warning method based on a binocular camera comprises the following steps:
the method comprises the following steps: the image acquisition module is used for acquiring a front road image in real time, carrying out graying processing on the image, extracting ROI (region of interest), median filtering, histogram equalization, inversion processing and the like, enhancing contrast, carrying out grayscale morphological processing operation, and enhancing road boundary characteristics;
and step two, carrying out straight line detection by utilizing Hough transformation to obtain straight line parameters, and then extracting the inner boundary of the lane. An improved river point searching algorithm is utilized, and the algorithm is used for extracting boundary characteristic points by using straight lines of Hough transformation as guide lines. After the boundary feature points are obtained, the curvature of the curve is calculated and fitted by using a random sample consensus (RANSAC) algorithm in combination with a hyperbolic model. And judging whether a curve exists in the front or not by utilizing the angles of the two lines, considering that the curve exists in the front when the angle is more than 10 degrees, and calculating the coordinates of a curve entering point in a pixel coordinate system by combining a straight line and the curve which are fitted by Hough transformation. And calculating the distance from the curve by using a binocular transverse parallel mode stereoscopic vision model, and improving the ranging precision through three-dimensional calibration.
And step three, providing a vehicle rollover index (CLRI) with prediction capability according to the load transfer ratio contour line. And (3) obtaining a critical side inclination angle, a critical side inclination angle speed and a critical side inclination angle acceleration, combining the curve curvature obtained in the step two to obtain a critical speed of the curve, and then combining the vehicle running speed measured by an Inertia Measurement Unit (IMU) to early warn a driver.
Further, the method for extracting the inner boundary in the second step comprises the following steps:
extracting the inner boundary of the lane by using a method based on the vector cross multiplication principle, and assuming that n boundary lines L existxi right sideLyi right side，Lxi left sideLLeft of yiI is 1,2,3 … n, and any starting point L is selectedxj RightAnd an end point Lyk right sideJ, k is 1,2,3 … n, and cross-product the operation.
Fjk right side＝Lxj RightLxi right side×Lxj RightLyk right side
Fjk right side＝Lxj RightLyi right side×Lxj RightLyk right side
In the scope of the definition domain, Fjk right sideAll being positive, the boundary line Lxi right sideLyi right sideIs the inner boundary.
Further, the hyperbolic model in the second step is as follows:
Figure BDA0002955806940000031
Figure BDA0002955806940000032
wherein (u, v) representsCoordinates of points on the curve, h represents the v coordinate of the horizontal asymptote, i.e. the vertical coordinate of the vanishing line, bx,byIs a left and right boundary, kx,kyThe slope of the left and right borders of the road, t the curvature of the road curve, and vp the abscissa of the vanishing line.
Further, the vehicle rollover index in the third step is as follows:
Figure BDA0002955806940000033
in the formula (I), the compound is shown in the specification,
Figure BDA0002955806940000034
is the roll angle velocity at point a,
Figure BDA0002955806940000039
is the roll angle velocity at point B,
Figure BDA0002955806940000035
is the roll angular acceleration at point a. Point a represents the roll state of the vehicle in the roll phase plane at the present time, and the tangent of the roll phase trajectory at point a is
Figure BDA0002955806940000036
Point B is the intersection of the tangent and the load transfer ratio contour.
Figure BDA0002955806940000037
Figure BDA0002955806940000038
LTR, the load transfer ratio.
The invention has the beneficial effects that:
1. the method fully considers the influences of the unsmooth road boundary, the complex environment of the original visual area and the like, can avoid the influence of multiple boundary lines, and has the characteristics of high road identification precision, strong adaptability and the like.
2. The invention fully considers the dynamic performance of the suspension and the dynamic motion of the vehicle, does not treat the vehicle as a rigid body model, and improves the practical application in rollover prevention early warning.
3. The invention has the advantages of low cost, less used devices and strong practicability.
Drawings
FIG. 1 is a schematic view of the mounting location of various sensors of the present invention;
FIG. 2 is a flow chart of an implementation of the real-time curve rollover warning system of the present invention;
FIG. 3 is a flowchart of an image processing method of a road of the present invention;
