CN107103185A - Intelligent vehicle active collision avoidance method based on grey forecasting model - Google Patents

Intelligent vehicle active collision avoidance method based on grey forecasting model Download PDF

Info

Publication number
CN107103185A
CN107103185A CN201710206122.XA CN201710206122A CN107103185A CN 107103185 A CN107103185 A CN 107103185A CN 201710206122 A CN201710206122 A CN 201710206122A CN 107103185 A CN107103185 A CN 107103185A
Authority
CN
China
Prior art keywords
vehicle
collision avoidance
sequence
grey forecasting
vehicles
Prior art date
Application number
CN201710206122.XA
Other languages
Chinese (zh)
Inventor
施凯津
江浩斌
曹福贵
朱畏畏
Original Assignee
江苏大学
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 江苏大学 filed Critical 江苏大学
Priority to CN201710206122.XA priority Critical patent/CN107103185A/en
Publication of CN107103185A publication Critical patent/CN107103185A/en

Links

Classifications

    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16ZINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
    • G16Z99/00Subject matter not provided for in other main groups of this subclass

Abstract

The invention discloses the intelligent vehicle active collision avoidance method based on grey forecasting model, belong to vehicle active safety technologies field.The present invention is using grey forecasting model by carrying out data processing to vehicle GPS original time series, draw the predicted value of cumulative sequence, and then regressive reduction draws the predicted value of original series, the position of main vehicle and surrounding vehicles future time instance is predicted with this, calculate the speed and acceleration of the main vehicle of future time instance and surrounding vehicles, risk of collision vehicle is judged whether, and reminds main vehicle drivers away from dangerous matter sources.Data needed for of the invention are few, and short-term forecast precision is higher, and real-time is good;The complexity of sensor-based collision prevention control system is avoided simultaneously, improves accuracy and robustness.Due to reducing number of sensors, cost reduction improves economy.

