CN110276988A - A kind of DAS (Driver Assistant System) based on collision warning algorithm - Google Patents

A kind of DAS (Driver Assistant System) based on collision warning algorithm Download PDF

Info

Publication number
CN110276988A
CN110276988A CN201910561949.1A CN201910561949A CN110276988A CN 110276988 A CN110276988 A CN 110276988A CN 201910561949 A CN201910561949 A CN 201910561949A CN 110276988 A CN110276988 A CN 110276988A
Authority
CN
China
Prior art keywords
vehicle
time
gps
collision
detection
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
CN201910561949.1A
Other languages
Chinese (zh)
Inventor
蔡林沁
周思桐
董建功
曹世洲
牟志豪
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.)
Chongqing University of Post and Telecommunications
Original Assignee
Chongqing University of Post and Telecommunications
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 Chongqing University of Post and Telecommunications filed Critical Chongqing University of Post and Telecommunications
Priority to CN201910561949.1A priority Critical patent/CN110276988A/en
Publication of CN110276988A publication Critical patent/CN110276988A/en
Pending legal-status Critical Current

Links

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095—Predicting travel path or likelihood of collision
    • B60W30/0956—Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W50/08—Interaction between the driver and the control system
    • B60W50/14—Means for informing the driver, warning the driver or prompting a driver intervention
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/0055—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot with safety arrangements
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0268—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means
    • G05D1/027—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means comprising intertial navigation means, e.g. azimuth detector
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0268—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means
    • G05D1/0274—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means using mapping information stored in a memory device
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
    • G05D1/0278—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle using satellite positioning signals, e.g. GPS
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/18—Status alarms
    • G08B21/24—Reminder alarms, e.g. anti-loss alarms
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/76—Television signal recording
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W50/08—Interaction between the driver and the control system
    • B60W50/14—Means for informing the driver, warning the driver or prompting a driver intervention
    • B60W2050/143—Alarm means
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00—Input parameters relating to objects

Abstract

The present invention relates to a kind of DAS (Driver Assistant System)s based on collision warning algorithm, belong to computer vision and intelligence auxiliary driving technology field.The system includes: detection and range finder module, acquires the traffic information in vehicle traveling process by camera, is detected, is identified and range measurement to barrier using YOLOv3 model；Anti-collision warning module carries out prediction of collision classification, and the time for calculating the needs that collide provides early warning judgement in time and carries out early warning casting to driver；Locating module: being acquired using traveling-position information of the GPS/IMU integrated navigation to vehicle, and when GPS signal missing, system is automatically converted to IMU and is positioned, and switchs to GPS positioning again when GPS signal is normal；GUI is shown and cloud video backup module, goes forward side by side to rack to identification video flowing, driving status and map software markup information progress real-time display and holds backup.The present invention can be improved the precision of prediction and real-time of DAS (Driver Assistant System).