FIG. 4 is a flow chart of a method of detecting a road curve and a calculation of curvature according to the invention;
FIG. 5 is a flow chart of the pre-determination and ranging of a road curve according to the present invention;
FIG. 6 is a flow chart of the present invention for predicting a critical speed of a curve rollover;
the device comprises a 001-image acquisition module and a 002-detection module.
Detailed Description
The present invention will be described in further detail with reference to the following drawings and specific embodiments. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. It should be noted that the terms "front", "back", "left", "right", "upper" and "lower" used in the following description refer to directions in the drawings, and the terms "inner" and "outer" refer to directions toward and away from the geometric center of a specific part, respectively, and those skilled in the art should not understand that the techniques beyond the scope of the present application are simple and need no inventive adjustment.
In the description of the present invention, unless otherwise specified or limited, the term "connected" is to be understood broadly, and may be, for example, a mechanical connection or an electrical connection, or a communication between two elements, or may be a direct connection or an indirect connection through an intermediate medium, and the specific meaning of the above terms may be understood by those skilled in the art according to specific situations.
As shown in fig. 1 and 2, the real-time curve rollover early warning system and method based on the binocular camera of the invention are applied to the automobile crane and comprise an image acquisition module 001, an image processing module, a detection module 002, an information processing module and a display module. The image acquisition module 001 is composed of a binocular camera and is installed below a windshield at the front end of the vehicle. The binocular camera is provided with depth information and used for acquiring road images in front of a vehicle in real time, and can be used for ranging a curve according to a curve characteristic point obtained by judging whether the curve exists in front or not, wherein the entrance of the curve is of a straight line structure, so that a driver can obtain information such as the curvature of the curve in front, the limit speed of the vehicle passing through the curve, the distance from the curve and the like before the vehicle reaches the curve, the driver has sufficient time to make adjustment, and the driver is warned in a voice mode. The image acquisition module 001 is connected with the image processing module and transmits the acquired image information to the image processing module for processing.
The detection module 002 is composed of an Inertial Measurement Unit (IMU), a roll angle sensor and an angular velocity sensor, and is installed at the position of the mass center of the vehicle inertial reference system for measuring the velocity, the roll angle and the roll angle velocity of the vehicle to be detected.
The information processing module records vehicle parameters, data information acquired by the detection module 002 and road information processed by the image processing module. The vehicle parameters include the height of the center of mass, the roll moment of inertia, the sprung mass, the wheelbase, and the roll center to center of mass distance. The curve rollover index can be obtained through the vehicle parameters and the data information acquired by the detection module 002, the speed of the vehicle at the limit of the curve can be obtained through prediction by combining the road information data acquired by the image processing module, whether the processing information of rollover danger exists or not is judged according to the speed, the obtained information is transmitted to the display module, the display module adopts an LED display and is installed in a cab, the processing information of rollover danger is displayed in the LED display, and the driver is warned by an additional warning device.
In other embodiments, the binocular camera is connected to the image processing module through a data line, the image processing module is a data acquisition card, the information processing module is a computer, and the computer is installed in the cab.
The real-time curve rollover early warning method based on the binocular camera comprises the following steps: as shown in fig. 3, after receiving an image transmitted by a binocular camera, an image processing module firstly grays the image, then divides an area of interest, filters noise through median filtering to enhance picture information, performs histogram equalization on the image, performs inversion processing to enhance contrast, and finally smoothes an edge through grayscale morphological processing.