Description

Intelligent vehicle active collision avoidance method based on grey forecasting model
Technical field
The invention belongs to vehicle active safety technologies field, and in particular to the intelligent vehicle based on grey forecasting model is actively Collision avoidance method.
Background technology
In recent years, frequent accidents occur, and traffic safety problem becomes a big public hazards of modern society, and China is One of most country of toll on traffic in the world.According to statistics, in all traffic accidents, automobile collision accident is Principal mode, accounts for the 60%~70% of traffic accident.In order to reduce traffic accident, collision avoidance system is arisen at the historic moment, and they lead If by collecting the information of adjacent vehicle on road, determine whether the vehicle of dangerous traveling, and remind driver avoid with It is collided.Typical collision avoidance system is measurement road information of the vehicle by actives such as radar, sound or vision sensors, These information mainly include the relative position and relative velocity between main vehicle and measurement vehicle.These relative motions are believed Breath is combined with the information that main vehicle is obtained from movable sensor, so as to estimate collision time.Automotive active anti-collision system Driver can be reminded to note and intervening measure can be taken before accident occurs, can effectively reduce the generation of automobile collision accident.
, can not be effectively pre- but sensor-based collision prevention control system can only monitor surrounding vehicles transport condition in real time Surrounding vehicles transport condition is surveyed, its economy, accuracy, actual effect and robustness is not ensured and controlled well.
The content of the invention
In order to overcome the above-mentioned deficiencies of the prior art, the invention provides the intelligent vehicle based on grey forecasting model actively Collision avoidance method.
Technical scheme uses following steps:
Step one, the single-chip microcomputer in collision avoidance controller receives vehicle ViGps time sequence be used as original series, represent For
Step 2, calculates the level of time series than λ (k), and judges level can held in covering than whether λ (k) falls, if falling It can hold in covering, then sequenceGray prediction can be carried out as the data of model GM (1,1);Otherwise, it is necessary to sequence Conversion process is done, falling into it can hold in covering;
Original series are done the cumulative sequence of 1 Accumulating generation by step 3
Step 4, orderFor the sequence that adds upClose to equal value sequence, be expressed as
Step 5 is rightSet up the differential equation and corresponding albinism differential equation is respectively: OrderFor parameter vector to be estimated, the differential equation is write as matrix form, by Least Square Method U, willBring albinism differential equation into and draw k+1 moment vehicles viThe predicted value of position, wherein k=2, n, a for development Coefficient, b is grey actuating quantity;
Step 6, according to k+1 moment vehicles V in step 5iThe predicted value of position, surrounding vehicles carry out instantaneous velocity and wink Brief acceleration estimates that the gps time sequence of the speed estimated and collection is issued main vehicle by surrounding vehicles, and main vehicle can be pre- Survey position and the speed of surrounding vehicles and itself subsequent time, thus calculate main vehicle and surrounding vehicles relative distance L and Relative velocity v, whether be hazardous vehicles, remind main vehicle drivers away from dangerous matter sources if judging surrounding vehicles.
Further, fall into a trap formula of the level than λ (k) of evaluation time sequence of the step 2 is
Further, holding in the step 2 is covered as
Further, the matrix form of the differential equation is in the step 5
Further, it is by Least Square Method U formula in the step 5:Its In
Further, k+1 moment vehicles v in the step 5iThe predicted value of position is:
Beneficial effects of the present invention are:The present invention predicts main vehicle and surrounding vehicles future time instance using grey forecasting model Position, the speed and acceleration of the main vehicle of future time instance and surrounding vehicles are calculated with this, risk of collision is judged whether Vehicle, and main vehicle drivers are reminded away from dangerous matter sources.Data needed for of the invention are few, and short-term forecast precision is higher, and real-time is good;Together When avoid the complexity of sensor-based collision prevention control system, improve accuracy and robustness.Due to reducing sensing Device quantity, cost reduction, improves economy.
Brief description of the drawings
Fig. 1 is the structured flowchart of the intelligent vehicle active collision avoidance method of the invention based on grey forecasting model;
Fig. 2 carries out the flow chart of gray prediction for the present invention;
Fig. 3 vehicles operating range schematic diagram within each sampling period;
Fig. 4 is different collision avoidance modes and friction speed interval and linear relationship chart of the safety time away from TTC values.
Embodiment
It is below in conjunction with accompanying drawing and specifically real to make the purpose of the present invention, technical scheme and effect clearer, clear and definite Applying example, the invention will be further described.
Gray prediction is to carry out estimation prediction to the development and change rule of system action feature, while can also be special to behavior Estimation calculating is carried out at the time of the abnormal conditions levied occur, and feelings are distributed to the future event of event occurs in specific time zone The method that condition makes research.Gray model is to be changed into randomness by production using Discrete Stochastic number to be significantly weakened and relatively have The generation number of rule, it is established that the model of differential equation form its change procedure is described.The model of gray system theory has Many kinds, but single order single argument GM (1,1) model in gray system theory is mainly used in gray prediction.
As shown in Figure 1, 2, the intelligent vehicle active collision avoidance method based on grey forecasting model, including step:
1) single-chip microcomputer in collision avoidance controller receives vehicle ViGps time sequence as original series, be expressed as:
Wherein:I=1,2,3n;
2) level of time series is calculated than λ (k):
If the level of the time series calculated all falls that covering can be being held than λ (k)It is interior, then original series Gray prediction can be carried out as the data of model GM (1,1);Otherwise, it is necessary to original seriesConversion process is done, makes it Falling into can hold in covering;Appropriate constant c is taken to do translation transformation so that level falls than λ (k) can hold coveringIt is interior:
Wherein:K=1,2n;
Original seriesThe new sequence of generation:
3) to original seriesDo 1 cumulative sequence of cumulative (AGO) generation:
Wherein:
4) makeFor the sequence that adds upClose to average (MEAN) sequence:
Wherein:
5) it is rightSet up the differential equation:
Albinism differential equation is accordingly:
Wherein:A is development coefficient, and b is grey actuating quantity;
OrderFor parameter vector to be estimated, then the matrix form of equation (9) is:
OrderThen by Least Square Method U:
WillBring equation (10) into, obtain:
Therefore, k+1 moment vehicles ViThe predicted value of position is:
6) according to step 5) in k+1 moment vehicles ViThe predicted value of position, estimation whether the event of dangerous traveling will Occur;
Assuming that vehicle ViCollecting GPS sequences from the beginning isSampling period For T, then vehicle V in each cycleiOperating range is the difference DELTA L of two gps time sequence of pointsm, as shown in Figure 3;Assuming that vehicle ViAcceleration and initial velocity within the ith sample cycle are respectively ai、vi, then have:
According to the formula (21) derived and (22), main vehicle and surrounding vehicles progress instantaneous velocity and instantaneous acceleration are estimated The gps time sequence of the speed estimated and collection is issued main vehicle by meter, surrounding vehicles, and main vehicle can predict surrounding vehicles And position and the speed of itself subsequent time;It therefore, it can calculate main vehicle and the phase of surrounding vehicles by formula (21), (22) Adjust the distance L and relative velocity v, when there is car speed to be more than safety time away from TTC more than road speeds limitation, or L/v, TTC Refer to two cars travelled in the same direction on same path and kept for time of the present speed required for collision occurs, as shown in figure 4, judging It is hazardous vehicles, and reminds driver away from dangerous matter sources.
TTC=Xr/Vr (23)
Wherein:XrFor the relative distance in two workshops, VrFor the relative velocity in two workshops.
It is described above that the present invention is briefly described, not by above-mentioned working range limit value, as long as taking the present invention Thinking and method of work carry out simple modification and apply to other equipment, or make and changing in the case where not changing central scope principle of the present invention Enter and retouch etc. behavior, within protection scope of the present invention.