Description

A kind of DAS (Driver Assistant System) based on collision warning algorithm
Technical field
The invention belongs to computer visions and intelligence auxiliary driving technology field, are related to a kind of based on collision warning algorithm DAS (Driver Assistant System).
Background technique
Recently as the rapid development of China's economy, highway and highway mileage number are constantly bettered a record, Ren Men While enjoyment automobile belt is convenient, traffic accident quantity is also being dramatically increased, in order to can be reduced the hair of traffic accident It is raw, more advanced automotive safety DAS (Driver Assistant System) is developed, reminds driver or the actively partial function of adapter tube automobile, energy in time Enough generations effectively to avoid traffic accident.
In the application study of DAS (Driver Assistant System), it is always wherein difficult for how improving the accuracy rate of collision warning algorithm Point.Many effective natural sentence comprehension models based on deep learning are had proposed in contemporary literature to solve the problems, such as this, Wherein just there is the safety driving assist system based on electroculogram, the special vehicle DAS (Driver Assistant System) based on bus or train route collaboration, be based on The signal lamp intersection auxiliary driving method of image recognition, a kind of greasy weather highway DAS (Driver Assistant System) etc., these models are all Good experiment effect is achieved under different application scenarios.However all there is corresponding at present for these DAS (Driver Assistant System)s Defect, major embodiment in the following areas: one is the existing DAS (Driver Assistant System) shortcoming based under different scenes is specifically touched Hit warning algorithm；The second is center of gravity has all been placed on different scene applications above by most DAS (Driver Assistant System)s, ignore wherein GPS signal obtains the problem of how this is handled with dropout in real time；The third is existing most of DAS (Driver Assistant System)s are ignored GUI is shown in human-computer interaction fluency application.
Although in conclusion being achieved based on the DAS (Driver Assistant System) of computer vision technique in intelligently auxiliary driving field A degree of progress, but interference of the outside environmental elements to detection effect how is reduced, inspection of the lifting system to target Survey precision；How higher verification and measurement ratio and real-time performance greatly promote auxiliary driving performance, promote driving safety performance, reduce Driving accident occurs；How to solve the problems, such as to lose GPS signal when entering tunnel in vehicle travel process；Auxiliary how is expanded to drive Sailing systematic difference range etc. is all the project for needing to continue to go into seriously and practice.
Summary of the invention
In view of this, being used for the purpose of the present invention is to provide a kind of DAS (Driver Assistant System) based on collision warning algorithm The detection accuracy of DAS (Driver Assistant System) is improved, and solves the problems, such as to lose GPS signal when entering tunnel in vehicle travel process.
In order to achieve the above objectives, the invention provides the following technical scheme:
A kind of DAS (Driver Assistant System) based on collision warning algorithm, comprising:
Detection and range finder module: using the detection of the objects such as YOLOv3 model realization vehicle front pedestrian and Che, pass through base The range measurement of all barriers in front is realized in the vehicle vehicle distance measurement method of monocular vision；
Anti-collision warning module: the front of the vehicle location coordinate of GPS detection, car speed and YOLOv3 model inspection is utilized Barrier and vehicle distances, classification prediction collide time of needs；By considering time of driver's reaction and abrupt deceleration vehicle Required time obtains the anticollision time, when prediction collision time is less than anticollision time, the timely early warning of voice broadcast；
Locating module: the traveling-position information of vehicle is acquired in the way of GPS/IMU integrated navigation；By right The extraction of GPS signal shows location information of the user on map, and when GPS signal missing, system will be automatically converted to used Property navigation (IMU) vehicle is positioned, switch to GPS again when GPS signal is normal and positioned；
GUI show with cloud video backup module: using camera acquisition video display area to current vehicle in front of Traffic information carries out the traffic information data of real-time display and navigation system, enterprising in map software to vehicle real time position Rower note, is visualized.
Further, the detection in range finder module, the detection of vehicle front pedestrian and vehicle and ranging specifically include with Lower step:
S11: monocular location algorithm is programmed into YOLOv3 model, trains the weight and other parameters and will be single of model Mesh location algorithm is programmed into model；
S12: carrying high definition monocular cam in vehicle-mounted end, realizes front vehicles and pedestrian etc. using the method for monocular ranging Detection, camera acquires vehicle front Video stream information in real time；
S13: the live video stream of acquisition is uploaded in YOLOv3 model, exports real-time detection of obstacles and ranging is believed Breath, shows for subsequent anti-collision warning and GUI and provides support.
Further, it in the anti-collision warning module, is examined using the front obstacle and vehicle distances and GPS of YOLOv3 detection Vehicle location coordinate, the car speed of survey, classification prediction collides time of needs, and provides early warning in time, specifically The following steps are included:
S21: by the YOLOv3 target based on monocular cam and apart from detection, pedestrian and the vehicle of vehicle front are detected Real-time range, calculate the speed v of target vehicle1, v2；
S22: the classification of collision situation is carried out according to the angular relationship between two vehicles: being setRespectively two cars are respectively The angle of driving direction and two vehicle lines；
IfIndicate that two vehicles are parallel and opposite traveling, it is understood that there may be head-on crash；
IfIndicate that two vehicles are parallel and with mutually traveling, it is understood that there may be rear-end impact；
IfAndWithFor contrary sign, indicate that two vehicles are ipsilateral and opposite traveling, it is understood that there may be side Collision；
S23: corresponding calculation processing is carried out to different classification situations respectively after the completion of vehicle collision classification, and then pre- Survey collision time；If vehicle body width is L, the distance between two vehicles are l.
(1) head-on crash: when detectingWhen, it carries out prediction collision time and the anticollision time compares；
(2) rear-end impact: when detectingWhen, it carries out prediction collision time and the anticollision time compares；
(3) side collision: pre- according to current motion state in the case where two vehicles maintain current vehicle speed and the constant situation of driving direction Measuring car future running track determines that the point of impingement and vehicle collision time, time difference of two vehicles of calculating apart from the point of impingement are sentenced It is disconnected to whether there is risk of collision；
S24: calculating the anticollision time and early warning judgement, system pass through mobile unit real-time reception car status information, When meet straight line collision and side collision precondition when, in real time calculate vehicle collide needs time and anticollision Time is compared, and there are risk of collision then upper progress early warning language promptings.
Further, in the locating module, the measurement equation of system state equation and GPS/IMU integrated navigation is combined High-precision real-time Position Fixing Navigation System is obtained, the GPS signal of target vehicle is acquired, specifically includes the following steps:
S31: use navigational coordinate system for northeast day geographic coordinate system, vehicle GPS/IMU integrated navigation system navigation ginseng Several error model parameters matrix X (t) are as follows:
X (t)=[θE,θN,θU,δL,δλ,δh,δυE,δυN,δυU,ψx,ψy,ψz,▽x,▽y,▽z]T
Wherein, δ L, δ λ, δ h are respectively latitude error, longitude error and height error；δvE,δvN,δvURespectively carrier East orientation, north orientation and sky orientation speed error；θE,θN,θUQuasi- angle is sweared for mathematical platform；ψx,ψy,ψz,▽x,▽y,▽zRespectively gyro Random Constant Drift and the random constant value zero bias of accelerometer；
System state equation are as follows:
Wherein, F (t) is the state-transition matrix of system, and G (t) is the noise variance matrix of system.
S32: the difference of the position of IMU and GPS output and the difference of speed are chosen as measurement, obtains velocity measurement vector Zυ (t) and position measures vector ZS(t) it is respectively as follows:
V in above-mentioned measurement vectorS(t)=[PGE PGN PGU], Vv(t)=[VGE VGN VGU],
HS=[03×6 diag(RM RNcos L 1) 03×6]
Hv=[03×6 diag(1 1 1) 03×6]；
Wherein, subscript I indicates that IMU, subscript G indicate GPS；LGE,λGN,hGURespectively GPS along northeast day direction latitude, Longitude, height, PGE,PGN,PGUIt is GPS along the location error in northeast day direction, VGE,VGN,VGUIt is GPS along the speed in northeast day direction Spend error, RM、RNRespectively meridian circle where carrier and radius of curvature in prime vertical；
S33: co-location measures vector equation and tachometric survey equation obtains the measurement equation Z of GPS/IMU integrated navigation (t) are as follows:
Further, the GUI shows that gui interface is shown and video flowing under networked environment with cloud video backup module Cloud backup specifically includes the following steps:
S41: the camera carried using vehicle-mounted end is acquired the traffic information in front during vehicle driving, to adopting Collect obtained video to pre-process by YOLOv3 deep learning algorithm, obtains the video flowing comprising detection information and ranging information；
S42: the characteristic that can be networked by vehicle-mounted micro computer TX2 carries out cloud backup, detection to the video flowing handled well Recognition detection is carried out to barriers such as vehicle, pedestrians in front of running car with range finder module, and transmits the result to GUI and shows Terminal, the traffic information data of navigation system, vehicle real time position is labeled in map software, is visualized It shows.
The beneficial effects of the present invention are:
(1) present invention is high using the accuracy of YOLOv3 detection model, keeps target detection more acurrate, platform utilization is wider It is general.The present invention chooses the high YOLOv3 model of accuracy and is measured in real time to vehicle, pedestrian, bicycle etc., relies on depth Habit technology chooses suitable convolutional neural networks model, reduces interference of the outside environmental elements to detection effect, lifting system pair The detection accuracy of target.Detection algorithm real-time is high, and the detection accuracy classified on a small quantity is higher, and mark is more easier, and passes through model The technologies such as trimming, compression, are able to achieve higher real-time, and model can be built on wider platform and is used.
(2) present invention utilizes the reasonability of collision warning algorithm, keep early warning rate higher, system performance is more stable.This hair The bright collision prevention of vehicle time warning algorithm rate of false alarm based on speed of operation used is lower, and early warning opportunity is suitable for rate height, basic energy Enough reach the application standard of technical grade.A big branch of the anticollision early warning as DAS (Driver Assistant System) simultaneously, higher verification and measurement ratio Auxiliary driving performance is greatly promoted with real-time performance, is promoted driving safety performance, is reduced driving accident.In addition it can rely on This technology develops such as blind area lane change, bend overspeed warning, traffic light intersection early warning, traffic intersection scheduling planning, pedestrian's early warning Etc. technologies.
(3) present invention utilizes the frontier nature that tunnel GPS loses prediction algorithm, make vehicle location more complete and accurate.This hair Bright aiming at the problem that GPS can lose when vehicle driving is into tunnel, algorithm is adopted since GPS loses the moment using IMU module Collect car speed, acceleration and attitude angle, calculates relative position of the vehicle in tunnel when driving, predict vehicle GPS data.? GPS technology and inertial technology combine available high-precision navigation system, to realize accurately positioning in real time.
(4) present invention employs the conveniences that GUI is shown, keep human-computer interaction more smooth, subsequent development is more convenient.The present invention Using GUI real-time display, convenience is provided for subsequent exploitation and in the application of every profession and trade.GUI is a kind of user interface, is led to It crosses setting GUI to show, support can be provided for later unmanned remote control technology and other extendable functions technologies, simultaneously It is numerous that this technology still can be widely used in all-terrain vehicle automatic Pilot, military target vehicle, beach buggy and unmanned harbour etc. On the remote control of occasion, more considerable economic benefit is brought.
(5) present invention has cloud video flowing backup functionality, can be in the state that 4G communicates good, the view that will locally save Frequency flow data uploads to cloud.When accident occurs for vehicle, the video flowing of cloud backup is as insurance indemnity correlation foundation.
Other advantages, target and feature of the invention will be illustrated in the following description to a certain extent, and And to a certain extent, based on will be apparent to those skilled in the art to investigating hereafter, Huo Zheke To be instructed from the practice of the present invention.Target of the invention and other advantages can be realized by following specification and It obtains.
Detailed description of the invention
To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention is made below in conjunction with attached drawing excellent The detailed description of choosing, in which:
Fig. 1 is the frame diagram of DAS (Driver Assistant System) provided by the invention；
Fig. 2 is the model schematic of the crash classification of collision warning algorithm provided by the invention；
Fig. 3 is GPS/IMU Combinated navigation method implementation flow chart provided by the invention；
Fig. 4 is GUI display terminal effect picture provided in this embodiment.
Specific embodiment
Illustrate embodiments of the present invention below by way of specific specific example, those skilled in the art can be by this specification Other advantages and efficacy of the present invention can be easily understood for disclosed content.The present invention can also pass through in addition different specific realities The mode of applying is embodied or practiced, the various details in this specification can also based on different viewpoints and application, without departing from Various modifications or alterations are carried out under spirit of the invention.It should be noted that diagram provided in following embodiment is only to show Meaning mode illustrates basic conception of the invention, and in the absence of conflict, the feature in following embodiment and embodiment can phase Mutually combination.
As shown in Figure 1, the control method of the DAS (Driver Assistant System) of the present invention based on collision warning algorithm are as follows: detection Module acquires the traffic information in vehicle traveling process by monocular cam, and ranging code realization is write in YOLOv3 model The range measurement of front obstacle；The relative coordinate and speed between vehicle and vehicle (people) are detected, is closed according to the angle between two vehicles System carries out crash classification, to predict collision time, calculate the anticollision time and provide early warning；By system mode side Journey and the measurement equation of GPS/IMU integrated navigation are combined with navigational parameter error model, obtain high-precision navigation system, then It is labeled according to station acquisition information in Baidu map, to realize accurately positioning in real time；Finally mould is shown in GUI It goes forward side by side to rack to identification video flowing, running condition information and Baidu map markup information progress real-time display on block and holds backup.
Specific embodiment is as follows:
Step 1: YOLOv3 model structure cfg file being converted into caffe formatted file, is then introduced into caffe official The available model framework in website, so that error function is restrained, finally obtains optimal training weight by constantly training. Then the weight and other parameters of model are trained, and monocular location algorithm is programmed into model, is taken the photograph by vehicle-mounted end high definition As head acquires realtime graphic and uploads in the YOLOv3 model on TX2 industry plate, so that it may export real-time detection of obstacles With ranging information；
Step 2: carrying high definition monocular cam in vehicle-mounted end, front vehicles and pedestrian are realized using the method for monocular ranging Deng detection, camera acquires vehicle front Video stream information in real time, and video flowing is input in model, before finally obtaining Square barrier classification information, location information and range information.
Step 201: carrying high definition monocular cam in vehicle-mounted end, front vehicles and row are realized using the method for monocular ranging The detection of people etc., camera acquires vehicle front Video stream information in real time, and video flowing is input in model, finally obtains Front obstacle classification information, location information and range information.
Step 202: writing ranging code in YOLOv3 model, realize the range measurement of front obstacle.Test knot Fruit shows that the distance survey relative error of this method less than 3%, has higher detection accuracy.Distance survey result only with The actual range of near-sighted site to video camera in image is related, without demarcating to all camera parameters, to solve It has determined monocular vision distance survey problem.
Step 3: relative coordinate and speed between detection vehicle and vehicle (people) are touched according to the angular relationship between two vehicles Classification is hit, to predict collision time, calculate the anticollision time and provide early warning.Specific steps are as follows:
Step 301: by the YOLOv3 target based on monocular cam and apart from detection, can detecte out vehicle front The real-time range of pedestrian and vehicle calculates the speed v of target vehicle1, v2；
Step 302: the classification of collision situation is carried out according to the angular relationship between two vehicles.IfRespectively two cars The angle of respective driving direction and two vehicle lines；
IfIndicate that two vehicles are parallel and opposite traveling, it is understood that there may be head-on crash；
IfIndicate that two vehicles are parallel and with mutually traveling, it is understood that there may be rear-end impact；
IfAndWithFor contrary sign, indicate that two vehicles are ipsilateral and opposite traveling, it is understood that there may be side Collision；
Step 303: corresponding calculation processing is carried out to different classification situations respectively after the completion of vehicle collision classification, into And predict collision time, as shown in Fig. 2, setting the distance between two vehicles as l (m), two vehicle speeds are respectively v1(km/h), v2(km/ h)。
(1) head-on crash: setting vehicle body width as L (m), when detectingWhen, carry out prediction collision time and peace The full anticollision time compares.
Work as satisfactionAndWhen, prediction collision time is
(2) rear-end impact: setting vehicle body width as L (m), when detectingWhen, carry out prediction collision time and safety The anticollision time compares.
Work as v1< v2AndWhen, prediction collision time isIt is at this time to chase after Tail；
Work as v1> v2AndWhen, prediction collision time isIt is at this time to be chased after Tail；
(3) side collision: pre- according to current motion state in the case where two vehicles maintain current vehicle speed and the constant situation of driving direction Measuring car future running track determines that the point of impingement and vehicle collision time, time difference of two vehicles of calculating apart from the point of impingement are sentenced It is disconnected to whether there is risk of collision.
If setting from vehicle at a distance from the point of impingement as S1, Ta Che is S at a distance from the point of impingement2,
Then whenWhen,
WhenWhen,
WhenWhen,
It is available from vehicle and reaches the time of the point of impingement as t1=3.6 × (S1-2.5)/v1；
Time needed for his vehicle reaches the point of impingement is t2=3.6 × (S2-2.5)/v2；
Time difference Δ t of two vehicles apart from the point of impingement >=| t1-t2|, the time for reaching the point of impingement from vehicle is t1, he touches in vehicle arrival Time needed for hitting a little is t2.If setting Δ t as 2s, when meet Δ t >=| t1-t2| when condition, it will be touched from vehicle and his vehicle arrival The time hit a little is compared with the anticollision time of respective vehicle, judges whether that needing to carry out early warning shows；
Step 304: calculating anticollision time and early warning judgement, the anticollision time is that driver takes measures to keep away just Exempt from the dangerous required shortest time, including the time required to time of driver's reaction and control vehicle (turning to, braking).
The calculation formula of anticollision time is as follows:
Wherein v is the travel speed of vehicle, km/h；G is acceleration of gravity, takes 9.8m/s2；μ is tire-road attachment system Number；The continuous braking time changes with speed and is changed, and g μ is braking maximum deceleration, different coefficient of road adhesion braking effects It is different；t1It is to be connected to pre-warning signal to the time made a response for time of driver's reaction；t2For reaction time of braking device, that is, make Dynamic pedal starts to be depressed to the time t that braking comes into force3For build-up time of braking force, i.e., braking, which comes into force, generates braking to hydraulic press Time；t4For the continuous braking time.
Under normal conditions: TTC≤3s is set as warning against danger；3s < TTC≤5s is set as dangerous tip；TTC > 5s is set It is set to no danger.
Step 4: as shown in figure 3, system state equation and the measurement equation of GPS/IMU integrated navigation and navigational parameter are missed Differential mode type combines, and is acquired to the traveling-position information of vehicle, and when GPS signal missing, it is fixed that system is automatically converted to IMU Position switchs to GPS positioning again when GPS signal is normal to obtain high-precision navigation system, realizes accurately real-time Positioning.Specific steps are as follows:
Step 401: using navigational coordinate system for northeast day geographic coordinate system, vehicle GPS/IMU integrated navigation in this system The error model parameters matrix X (t) of the navigational parameter of system are as follows:
X (t)=[θE,θN,θU,δL,δλ,δh,δυE,δυN,δυU,ψx,ψy,ψz,▽x,▽y,▽z]T
Wherein, δ L, δ λ, δ h are respectively latitude error, longitude error and height error；δvE,δvN,δvURespectively carrier East orientation, north orientation and sky orientation speed error；θE,θN,θUQuasi- angle is sweared for mathematical platform；ψx,ψy,ψz,▽x,▽y,▽zRespectively gyro Random Constant Drift and the random constant value zero bias of accelerometer；
System state equation are as follows:
Wherein, F (t) is the state-transition matrix of system, and G (t) is the noise variance matrix of system.
Step 402: choosing the difference of the position of IMU and GPS output and the difference of speed as measurement, show that velocity measurement is sweared Measure Zυ(t) and position measures vector ZS(t) it is respectively as follows:
V in above-mentioned measurement vectorS(t)=[PGE PGN PGU], Vv(t)=[VGE VGN VGU],
HS=[03×6 diag(RM RNcos L 1) 03×6]
Hv=[03×6 diag(1 1 1) 03×6]；
Wherein, subscript I indicates that IMU, subscript G indicate GPS；LGE,λGN,hGURespectively GPS along northeast day direction latitude, Longitude, height, PGE,PGN,PGUIt is GPS along the location error in northeast day direction, VGE,VGN,VGUIt is GPS along the speed in northeast day direction Spend error, RM、RNRespectively meridian circle where carrier and radius of curvature in prime vertical.
Step 403: co-location measures vector equation and tachometric survey equation obtains the measurement side of GPS/IMU integrated navigation Journey Z (t) is as follows
According to the measurement equation of above system state equation and GPS/IMU integrated navigation, this system is made using STM32F103 For master controller, I2C universal serial bus connects IMU module, and serial ports receives GPS information, acquires GPS data and IMU data in real time, leads to Cross data prediction, attitude algorithm achievees the purpose that integrated navigation, and available current vehicle more it is accurate in real time Position, velocity information and posture information.
Step 5: real-time display being carried out to the traffic information in front of current vehicle using camera acquisition video display area And the traffic information data for combining Baidu's navigation system powerful, vehicle real time position is labeled in Baidu map, is carried out It visualizes, as shown in Figure 4.Specifically includes the following steps:
Step 501: being adopted using traffic information of the camera on automobile data recorder to front during vehicle driving Collection pre-processes the video collected by YOLOv3 deep learning algorithm, obtains including detection information and ranging information Video flowing.The characteristic that can be networked by vehicle-mounted micro computer TX2 simultaneously carries out cloud backup to the video flowing handled well, detects mould Block carries out recognition detection to barriers such as vehicle, pedestrians in front of running car, and transmits the result to GUI display terminal, Camera acquires video display area and carries out real-time display to the traffic information in front of current vehicle.
Step 502: by the collected vehicle position information of GPS module, in conjunction with the powerful road conditions letter of Baidu's navigation system Data are ceased, vehicle actual geographic position is labeled in Baidu map, is visualized.
Step 503:GUI display function key area includes that three keys are respectively " status information ", Reset and " move back Out "." status information " key mainly opens specific traffic information in vehicle travel process (comprising mileage, time, weather etc.) Subpage frame；The Reset button mainly provides the reset function of system；" exiting " button mainly provides the function of logging off.
Finally, it is stated that the above examples are only used to illustrate the technical scheme of the present invention and are not limiting, although referring to compared with Good embodiment describes the invention in detail, those skilled in the art should understand that, it can be to skill of the invention Art scheme is modified or replaced equivalently, and without departing from the objective and range of the technical program, should all be covered in the present invention Scope of the claims in.

Claims (5)

1. a kind of DAS (Driver Assistant System) based on collision warning algorithm, which is characterized in that the system includes:
Detection and range finder module: using the detection of YOLOv3 model realization vehicle front pedestrian and Che, by being based on monocular vision Vehicle vehicle distance measurement method realize the range measurements of all barriers in front；
Anti-collision warning module: the vehicle location coordinate of GPS detection, the preceding object of car speed and YOLOv3 model inspection are utilized Object and vehicle distances, classification prediction collide time of needs；By considering needed for time of driver's reaction and abrupt deceleration vehicle Time obtains the anticollision time, when prediction collision time is less than anticollision time, the timely early warning of voice broadcast；
Locating module: the traveling-position information of vehicle is acquired in the way of GPS/IMU integrated navigation；By believing GPS Number extraction show location information of the user on map, when GPS signal missing when, system will be automatically converted to IMU to vehicle It is positioned, switchs to GPS again when GPS signal is normal and positioned；
GUI is shown and cloud video backup module: using camera acquisition video display area to the road conditions in front of current vehicle Information carries out the traffic information data of real-time display and navigation system, to vehicle real time position in the enterprising rower of map software Note, is visualized.
2. a kind of DAS (Driver Assistant System) based on collision warning algorithm according to claim 1, which is characterized in that the inspection Survey in range finder module, the detection of vehicle front pedestrian and vehicle and ranging specifically includes the following steps:
S11: monocular location algorithm is programmed into YOLOv3 model, is trained the weight of model and other parameters and is surveyed monocular It is programmed into model away from algorithm；
S12: high definition monocular cam is carried in vehicle-mounted end, the inspection of front vehicles and pedestrian is realized using the method for monocular ranging It surveys, camera acquires vehicle front Video stream information in real time；
S13: the live video stream of acquisition is uploaded in YOLOv3 model, exports real-time detection of obstacles and ranging information.
3. a kind of DAS (Driver Assistant System) based on collision warning algorithm according to claim 1, which is characterized in that described to touch It hits in warning module, utilizes the front obstacle and vehicle distances of YOLOv3 detection and vehicle location coordinate, the vehicle of GPS detection Speed, classification prediction collides time of needs, and provides early warning in time, specifically includes the following steps:
S21: by YOLOv3 target based on monocular cam and apart from detection, pedestrian and the vehicle of vehicle front are detected Real-time range calculates the speed v of target vehicle1, v2；
S22: the classification of collision situation is carried out according to the angular relationship between two vehicles: being setRespectively two cars respectively travel The angle in direction and two vehicle lines；
IfIndicate that two vehicles are parallel and opposite traveling, it is understood that there may be head-on crash；
IfIndicate that two vehicles are parallel and with mutually traveling, it is understood that there may be rear-end impact；
IfAndWithFor contrary sign, indicate that two vehicles are ipsilateral and opposite traveling, it is understood that there may be side collision；
S23: carrying out corresponding calculation processing to different classification situations respectively after the completion of vehicle collision classification, and then predicts to touch Hit the time；If vehicle body width is L, the distance between two vehicles are l；
(1) head-on crash: when detectingWhen, it carries out prediction collision time and the anticollision time compares；
(2) rear-end impact: when detectingWhen, it carries out prediction collision time and the anticollision time compares；
(3) side collision: in the case where two vehicles maintain current vehicle speed and the constant situation of driving direction, according to the pre- measuring car of current motion state The following running track, determines the point of impingement and vehicle collision time, calculates time difference of two vehicles apart from the point of impingement and carries out judgement and is It is no that there are risks of collision；
S24: anticollision time and early warning judgement are calculated, system is by mobile unit real-time reception car status information, when full When the precondition of sufficient straight line collision and side collision, calculating vehicle collides the time and anticollision time of needs in real time It is compared, there are risk of collision then upper progress early warning language promptings.
4. a kind of DAS (Driver Assistant System) based on collision warning algorithm according to claim 1, which is characterized in that described fixed In the module of position, the measurement equation of system state equation and GPS/IMU integrated navigation is combined to obtain high-precision real-time positioning Navigation system acquires the GPS signal of target vehicle, specifically includes the following steps:
S31: using navigational coordinate system for northeast day geographic coordinate system, vehicle GPS/IMU integrated navigation system navigational parameter Error model parameters matrix X (t) are as follows:
Wherein, δ L, δ λ, δ h are respectively latitude error, longitude error and height error；δvE,δvN,δvURespectively the east orientation of carrier, North orientation and sky orientation speed error；θE,θN,θUQuasi- angle is sweared for mathematical platform；ψx,ψy,ψz,Respectively Gyro Random is normal Value drift and the random constant value zero bias of accelerometer；
System state equation are as follows:
Wherein, F (t) is the state-transition matrix of system, and G (t) is the noise variance matrix of system；
S32: the difference of the position of IMU and GPS output and the difference of speed are chosen as measurement, obtains velocity measurement vector Zυ(t) and Position measures vector ZS(t) it is respectively as follows:
V in above-mentioned measurement vectorS(t)=[PGE PGN PGU], Vv(t)=[VGE VGN VGU],
HS=[03×6 diag(RM RNcosL 1) 03×6]
Hv=[03×6 diag(1 1 1) 03×6]；
Wherein, subscript I indicates that IMU, subscript G indicate GPS；LGE,λGN,hGURespectively GPS along the latitude in northeast day direction, longitude, Highly, PGE,PGN,PGUIt is GPS along the location error in northeast day direction, VGE,VGN,VGUSpeed for GPS along northeast day direction is missed Difference, RM、RNRespectively meridian circle where carrier and radius of curvature in prime vertical；
S33: co-location measures vector equation and tachometric survey equation obtains the measurement equation Z (t) of GPS/IMU integrated navigation Are as follows:
5. a kind of DAS (Driver Assistant System) based on collision warning algorithm according to claim 1, which is characterized in that described GUI show with cloud video backup module in, gui interface show under networked environment video flowing cloud back up specifically include it is following Step:
S41: the camera carried using vehicle-mounted end is acquired the traffic information in front during vehicle driving, to acquiring The video arrived is pre-processed by YOLOv3 deep learning algorithm, obtains the video flowing comprising detection information and ranging information；
S42: cloud backup is carried out to the video flowing handled well by vehicle-mounted micro computer, detection and range finder module are to running car The barrier in front carries out recognition detection, and transmits the result to GUI display terminal, the traffic information number of navigation system According to being labeled, visualized in map software to vehicle real time position.
CN201910561949.1A 2019-06-26 2019-06-26 A kind of DAS (Driver Assistant System) based on collision warning algorithm Pending CN110276988A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910561949.1A CN110276988A (en) 2019-06-26 2019-06-26 A kind of DAS (Driver Assistant System) based on collision warning algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910561949.1A CN110276988A (en) 2019-06-26 2019-06-26 A kind of DAS (Driver Assistant System) based on collision warning algorithm

Publications (1)

Publication Number Publication Date
CN110276988A true CN110276988A (en) 2019-09-24

Family

ID=67963290

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910561949.1A Pending CN110276988A (en) 2019-06-26 2019-06-26 A kind of DAS (Driver Assistant System) based on collision warning algorithm

Country Status (1)

Country Link
CN (1) CN110276988A (en)

Cited By (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110782702A (en) * 2019-10-25 2020-02-11 安徽三联学院 Vehicle anti-collision early warning system and method
CN110853399A (en) * 2019-10-12 2020-02-28 惠州市德赛西威智能交通技术研究院有限公司 Parking space identification compensation method based on ultrasonic sensor parking space detection system
CN111338345A (en) * 2020-03-03 2020-06-26 上海工程技术大学 Auxiliary intelligent driving device based on visual information
CN111645680A (en) * 2020-04-22 2020-09-11 浙江吉利汽车研究院有限公司 Method, device, terminal and storage medium for determining vehicle trafficability
CN112215198A (en) * 2020-10-28 2021-01-12 郑小云 Self-adaptive man-machine interaction system and method based on big data
CN112349144A (en) * 2020-11-10 2021-02-09 中科海微(北京)科技有限公司 Monocular vision-based vehicle collision early warning method and system
CN112356815A (en) * 2020-12-01 2021-02-12 吉林大学 Pedestrian active collision avoidance system and method based on monocular camera
CN112389448A (en) * 2020-11-23 2021-02-23 重庆邮电大学 Abnormal driving behavior identification method based on vehicle state and driver state
CN112598932A (en) * 2020-12-11 2021-04-02 湖南汽车工程职业学院 Automobile anti-collision early warning model based on vehicle-road cooperation technology
WO2021063006A1 (en) * 2019-09-30 2021-04-08 上海商汤临港智能科技有限公司 Driving early warning method and apparatus, electronic device, and computer storage medium
CN112660122A (en) * 2020-12-29 2021-04-16 山东华宇工学院 Active safe driving early warning method based on 5G low delay and Beidou navigation
CN112758089A (en) * 2019-11-05 2021-05-07 深圳市大富科技股份有限公司 Voice reminding method, advanced driving assistance system and computer storage medium
CN112977369A (en) * 2021-04-19 2021-06-18 天津所托瑞安汽车科技有限公司 Vehicle response method, device, equipment and medium for blind area target object
CN113192363A (en) * 2021-04-29 2021-07-30 杭州电子科技大学 Video data edge calculation method based on artificial intelligence algorithm
CN113192331A (en) * 2021-04-26 2021-07-30 吉林大学 Intelligent early warning system and early warning method for riding safety in internet environment
CN113256981A (en) * 2021-06-09 2021-08-13 天津所托瑞安汽车科技有限公司 Alarm analysis method, device, equipment and medium based on vehicle driving data
CN113299147A (en) * 2021-06-30 2021-08-24 中国汽车工程研究院股份有限公司 Training system and training method based on traffic accident deep investigation
CN113447972A (en) * 2021-06-07 2021-09-28 华东师范大学 Automatic driving GPS deception detection method and system based on vehicle-mounted IMU
CN113442914A (en) * 2020-03-26 2021-09-28 现代摩比斯株式会社 Collision distance estimation apparatus and advanced driver assistance system using the same
CN113624246A (en) * 2020-10-19 2021-11-09 株式会社电装 Vehicle position correction device and vehicle position correction method
CN113837127A (en) * 2021-09-28 2021-12-24 中国科学技术大学先进技术研究院 Map and V2V data fusion model, method, system and medium
CN114170846A (en) * 2021-12-24 2022-03-11 北京百度网讯科技有限公司 Vehicle lane change early warning method, device, equipment and storage medium
CN114199265A (en) * 2021-11-22 2022-03-18 南京邮电大学 Auxiliary navigation system and navigation method based on target detection algorithm
CN114355876A (en) * 2021-11-22 2022-04-15 中国煤炭科工集团太原研究院有限公司 Safety control system and method
CN117746397A (en) * 2024-02-19 2024-03-22 中印云端(深圳)科技有限公司 Visual early warning management system of new energy automobile
CN113624246B (en) * 2020-10-19 2024-05-03 株式会社电装 Vehicle position correction device and vehicle position correction method

Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102328619A (en) * 2010-07-12 2012-01-25 叶春林 Method and structure for expanding front visual field of vehicle
CN105438066A (en) * 2014-08-22 2016-03-30 怡利电子工业股份有限公司 Device for simultaneously displaying navigation information and traffic safety prompting information
CN106056972A (en) * 2016-06-29 2016-10-26 江苏科技大学 Security anti-collision early-warning method based on vehicle driving speed and position information fusion
CN205788804U (en) * 2016-06-23 2016-12-07 北京航天益森风洞工程技术有限公司 Vehicle short-distance wireless communication anticollision on-vehicle terminal device
CN106240458A (en) * 2016-07-22 2016-12-21 浙江零跑科技有限公司 A kind of vehicular frontal impact method for early warning based on vehicle-mounted binocular camera
CN106611513A (en) * 2015-10-27 2017-05-03 株式会社日立制作所 Apparatus and method for providing traffic information
CN106949889A (en) * 2017-03-17 2017-07-14 南京航空航天大学 For the inexpensive MEMS/GPS integrated navigation systems and method of pedestrian navigation
CN107609486A (en) * 2017-08-16 2018-01-19 中国地质大学（武汉） To anti-collision early warning method and system before a kind of vehicle
CN107710751A (en) * 2015-04-17 2018-02-16 三菱电机株式会社 Display control unit, display system, display control method and display control program
CN108099906A (en) * 2017-12-21 2018-06-01 奇瑞汽车股份有限公司 Collision warning systems and automobile on automobile
CN108629328A (en) * 2018-05-15 2018-10-09 安徽师范大学 Intelligent vehicle collision-proof method based on monocular vision and device
WO2019042345A1 (en) * 2017-08-30 2019-03-07 苏州宝时得电动工具有限公司 Self-mobile device and control method for moving path thereof
CN109686125A (en) * 2019-01-11 2019-04-26 重庆邮电大学 A kind of V2X car networking collision prevention of vehicle early warning system based on HMM
WO2019082505A1 (en) * 2017-10-23 2019-05-02 株式会社デンソー Automated driving control device and method for controlling automated driving of vehicle
CN109829403A (en) * 2019-01-22 2019-05-31 淮阴工学院 A kind of vehicle collision avoidance method for early warning and system based on deep learning

Patent Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102328619A (en) * 2010-07-12 2012-01-25 叶春林 Method and structure for expanding front visual field of vehicle
CN105438066A (en) * 2014-08-22 2016-03-30 怡利电子工业股份有限公司 Device for simultaneously displaying navigation information and traffic safety prompting information
CN107710751A (en) * 2015-04-17 2018-02-16 三菱电机株式会社 Display control unit, display system, display control method and display control program
CN106611513A (en) * 2015-10-27 2017-05-03 株式会社日立制作所 Apparatus and method for providing traffic information
CN205788804U (en) * 2016-06-23 2016-12-07 北京航天益森风洞工程技术有限公司 Vehicle short-distance wireless communication anticollision on-vehicle terminal device
CN106056972A (en) * 2016-06-29 2016-10-26 江苏科技大学 Security anti-collision early-warning method based on vehicle driving speed and position information fusion
CN106240458A (en) * 2016-07-22 2016-12-21 浙江零跑科技有限公司 A kind of vehicular frontal impact method for early warning based on vehicle-mounted binocular camera
CN106949889A (en) * 2017-03-17 2017-07-14 南京航空航天大学 For the inexpensive MEMS/GPS integrated navigation systems and method of pedestrian navigation
CN107609486A (en) * 2017-08-16 2018-01-19 中国地质大学（武汉） To anti-collision early warning method and system before a kind of vehicle
WO2019042345A1 (en) * 2017-08-30 2019-03-07 苏州宝时得电动工具有限公司 Self-mobile device and control method for moving path thereof
WO2019082505A1 (en) * 2017-10-23 2019-05-02 株式会社デンソー Automated driving control device and method for controlling automated driving of vehicle
CN108099906A (en) * 2017-12-21 2018-06-01 奇瑞汽车股份有限公司 Collision warning systems and automobile on automobile
CN108629328A (en) * 2018-05-15 2018-10-09 安徽师范大学 Intelligent vehicle collision-proof method based on monocular vision and device
CN109686125A (en) * 2019-01-11 2019-04-26 重庆邮电大学 A kind of V2X car networking collision prevention of vehicle early warning system based on HMM
CN109829403A (en) * 2019-01-22 2019-05-31 淮阴工学院 A kind of vehicle collision avoidance method for early warning and system based on deep learning

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
继峰聊车: "IMU自动驾驶中最后一道安全防线", 《IMU自动驾驶中最后一道安全防线》 *

Cited By (34)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021063006A1 (en) * 2019-09-30 2021-04-08 上海商汤临港智能科技有限公司 Driving early warning method and apparatus, electronic device, and computer storage medium
CN110853399A (en) * 2019-10-12 2020-02-28 惠州市德赛西威智能交通技术研究院有限公司 Parking space identification compensation method based on ultrasonic sensor parking space detection system
CN110782702A (en) * 2019-10-25 2020-02-11 安徽三联学院 Vehicle anti-collision early warning system and method
CN112758089A (en) * 2019-11-05 2021-05-07 深圳市大富科技股份有限公司 Voice reminding method, advanced driving assistance system and computer storage medium
CN111338345A (en) * 2020-03-03 2020-06-26 上海工程技术大学 Auxiliary intelligent driving device based on visual information
CN113442914B (en) * 2020-03-26 2024-04-09 现代摩比斯株式会社 Collision distance estimation apparatus and advanced driver assistance system using the same
CN113442914A (en) * 2020-03-26 2021-09-28 现代摩比斯株式会社 Collision distance estimation apparatus and advanced driver assistance system using the same
US11702068B2 (en) 2020-03-26 2023-07-18 Hyundai Mobis Co., Ltd. Collision distance estimation device and advanced driver assistance system using the same
CN111645680B (en) * 2020-04-22 2022-06-14 浙江吉利汽车研究院有限公司 Method, device, terminal and storage medium for determining vehicle trafficability
CN111645680A (en) * 2020-04-22 2020-09-11 浙江吉利汽车研究院有限公司 Method, device, terminal and storage medium for determining vehicle trafficability
CN113624246B (en) * 2020-10-19 2024-05-03 株式会社电装 Vehicle position correction device and vehicle position correction method
CN113624246A (en) * 2020-10-19 2021-11-09 株式会社电装 Vehicle position correction device and vehicle position correction method
CN112215198A (en) * 2020-10-28 2021-01-12 郑小云 Self-adaptive man-machine interaction system and method based on big data
CN112349144B (en) * 2020-11-10 2022-04-19 中科海微(北京)科技有限公司 Monocular vision-based vehicle collision early warning method and system
CN112349144A (en) * 2020-11-10 2021-02-09 中科海微(北京)科技有限公司 Monocular vision-based vehicle collision early warning method and system
CN112389448B (en) * 2020-11-23 2022-07-01 重庆邮电大学 Abnormal driving behavior identification method based on vehicle state and driver state
CN112389448A (en) * 2020-11-23 2021-02-23 重庆邮电大学 Abnormal driving behavior identification method based on vehicle state and driver state
CN112356815A (en) * 2020-12-01 2021-02-12 吉林大学 Pedestrian active collision avoidance system and method based on monocular camera
CN112598932A (en) * 2020-12-11 2021-04-02 湖南汽车工程职业学院 Automobile anti-collision early warning model based on vehicle-road cooperation technology
CN112660122A (en) * 2020-12-29 2021-04-16 山东华宇工学院 Active safe driving early warning method based on 5G low delay and Beidou navigation
CN112977369B (en) * 2021-04-19 2021-08-10 天津所托瑞安汽车科技有限公司 Vehicle response method, device, equipment and medium for blind area target object
CN112977369A (en) * 2021-04-19 2021-06-18 天津所托瑞安汽车科技有限公司 Vehicle response method, device, equipment and medium for blind area target object
CN113192331A (en) * 2021-04-26 2021-07-30 吉林大学 Intelligent early warning system and early warning method for riding safety in internet environment
CN113192363A (en) * 2021-04-29 2021-07-30 杭州电子科技大学 Video data edge calculation method based on artificial intelligence algorithm
CN113447972A (en) * 2021-06-07 2021-09-28 华东师范大学 Automatic driving GPS deception detection method and system based on vehicle-mounted IMU
CN113447972B (en) * 2021-06-07 2022-09-20 华东师范大学 Automatic driving GPS deception detection method and system based on vehicle-mounted IMU
CN113256981A (en) * 2021-06-09 2021-08-13 天津所托瑞安汽车科技有限公司 Alarm analysis method, device, equipment and medium based on vehicle driving data
CN113256981B (en) * 2021-06-09 2021-09-21 天津所托瑞安汽车科技有限公司 Alarm analysis method, device, equipment and medium based on vehicle driving data
CN113299147A (en) * 2021-06-30 2021-08-24 中国汽车工程研究院股份有限公司 Training system and training method based on traffic accident deep investigation
CN113837127A (en) * 2021-09-28 2021-12-24 中国科学技术大学先进技术研究院 Map and V2V data fusion model, method, system and medium
CN114355876A (en) * 2021-11-22 2022-04-15 中国煤炭科工集团太原研究院有限公司 Safety control system and method
CN114199265A (en) * 2021-11-22 2022-03-18 南京邮电大学 Auxiliary navigation system and navigation method based on target detection algorithm
CN114170846A (en) * 2021-12-24 2022-03-11 北京百度网讯科技有限公司 Vehicle lane change early warning method, device, equipment and storage medium
CN117746397A (en) * 2024-02-19 2024-03-22 中印云端(深圳)科技有限公司 Visual early warning management system of new energy automobile

Similar Documents

Publication Publication Date Title
CN110276988A (en) A kind of DAS (Driver Assistant System) based on collision warning algorithm
CN108394411B (en) Vehicle follows start control device
EP3972882B1 (en) Systems and methods for predicting blind spot incursions
CN106537271B (en) The traffic signals of autonomous vehicle respond
CN102815300B (en) Cruise control apparatus and control method thereof
JP5407764B2 (en) Driving assistance device
CN108534792A (en) Change estimating device, track change presumption method and storage medium in track
CN109062209A (en) A kind of intelligently auxiliary Ride Control System and its control method
CN110392652A (en) Prepare autonomous land vehicle for turning
CN109841088A (en) Vehicle drive assist system and method
CN110001634A (en) Controller of vehicle, control method for vehicle and storage medium
CN106364480A (en) Vehicle control device
CN108437986A (en) Vehicle drive assist system and householder method
CN109389864A (en) A kind of vehicle relative positioning and anti-collision warning method
CN108628324A (en) Unmanned vehicle navigation method, device, equipment based on map vector and storage medium
CN111071249A (en) Unmanned vehicle emergency steering obstacle avoidance method based on obstacle motion prediction
CN106573571A (en) Driver assist system utilizing an inertial sensor
CN110588623A (en) Large automobile safe driving method and system based on neural network
CN110119138A (en) For the method for self-locating of automatic driving vehicle, system and machine readable media
CN114999228B (en) Anti-collision method for automatic driving vehicle in severe weather
CN108466621A (en) effective rolling radius
EP3869341A1 (en) Play-forward planning and control system for an autonomous vehicle
CN104773190A (en) Auxiliary traction control system for train
JPWO2018230640A1 (en) Vehicle control system, vehicle control method, and program
WO2004006207A1 (en) Automatic guide apparatus for public transport

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20190924