In the process, firstly, the acquired RGB color image is grayed by a weighted average method, different weights are set for R, G, B three channels, the gray value of the point in the image is calculated, and the complexity of the image is reduced. And ROI extraction is carried out on the image, so that an invalid region in the image is effectively eliminated, and the running speed is improved.
Then, as shown in fig. 4, line detection is performed by using Hough transformation to obtain line parameters, and then a method based on the vector cross-multiplication principle is used to extract the inner boundary of the lane. The improved river searching point algorithm is used, and the straight line of Hough transformation performed by the people in the foregoing is used as a guide line to extract boundary characteristic points. After the boundary feature points are obtained, the invention utilizes a Random Sample Consensus algorithm (RANSAC) in combination with a hyperbolic model to calculate the curvature of the curve and perform fitting. And judging whether a curve exists in the front or not by utilizing the angles of the two lines, considering that the curve exists in the front when the angle is more than 10 degrees, and calculating the coordinates of a curve entering point in a pixel coordinate system by combining a straight line and the curve which are fitted by Hough transformation. And calculating the distance from the curve by using a binocular transverse parallel mode stereoscopic vision model, and improving the ranging precision through three-dimensional calibration.
Extracting the inner boundary of the lane by using a method based on the vector cross multiplication principle, and assuming that n boundary lines L existxi right sideLyi right side，Lxi left sideLLeft of yiI is 1,2,3 … n. Selecting any one starting point Lxj RightAnd an end point Lyk right sideJ, k is 1,2,3 … n. It is cross-multiplied.
Figure BDA0002955806940000052
In the scope of the definition domain, Fjk right sideAll being positive, the boundary line Lxi right sideLyi right sideIs the inner boundary.
As shown in the table I, the improved river searching algorithm starts to search points from bottom to top, the searching process can only move forwards or in parallel, pixel values of a plurality of points around the point are compared through a 5 x 2 template, and the point with the largest mutation of the pixel values is used as a boundary characteristic point.
qx-2,y-1 qx-1,y-1 qx,y-1 qx+1,y-1 qx+2,y-1
qx-2,y qx-1,y qx,y qx+1,y qx+2,y
Table one search template, left and right boundaries general
qx,yThe pixel values of the search starting point are represented, and the other nine are the pixel values of the surrounding nine points, and the pixel values of the nine points are compared to determine the next search starting point as the next search starting point：
qx,y＝max{qx-2,y-1,qx-1,y-1,qx,y-1,qx+1,y-1,qx+2,y-1,qx-2,y,qx-1,y,qx+1,y,qx+2,y} (2)
The hyperbolic model is:
Figure BDA0002955806940000051
in the formula, (u, v) represents coordinates of points on the curve, h represents a v coordinate of a horizontal asymptote, namely a vertical coordinate of a vanishing line, and bx,byIs a left and right boundary, kx,kyThe slope of the left and right borders of the road, t the curvature of the road curve, and vp the abscissa of the vanishing line.
The set of boundary feature points of the left boundary of the road is Q, and the set of boundary feature points of the right boundary of the road is W. And fitting the curve by using a random sampling consistency algorithm, wherein for the characteristic points with the distance less than the set threshold value U, the curve model passes through the characteristic points, N curve models are established after repeating N times, and the curve model with the largest passing number, namely the optimal model, can be selected by comparing the number of the characteristic points passed by each curve model. The more times of repeated iteration, the more accurate the established optimal model is. Curve curvature of optimal model is kq。
Wherein the number of cycles N may be based on the minimum number of iterations N1To determine, assuming that the estimation model needs to select n points, epsilon is the data accuracy (the proportion of local interior points in the original data), P represents the probability that the randomly selected points in the data set are local interior points in the iterative process, and in a certain test, the probability that n points are local interior points is epsilonn；1-εnIs the probability that at least 1 of the n points is an outlier, i.e., a bad model is estimated from the data set. The minimum number of samples N meeting the requirement can be obtained1The method comprises the following steps:
Figure BDA0002955806940000061