Claims (6)

1. the intelligent vehicle active collision avoidance method based on grey forecasting model, it is characterised in that comprise the following steps:
Step one, the single-chip microcomputer in collision avoidance controller receives vehicle ViGps time sequence as original series, be expressed as
Step 2, calculates the level of time series than λ (k), and judges level can held in covering than whether λ (k) falls, if falling to hold Covering is interior, then sequenceGray prediction can be carried out as the data of model GM (1,1);Otherwise, it is necessary to sequenceDo and become Processing is changed, falling into it can hold in covering;
Original series are done the cumulative sequence of 1 Accumulating generation by step 3
Step 4, orderFor the sequence that adds upClose to equal value sequence, be expressed as
Step 5 is rightSet up the differential equation and corresponding albinism differential equation is respectively: OrderFor parameter vector to be estimated, the differential equation is write as matrix form, by Least Square Method U, willBring albinism differential equation into and draw k+1 moment vehicles viThe predicted value of position, wherein k=2 ..., n, a for development be Number, b is grey actuating quantity;
Step 6, according to k+1 moment vehicles V in step 5iThe predicted value of position, surrounding vehicles carry out instantaneous velocity and instantaneously added The gps time sequence of the speed estimated and collection is issued main vehicle by velocity estimation, surrounding vehicles, and main vehicle can predict week Vehicle and position and the speed of itself subsequent time are enclosed, so as to calculate the relative distance L of main vehicle and surrounding vehicles and relative Speed v, whether be hazardous vehicles, remind main vehicle drivers away from dangerous matter sources if judging surrounding vehicles.
2. the intelligent vehicle active collision avoidance method according to claim 1 based on grey forecasting model, it is characterised in that institute Stating fall into a trap formula of the level than λ (k) of evaluation time sequence of step 2 is
3. the intelligent vehicle active collision avoidance method according to claim 1 based on grey forecasting model, it is characterised in that institute State and be covered as holding in step 2
4. the intelligent vehicle active collision avoidance method according to claim 1 based on grey forecasting model, it is characterised in that institute The matrix form for stating the differential equation in step 5 is
5. the intelligent vehicle active collision avoidance method according to claim 4 based on grey forecasting model, it is characterised in that institute State in step 5 by Least Square Method U formula and be:Wherein
6. the intelligent vehicle active collision avoidance method according to claim 1 based on grey forecasting model, it is characterised in that institute State k+1 moment vehicles v in step 5iThe predicted value of position is:
CN201710206122.XA 2017-03-31 2017-03-31 Intelligent vehicle active collision avoidance method based on grey forecasting model CN107103185A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710206122.XA CN107103185A (en) 2017-03-31 2017-03-31 Intelligent vehicle active collision avoidance method based on grey forecasting model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710206122.XA CN107103185A (en) 2017-03-31 2017-03-31 Intelligent vehicle active collision avoidance method based on grey forecasting model

Publications (1)

Publication Number Publication Date
CN107103185A true CN107103185A (en) 2017-08-29

Family

ID=59675370

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710206122.XA CN107103185A (en) 2017-03-31 2017-03-31 Intelligent vehicle active collision avoidance method based on grey forecasting model

Country Status (1)

Country Link
CN (1) CN107103185A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109164799A (en) * 2018-07-24 2019-01-08 江苏大学 A kind of driving Active collision avoidance system and method
WO2019213980A1 (en) * 2018-05-08 2019-11-14 清华大学 Intelligent vehicle safety decision-making method employing driving safety field

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102069800A (en) * 2010-12-28 2011-05-25 奇瑞汽车股份有限公司 Vehicle running control method and control device
CN103465907A (en) * 2013-08-27 2013-12-25 奇瑞汽车股份有限公司 Automotive collision avoidance and method
CN104134371A (en) * 2014-07-09 2014-11-05 奇瑞汽车股份有限公司 Method and device for early warning of vehicle collision
CN104494550A (en) * 2014-12-15 2015-04-08 江苏大学 Active anticollision device of vehicle and calculating method of anticollision distance
CN104715158A (en) * 2015-03-30 2015-06-17 北京航空航天大学 Empirical mode decomposition endpoint effect inhibition method based on grey GM (1,1) forecasting model
CN105270398A (en) * 2015-10-21 2016-01-27 北京新能源汽车股份有限公司 Vehicle distance control method and device for electric vehicle and electric vehicle
CN105679093A (en) * 2016-02-23 2016-06-15 江苏大学 Multi-vehicle coordination collision avoidance system and method based on vehicle-vehicle communication

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102069800A (en) * 2010-12-28 2011-05-25 奇瑞汽车股份有限公司 Vehicle running control method and control device
CN103465907A (en) * 2013-08-27 2013-12-25 奇瑞汽车股份有限公司 Automotive collision avoidance and method
CN104134371A (en) * 2014-07-09 2014-11-05 奇瑞汽车股份有限公司 Method and device for early warning of vehicle collision
CN104494550A (en) * 2014-12-15 2015-04-08 江苏大学 Active anticollision device of vehicle and calculating method of anticollision distance
CN104715158A (en) * 2015-03-30 2015-06-17 北京航空航天大学 Empirical mode decomposition endpoint effect inhibition method based on grey GM (1,1) forecasting model
CN105270398A (en) * 2015-10-21 2016-01-27 北京新能源汽车股份有限公司 Vehicle distance control method and device for electric vehicle and electric vehicle
CN105679093A (en) * 2016-02-23 2016-06-15 江苏大学 Multi-vehicle coordination collision avoidance system and method based on vehicle-vehicle communication

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
王洁 等: "基于最小二乘法的GM(1,1)模型在我国蔬菜产量预测中的应用研究", 《数学理论与应用》 *
袁基炜 等: "一种基于灰色预测模型GM( 1, 1) 的运动车辆跟踪方法", 《控制与决策》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019213980A1 (en) * 2018-05-08 2019-11-14 清华大学 Intelligent vehicle safety decision-making method employing driving safety field
CN109164799A (en) * 2018-07-24 2019-01-08 江苏大学 A kind of driving Active collision avoidance system and method

Similar Documents

Publication Publication Date Title
Mahmud et al. Application of proximal surrogate indicators for safety evaluation: A review of recent developments and research needs
CN107257756B (en) Techniques for assisting a vehicle in situations of varying road conditions
Liu et al. Impact of road gradient on energy consumption of electric vehicles
CN106240458B (en) A kind of vehicular frontal impact method for early warning based on vehicle-mounted binocular camera
CN104164829B (en) Detection method of road-surface evenness and intelligent information of road surface real-time monitoring system based on mobile terminal
JP2018509705A (en) How to control the distance between vehicles
CN104809879B (en) Expressway road traffic state estimation method based on dynamic Bayesian network
Thiemann et al. Estimating acceleration and lane-changing dynamics from next generation simulation trajectory data
CN102666240B (en) Collision prevention device
Nouveliere Experimental vehicle longitudinal control using a second order sliding mode technique
Liebner et al. Velocity-based driver intent inference at urban intersections in the presence of preceding vehicles
Risack et al. A video-based lane keeping assistant
US7561054B2 (en) System and method for displaying a driving profile
CN102320280B (en) Automatic alarm method for preventing front crash of vehicles at turning
US8947218B2 (en) Driving support device
EP2710572B1 (en) Vehicle data analysis method and vehicle data analysis system
CN102745194B (en) Self-adaption alarming method for preventing tailgating with front car on expressway
CN107672589B (en) A kind of track of vehicle real-time predicting method and device based on GPR Detection Data
Liu et al. A virtual vehicle probe model for time-dependent travel time estimation on signalized arterials
CN103935364B (en) Automobile actively anti-collision early warning system based on millimetre-wave radar
US7739087B2 (en) Method and device for analying collision driving data
CN103164962B (en) Mountain road sharp turn section real-time vehicle speed early warning method
CN102795225B (en) Method for detecting disturbance state of driver by utilizing driver-side longitudinal control model
EP2354762B1 (en) Navigation system and method for determining parameters in a navigation system
CN103395419B (en) Based on vehicle platoon drive-control system and the control method thereof of safe spacing strategy

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: 20170829