assuming that the mean of the outliers is 0 and the variance is σ, the residual of the error fits into the chi-square distribution in dimension r. And the selection of the residual error threshold value U is as follows:
Figure BDA0002955806940000062
in the formula, alpha is confidence probability, U is residual threshold value, and r is dimension of chi-square distribution.
As shown in fig. 5, the straight line and the curved line of both boundaries are compared, whether or not there is a curve ahead is determined based on the angle between the straight line and the curved line, and when the angle is greater than 10 degrees, it is considered that there is a curve ahead. When the angle is judged, according to the included angle between the straight line and the curve of the left boundary and the right boundary of the road, when the two boundaries are larger than 10 degrees, the curve in front is considered to exist, misjudgment caused by errors on one side is prevented, and the straight line and the curve which are fitted by combining Hough transformation are combined for operation, so that the coordinates of a bending point under a pixel coordinate system are calculated.
And finally, calculating the distance from the curve by using a binocular transverse parallel mode stereoscopic vision model, and improving the ranging precision through three-dimensional calibration.
As shown in fig. 6, a Vehicle Rollover Index (cli) with predictive capability can be provided based on the Load Transfer Ratio Contour by the data measured by the detection module 002 and the parameters of the Vehicle itself. The critical speed of the curve can be obtained by obtaining the critical side inclination angle, the side inclination angle speed and the side inclination angle acceleration and combining the curve curvature obtained by the image processing module, and the driver is warned in advance by combining the vehicle running speed measured by an Inertia Measurement Unit (IMU).
If the speed of the vehicle is greater than or equal to the critical vehicle speed of the curve rollover when the vehicle is about to enter the curve, the vehicle is likely to have a rollover accident at the moment, and the display module displays the current road edge, the curvature of the curve, the distance from the curve, the real-time vehicle speed of the vehicle and the critical vehicle speed of the rollover and sends out a voice alarm to prompt the driver to slow down and prevent the vehicle from rollover.
In the prior art, to calculate the Load Transfer Ratio (LTR), the moment when one side wheel lifts off is generally defined as the critical rollover point. In fact, due to the inertia of the vehicle, most high-speed vehicles turn 90 degrees around the axis after the wheels on one side lift off, and the vehicle turns over in the true sense.
Figure BDA0002955806940000071
Wherein LTR is the load transfer ratio, ayLateral acceleration, hrcBeing the roll center-to-center distance,
Figure BDA0002955806940000072
is the roll angle, g is the acceleration of gravity, IxThe moment of inertia of the roll is,
Figure BDA0002955806940000073
acceleration of roll angle, msThe mass of the automobile crane.
The rollover index CLRI is defined as:
Figure BDA0002955806940000074
point a represents the roll state of the vehicle in the roll phase plane at the present time, and the tangent of the roll phase trajectory at point a is
Figure BDA0002955806940000075
Point B is the intersection of the tangent and the load transfer ratio contour.
Figure BDA0002955806940000076
Figure BDA0002955806940000077
LTR, the load transfer ratio.
The rollover index CLRI adopts time scales to quantify the risk of the vehicle rollover. Therefore, when using the CLRI, a prediction time t needs to be set. When the CLRI is greater than the predicted time t, it indicates that no rollover will occur within the predicted time. When the vehicle is just turned over in the prediction time, the data detected by the detection module 002 is recorded, the critical speed of the turning over of the curve can be obtained by combining the vehicle parameters and the road information data obtained by the image processing module, and then the driver is early warned, so that the driving safety of the curve is improved.
It should be understood that although the present invention has been described in terms of embodiments, not every embodiment includes only a single embodiment, and such description is for clarity only, and those skilled in the art will recognize that the embodiments described herein may be combined as a whole to form other embodiments as would be understood by those skilled in the art.
It should be noted that the above embodiments are only for explaining the technical solutions of the present invention, and are not intended to limit the scope of the present invention, and equivalent embodiments or modifications made without departing from the technical spirit of the present invention should be included in the scope of the present invention.

Claims (10)

1. The utility model provides a real-time bend early warning system of turning on one's side based on binocular camera which characterized in that includes: the device comprises an image acquisition module, an image processing module, a detection module, an information processing module and a display module;
the image acquisition module transmits a road image in front of a vehicle acquired in real time to the image processing module;
the image processing module judges whether the front is a curve or not, and transmits the obtained road surface edge of the front road, curve curvature information and distance information from the curve to the information processing module;
the detection module comprises an Inertial Measurement Unit (IMU), a roll angle sensor and an angular velocity sensor, and transmits the measured information to the information processing module;
the information processing module receives the vehicle parameters, the information transmitted by the image processing module and the detection module, calculates and generates rollover early warning information and transmits the rollover early warning information to the display module;
the display module is used for displaying the rollover early warning information and carrying out voice reminding on the driver.
2. The binocular camera based real-time curve rollover early warning system according to claim 1, wherein the image acquisition module is composed of a binocular camera and is installed below a windshield at the front end of the vehicle.
3. The binocular camera based real-time curve rollover warning system according to claim 1, wherein the detection module is installed at a centroid position of a vehicle inertial reference system.
4. The binocular camera based real-time curve rollover warning system according to claim 1, wherein the display module comprises an LED display and an alarm.
5. The binocular camera based real-time curve rollover early warning system as recited in claim 1, wherein the information processing module is composed of a data acquisition card and a computer.
6. The binocular camera based real-time curve rollover early warning system as recited in claim 1, wherein the image processing module is a data acquisition card, and the information processing module is a computer.
7. A real-time curve rollover early warning method based on a binocular camera comprises the following steps:
the method comprises the following steps: the image acquisition module is used for acquiring a front road image in real time, carrying out graying processing on the image, extracting ROI (region of interest), carrying out median filtering, carrying out histogram equalization and negation processing, enhancing contrast, carrying out grayscale morphological processing operation and enhancing road boundary characteristics;
and step two, carrying out straight line detection by utilizing Hough transformation to obtain straight line parameters, and then extracting the inner boundary of the lane. An improved river point searching algorithm is utilized, and the algorithm is used for extracting boundary characteristic points by using straight lines of Hough transformation as guide lines. After the boundary feature points are obtained, the curvature of the curve is calculated and fitted by using a random sample consensus (RANSAC) algorithm in combination with a hyperbolic model. And judging whether a curve exists in the front or not by utilizing the angles of the two lines, considering that the curve exists in the front when the angle is more than 10 degrees, and calculating the coordinates of a curve entering point in a pixel coordinate system by combining a straight line and the curve which are fitted by Hough transformation. And calculating the distance from the curve by using a binocular transverse parallel mode stereoscopic vision model, and improving the ranging precision through three-dimensional calibration.
And step three, providing a vehicle rollover index (CLRI) with prediction capability according to the load transfer ratio contour line. And (3) obtaining a critical side inclination angle, a critical side inclination angle speed and a critical side inclination angle acceleration, combining the curve curvature obtained in the step two to obtain a critical speed of the curve, and then combining the vehicle running speed measured by an Inertia Measurement Unit (IMU) to early warn a driver.
8. The binocular camera-based real-time curve rollover early warning method according to claim 7, wherein the method for extracting the inner boundary in the second step comprises the following steps:
extracting the inner boundary of the lane by using a method based on the vector cross multiplication principle, and assuming that n boundary lines L existxi right sideLyi right side，Lxi left sideLLeft of yiI is 1,2,3 … n, and any starting point L is selectedxj RightAnd an end point Lyk right sideJ, k is 1,2,3 … n, and cross-product the operation.
Fjk right side＝Lxj RightLxi right side×Lxj RightLyk right side
Fjk right side＝Lxj RightLyi right side×Lxj RightLyk right side
In the scope of the definition domain, Fjk right sideAll being positive, the boundary line Lxi right sideLyi right sideIs the inner boundary.
9. The binocular camera-based real-time curve rollover early warning method according to claim 7, wherein the hyperbolic model in the second step is as follows:
Figure FDA0002955806930000021
Figure FDA0002955806930000022
in the formula, (u, v) represents coordinates of points on the curve, h represents a v coordinate of a horizontal asymptote, namely a vertical coordinate of a vanishing line, and bx,byIs a left and right boundary, kx,kyThe slope of the left and right borders of the road, t the curvature of the road curve, and vp the abscissa of the vanishing line.
10. The binocular camera-based real-time curve rollover early warning method according to claim 7, wherein the vehicle rollover indicators in step three are as follows:
Figure FDA0002955806930000023
point a represents the roll state of the vehicle in the roll phase plane at the present time, and the tangent of the roll phase trajectory at point a is
Figure FDA0002955806930000024
Point B is the intersection of the tangent and the load transfer ratio contour.
Figure FDA0002955806930000025
Figure FDA0002955806930000026
LTR, the load transfer ratio.
CN202110225599.9A 2021-03-01 2021-03-01 Real-time curve rollover early warning system and method based on binocular camera Pending CN112991732A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110225599.9A CN112991732A (en) 2021-03-01 2021-03-01 Real-time curve rollover early warning system and method based on binocular camera

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110225599.9A CN112991732A (en) 2021-03-01 2021-03-01 Real-time curve rollover early warning system and method based on binocular camera

Publications (1)

Publication Number Publication Date
CN112991732A true CN112991732A (en) 2021-06-18

Family

ID=76351822

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110225599.9A Pending CN112991732A (en) 2021-03-01 2021-03-01 Real-time curve rollover early warning system and method based on binocular camera

Country Status (1)

Country Link
CN (1) CN112991732A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113263994A (en) * 2021-06-21 2021-08-17 三一汽车起重机械有限公司 All-terrain crane steering protection method and device and all-terrain crane
CN113409591A (en) * 2021-06-23 2021-09-17 广州小鹏自动驾驶科技有限公司 Curve speed limiting method, vehicle-mounted terminal, vehicle and computer readable storage medium
CN114822029A (en) * 2022-04-26 2022-07-29 广州大学 Bridge deck traffic flow load space-time distribution reconstruction method, system and device
CN117346723A (en) * 2021-10-19 2024-01-05 郑州大学 Vehicle-mounted track curvature data full-line detection device

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113263994A (en) * 2021-06-21 2021-08-17 三一汽车起重机械有限公司 All-terrain crane steering protection method and device and all-terrain crane
CN113409591A (en) * 2021-06-23 2021-09-17 广州小鹏自动驾驶科技有限公司 Curve speed limiting method, vehicle-mounted terminal, vehicle and computer readable storage medium
CN117346723A (en) * 2021-10-19 2024-01-05 郑州大学 Vehicle-mounted track curvature data full-line detection device
CN117346723B (en) * 2021-10-19 2024-05-17 郑州大学 Vehicle-mounted track curvature data full-line detection device
CN114822029A (en) * 2022-04-26 2022-07-29 广州大学 Bridge deck traffic flow load space-time distribution reconstruction method, system and device
CN114822029B (en) * 2022-04-26 2023-04-07 广州大学 Bridge deck traffic flow load space-time distribution reconstruction method, system and device

Similar Documents

Publication Publication Date Title
CN112991732A (en) Real-time curve rollover early warning system and method based on binocular camera
CN106919915B (en) Map road marking and road quality acquisition device and method based on ADAS system
CN107038723B (en) Method and system for estimating rod-shaped pixels
JP5297078B2 (en) Method for detecting moving object in blind spot of vehicle, and blind spot detection device
CN104916163B (en) Parking space detection method
CN105206109B (en) A kind of vehicle greasy weather identification early warning system and method based on infrared CCD
CN107972662A (en) To anti-collision warning method before a kind of vehicle based on deep learning
US20230174056A1 (en) Dynamic Distance Estimation Output Generation Based on Monocular Video
CN111002990B (en) Lane departure early warning method and system based on dynamic departure threshold
Labayrade et al. In-vehicle obstacles detection and characterization by stereovision
Guo et al. A multimodal ADAS system for unmarked urban scenarios based on road context understanding
CN104573646A (en) Detection method and system, based on laser radar and binocular camera, for pedestrian in front of vehicle
KR100816377B1 (en) Method and Apparatus for Recognizing Parking Slot Marking by Using Hough Transformation and Parking Assist System Using Same
CN110909705B (en) Road side parking space sensing method and system based on vehicle-mounted camera
JP2015007953A (en) Apparatus, method, and computer readable medium for monitoring the number of passengers in automobile
CN105678287B (en) A kind of method for detecting lane lines based on ridge measurement
CN107097794B (en) System and method for detecting lane line of road
KR20150112656A (en) Method to calibrate camera and apparatus therefor
WO2018149539A1 (en) A method and apparatus for estimating a range of a moving object
CN114092683A (en) Tire deformation identification method and device based on visual feedback and depth network
EP3082069A1 (en) Stereoscopic object detection device and stereoscopic object detection method
CN113657265B (en) Vehicle distance detection method, system, equipment and medium
CN114663859A (en) Sensitive and accurate complex road condition lane deviation real-time early warning system
KR101998584B1 (en) Lane detection apparatus and lane detection method
CN116311136A (en) Lane line parameter calculation method for driving assistance

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication