CN106777776A - A kind of vehicle lane-changing decision-making technique based on supporting vector machine model - Google Patents
A kind of vehicle lane-changing decision-making technique based on supporting vector machine model Download PDFInfo
- Publication number
- CN106777776A CN106777776A CN201710015537.9A CN201710015537A CN106777776A CN 106777776 A CN106777776 A CN 106777776A CN 201710015537 A CN201710015537 A CN 201710015537A CN 106777776 A CN106777776 A CN 106777776A
- Authority
- CN
- China
- Prior art keywords
- lane
- decision
- making
- vehicle
- change
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000012360 testing method Methods 0.000 claims abstract description 12
- 238000001514 detection method Methods 0.000 claims abstract description 5
- 238000012706 support-vector machine Methods 0.000 claims description 7
- 241001269238 Data Species 0.000 claims description 6
- 238000004088 simulation Methods 0.000 claims description 4
- 230000001133 acceleration Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 2
- 238000012795 verification Methods 0.000 claims 2
- 235000013399 edible fruits Nutrition 0.000 claims 1
- 238000010276 construction Methods 0.000 abstract description 2
- 230000000694 effects Effects 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 8
- 230000008859 change Effects 0.000 description 4
- 206010039203 Road traffic accident Diseases 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Abstract
The invention discloses a kind of vehicle lane-changing decision-making technique based on supporting vector machine model, the method uses the following steps:First pass through related data when sensor reads vehicle lane-changing in real time;Then the data for obtaining are imported in the vehicle lane-changing decision-making module based on supporting vector machine model, the construction step of the module mainly includes training and the selection of test sample, the treatment of sample data, the training of model and detection;Decision-making judged result during vehicle lane-changing can be formed finally by decision-making module, such as decision-making judged result then sends alarm and reminding driver and be unable to lane-change in real time to be unable to lane-change.The present invention reduces warning algorithm complexity, the negative effect of the excessively multipair judged result of decision-making judgment rule, decision-making judges when improve vehicle lane-changing the degree of accuracy and reliability reduce rate of false alarm.
Description
Technical field
The present invention relates to vehicle driving security fields, more particularly to a kind of replacing vehicle based on supporting vector machine model
Road decision-making technique.
Background technology
Vehicle lane-changing error is always one of the major reason for causing traffic accident to occur, and statistics shows, every year by
A large amount of traffic accidents are result in incorrect lane changing driving behavior.In various driving behaviors, lane-change is driven because being related to
Change lane and risk factor is higher, therefore successfully lane-change behavior can not only support vehicles drive security, moreover it is possible to really
Protect the patency of traffic flow.
During lane-change, driver is general only to be obtained for information about by left and right rearview mirror, is done after comprehensive descision
Go out whether the decision of lane-change, yet with information imperfect and human factor the reason for, driver can not possibly consider from car with
The distance between front truck, the front truck of rear car and target track, rear car of original lane and relative velocity, thus driver analysis change
Error is susceptible to during road possibility.On the other hand, the blind area of rearview mirror also easily causes larger safety to be drawn
Suffer from.
At present, prior art has been used to assist the vehicle lane-changing danger early warning system of changing Lane.It is common at present to change
Road early warning system is divided into two classes, the first kind mainly for rearview mirror vision dead zone problem, by using ultrasonic sensor pair
Vehicle from car is lateral, in the close region of rear is monitored;Equations of The Second Kind system exists mainly for lane-change target track rear
At a high speed close to the situation of vehicle, by using range radar to the relative distance during lane-change from car and other vehicles, relative
Speed carries out monitor in real time, and the degree of risk of collision accident is triggered during analysis lane-change, right in the case of degree of risk is higher
Driver carries out early warning.However, device needed for existing vehicle lane-changing danger early warning system operation is more, warning algorithm is complicated,
Reliability is not high.Therefore, warning algorithm is simple when needing a kind of vehicle lane-changing at present, rate of false alarm turns low vehicle lane-changing decision-making party
Method.
The content of the invention
It is an object of the invention to provide a kind of vehicle lane-changing decision-making technique based on supporting vector machine model, using support
Vector machine model judges for driver provides decision-making in vehicle lane-changing, and lane-change is dangerous occur when send alarm in time and carry
Wake up.Methods described effectively can judge for driver provides correct reliable decision-making when vehicle lane-changing is carried out, so as to reduce
There is the possibility of traffic accident during lane-change.
To achieve the above object, the present invention is adopted the following technical scheme that:
A kind of vehicle lane-changing decision-making technique based on decision-tree model, comprises the following steps:
Step 1 obtains sample data, lane-change vehicle and front truck specifically on original lane by millimeter speed-measuring radar sensor
Apart from GPO, lane-change vehicle and front truck apart from G on target trackPT, lane-change vehicle and rear car apart from G on target trackFT、
Target track car of going forward is obtained in real time with five sample datas of speed V apart from D, lane-change vehicle of rear car;
Step 2 builds the vehicle lane-changing decision-making module based on SVMs, respectively by training and the collection of test sample with
Treatment, the foundation of the model of the SVMs based on Matlab platforms, model accuracy verify these links to build car
Lane-change decision-making module;
Step 3 decision-making judge, by obtain five sample datas imported into real time in vehicle lane-changing decision-making module, obtain whether
The decision-making for carrying out lane-change judges.
Further, in the step 2, the traffic data in training and the selection of test sample data and processing links be by
Next Generation Simulation (NGSIM) are provided, and the track data in NGSIM data sets provides the vertical seat of each car
Mark, abscissa, speed, acceleration and front and rear two cars interval time, the sampling interval are 0.1 second;And data set is divided, institute
A data set part is stated for model training, another part is used to test.
Advantages of the present invention mainly has:
1st, the method frame of the vehicle lane-changing decision-making based on supporting vector machine model has been built, has been that the vehicle of multitude of different ways is strong
Lane-change Analysis of Policy Making processed is laid a good foundation;
2nd, warning algorithm complexity, the influence of the excessively multipair judged result of decision-making judgment rule are reduced, vehicle compulsory lane-change is improve
When decision-making judge the degree of accuracy and reliability.
Brief description of the drawings
Fig. 1 is the expression figure of this method scope of application and involved relevant parameter.
Fig. 2 is the vehicle lane-changing decision-making technique schematic flow sheet based on supporting vector machine model.
Specific embodiment
Further specific embodiment of the invention is illustrated below in conjunction with the accompanying drawings:
Step 1 obtains sample data, lane-change vehicle and front truck specifically on original lane by millimeter speed-measuring radar sensor
Apart from GPO, lane-change vehicle and front truck apart from G on target trackPT, lane-change vehicle and rear car apart from G on target trackFT、
Target track car of going forward is obtained in real time with five sample datas of speed V apart from D, lane-change vehicle of rear car;
Step 2 builds the vehicle lane-changing decision-making module based on SVMs, respectively by training and the collection of test sample
Verify these links to build with treatment, the foundation of the model of the SVMs based on Matlab platforms, model accuracy
Vehicle lane-changing decision-making module;
Step 3 decision-making judge, by obtain five sample datas imported into real time in vehicle lane-changing decision-making module, obtain whether
The decision-making for carrying out lane-change judges.
Above three step is illustrated as follows below.
As shown in Figure 1, the lane-change behavior on highway is divided into two classes according to driver's motivation is unusual:Force
Property lane-change and initiative lane-change.Mandatory lane-change refers to the plan of travel according to traffic rules and driver, in order to complete it just
The lane-change behavior carried out when often traveling purpose driver has to;Initiative lane-change refers to driver in order to avoid collision, speed
Slow down, road bottleneck etc. and the lane-change behavior that carries out, its basic goal is that, in order to reduce delay, acquisition speed advantage, increase is driven
The comfort sailed.Even if initiative lane-change is that changing Lane can be on former track yet for vehicle with the main distinction of mandatory lane-change
Its traveling task is completed, it is that driver implements to reach more free, more preferable transport condition.This method
The scope of application is initiative lane-change.
Vehicle lane-changing behavior is related to the interaction between many cars, and the vehicle of participation will be to the lane-change decision-making of driver
Influence is produced, is found through research, during vehicle lane-changing, had to the maximum factor of lane-change Decision Making Effect:Original lane
Upper lane-change vehicle is with front truck apart from GPO, lane-change vehicle and front truck apart from G on target trackPT, lane-change vehicle on target track
With rear car apart from GFT, target track go forward car and rear car apart from D, the speed V of lane-change vehicle, lane-change influence factor data
Symbol implication is as shown in table 1.
The symbol implication of the lane-change influence factor data of table 1
The structure of the vehicle lane-changing decision model based on supporting vector machine model, is divided into three links, wherein having described in step 2
Body step is as follows.
The first step:Training and the selection of test sample
Training in this method is provided with test sample traffic data by Next Generation Simulation (NGSIM),
For the study and checking of decision-tree model.NGSIM be by Bureau of Public Road for the purpose of studying microscopic traffic simulation and
The project of initiation, data are taken pictures by using many altitude cameras and obtained, frequency of taking pictures be every 0.1 second once, obtain
Arrive each car and be spaced the precise trajectory data of 0.1 second, then by aiming at data processing software that NGSIM develops from vehicle rail
Mark extracting data correlated variables:Lateral direction of car lengthwise position, speed, acceleration, space headway, time headway, vehicle, width
Degree, length etc..
Used by this method be NGSIM on June 15th, 2005 on the expressway of Los Angeles,U.S 101, positioned at Ventura
The vehicle-mounted data gathered on about 650 meters of linear road and two ring roads of Cahuenga mouthful between, in order to check SVM models when
Between on versatility, 7 are chosen herein:50-8:05 and 8:05-8:The data set of 20 two time periods is analyzed, a work
It is training set, one collects as detection.
Second step:The treatment of sample data
Research vehicle running orbit data obtain 5281 group observationses altogether, are changed including 279 groups of lane-change data and 5002 groups are non-
Track data.Data set is further divided, 8 are will occur in:05-8:The data set of 20 time periods is used for model training, hair
Life is 7:50-8:The data set of 05 time period is used for model measurement.
In the data set of instantiation, the data variance between different dimensions parameter is larger, easily causes part number
According to being submerged.Meanwhile, data variation scope conference makes calculating process complex, causes the time of training pattern more long, can be right
Model Identification precision is adversely affected.Therefore in the processing procedure of sample data, using data normalization method reduction number
According to complexity.
3rd step:The training of model and detection
This method is based on the structure that Matlab platforms are supported vector machine model, and Matlab has the svmtrain functions for carrying to enter
Row model construction of SVM, training set sample is imported, and is trained using system default parameter, is used after obtaining model
Svmclassify function pair test set samples are detected that the accuracy situation of model is as shown in table 2.
The accuracy situation of the supporting vector machine model of table 2
Result shows that 115 samples of lane-change in test set, supporting vector machine model gives lane-change to wherein 95 samples
Prediction, accuracy 82.6%;2591 samples without lane-change in test set, model gives to wherein 1854 samples and does not change
The prediction in road, accuracy 71.6%.Can be obtained when showing the model for vehicle lane-changing behaviour decision making on a highway compared with
Good predicts the outcome.
The above, is not intended to limit the scope of the present invention, it will be appreciated that the present invention is not limited to be retouched here
The implementation stated, the purpose of these implementations description is to help the those of skill in the art practice present invention.It is any
Those of skill in the art be easy to be further improved without departing from the spirit and scope of the present invention with it is complete
It is apt to, therefore the present invention is only limited by the content and scope of the claims in the present invention, its intention covers all being included in by institute
Attached claim is limited in spirit and scope of the invention alternative and equivalent.
Claims (5)
1. a kind of vehicle lane-changing decision-making technique based on supporting vector machine model, it is characterised in that:Including supporting vector machine model
Training process and the process of lane-change identification is carried out using supporting vector machine model;Comprise the following steps:
Step 1:Sample data, lane-change vehicle and front truck specifically on original lane are obtained by millimeter speed-measuring radar sensor
Apart from GPO, lane-change vehicle and front truck apart from G on target trackPT, lane-change vehicle and rear car apart from G on target trackFT、
Target track car of going forward is obtained in real time with five sample datas of speed V apart from D, lane-change vehicle of rear car;
Step 2:Build the vehicle lane-changing decision-making module based on SVMs, respectively by training and the collection of test sample with
Treatment, the foundation of the model of the SVMs based on Matlab platforms, model accuracy verify these links to build car
Lane-change decision-making module;
Step 3:Decision-making judgement, five sample datas for obtaining are imported into vehicle lane-changing decision-making module in real time, are
The no decision-making for carrying out lane-change judges.
2. the vehicle lane-changing decision-making technique based on supporting vector machine model according to claim 1, it is characterised in that:It is described
In step 2, training is by Next Generation with the selection of test sample data and the traffic data in processing links
Simulation (NGSIM) is provided, track data in NGSIM data sets provide the ordinate of each car, abscissa, speed,
Acceleration and front and rear two cars interval time, the sampling interval are 0.1 second;And data set is divided.
3. the vehicle lane-changing decision-making technique based on supporting vector machine model according to claim 1, it is characterised in that:It is described
In step 2, used be NGSIM on June 15th, 2005 on the expressway of Los Angeles,U.S 101, positioned at Ventura and
The vehicle-mounted data gathered on about 650 meters of linear road between Cahuenga two ring roads mouthful, in order to check SVM models in the time
On versatility, 7 are chosen herein:50-8:05 and 8:05-8:The data set of 20 two time periods is analyzed, a conduct
Training set, one collects as detection.
4. the vehicle lane-changing decision-making technique based on supporting vector machine model according to claim 1, it is characterised in that:It is described
In step 2, the structure of vector machine model is supported based on Matlab platforms, training set sample is trained and obtains model.
5. the vehicle lane-changing decision-making technique based on supporting vector machine model according to claim 1, it is characterised in that:It is described
The precise verification link of supporting vector machine model in step 2, is identified, and identification is tied using model to detection collection sample
Fruit is contrasted with actual lane-change situation, with the validity of the Accuracy Verification model of identification.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710015537.9A CN106777776A (en) | 2017-01-10 | 2017-01-10 | A kind of vehicle lane-changing decision-making technique based on supporting vector machine model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710015537.9A CN106777776A (en) | 2017-01-10 | 2017-01-10 | A kind of vehicle lane-changing decision-making technique based on supporting vector machine model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106777776A true CN106777776A (en) | 2017-05-31 |
Family
ID=58948732
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710015537.9A Pending CN106777776A (en) | 2017-01-10 | 2017-01-10 | A kind of vehicle lane-changing decision-making technique based on supporting vector machine model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106777776A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108802769A (en) * | 2018-05-30 | 2018-11-13 | 千寻位置网络有限公司 | Detection method and device of the GNSS terminal on overhead or under overhead |
CN110609541A (en) * | 2018-06-15 | 2019-12-24 | 本田技研工业株式会社 | Vehicle control device, vehicle control method, and storage medium |
CN110619340A (en) * | 2018-06-19 | 2019-12-27 | 广州汽车集团股份有限公司 | Method for generating lane change rule of automatic driving automobile |
TWI690440B (en) * | 2018-10-17 | 2020-04-11 | 財團法人車輛研究測試中心 | Intelligent driving method for passing intersections based on support vector machine and intelligent driving system thereof |
CN111325230A (en) * | 2018-12-17 | 2020-06-23 | 上海汽车集团股份有限公司 | Online learning method and online learning device of vehicle lane change decision model |
CN111882870A (en) * | 2020-07-14 | 2020-11-03 | 吉林大学 | Quantification method of road traffic environment |
CN112560782A (en) * | 2020-12-26 | 2021-03-26 | 浙江天行健智能科技有限公司 | Vehicle lane changing behavior identification method based on random forest algorithm |
CN112793576A (en) * | 2021-01-26 | 2021-05-14 | 北京理工大学 | Lane change decision method and system based on rule and machine learning fusion |
CN112907987A (en) * | 2021-01-19 | 2021-06-04 | 吉林大学 | Multi-lane express way exit ramp shunting area intelligent motorcade lane change guiding method and system |
TWI778428B (en) * | 2020-08-26 | 2022-09-21 | 中國商宸展光電(廈門)股份有限公司 | Method, device and system for detecting memory installation state |
CN115731261A (en) * | 2021-08-27 | 2023-03-03 | 河北省交通规划设计研究院有限公司 | Method and system for identifying lane changing behavior of vehicle based on expressway radar data |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103996287A (en) * | 2014-05-26 | 2014-08-20 | 江苏大学 | Vehicle forced lane changing decision-making method based on decision-making tree model |
-
2017
- 2017-01-10 CN CN201710015537.9A patent/CN106777776A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103996287A (en) * | 2014-05-26 | 2014-08-20 | 江苏大学 | Vehicle forced lane changing decision-making method based on decision-making tree model |
Non-Patent Citations (1)
Title |
---|
JIANQIANG NIE等: "Modeling of Decision-Making Behavior for Discretionary Lane-Changing Execution" * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108802769A (en) * | 2018-05-30 | 2018-11-13 | 千寻位置网络有限公司 | Detection method and device of the GNSS terminal on overhead or under overhead |
CN110609541A (en) * | 2018-06-15 | 2019-12-24 | 本田技研工业株式会社 | Vehicle control device, vehicle control method, and storage medium |
CN110609541B (en) * | 2018-06-15 | 2022-09-27 | 本田技研工业株式会社 | Vehicle control device, vehicle control method, and storage medium |
CN110619340B (en) * | 2018-06-19 | 2022-09-16 | 广州汽车集团股份有限公司 | Method for generating lane change rule of automatic driving automobile |
CN110619340A (en) * | 2018-06-19 | 2019-12-27 | 广州汽车集团股份有限公司 | Method for generating lane change rule of automatic driving automobile |
TWI690440B (en) * | 2018-10-17 | 2020-04-11 | 財團法人車輛研究測試中心 | Intelligent driving method for passing intersections based on support vector machine and intelligent driving system thereof |
CN111325230A (en) * | 2018-12-17 | 2020-06-23 | 上海汽车集团股份有限公司 | Online learning method and online learning device of vehicle lane change decision model |
CN111325230B (en) * | 2018-12-17 | 2023-09-12 | 上海汽车集团股份有限公司 | Online learning method and online learning device for vehicle lane change decision model |
CN111882870A (en) * | 2020-07-14 | 2020-11-03 | 吉林大学 | Quantification method of road traffic environment |
TWI778428B (en) * | 2020-08-26 | 2022-09-21 | 中國商宸展光電(廈門)股份有限公司 | Method, device and system for detecting memory installation state |
CN112560782A (en) * | 2020-12-26 | 2021-03-26 | 浙江天行健智能科技有限公司 | Vehicle lane changing behavior identification method based on random forest algorithm |
CN112907987B (en) * | 2021-01-19 | 2022-03-11 | 吉林大学 | Multi-lane express way exit ramp shunting area intelligent motorcade lane change guiding method and system |
CN112907987A (en) * | 2021-01-19 | 2021-06-04 | 吉林大学 | Multi-lane express way exit ramp shunting area intelligent motorcade lane change guiding method and system |
CN112793576B (en) * | 2021-01-26 | 2022-04-01 | 北京理工大学 | Lane change decision method and system based on rule and machine learning fusion |
CN112793576A (en) * | 2021-01-26 | 2021-05-14 | 北京理工大学 | Lane change decision method and system based on rule and machine learning fusion |
CN115731261A (en) * | 2021-08-27 | 2023-03-03 | 河北省交通规划设计研究院有限公司 | Method and system for identifying lane changing behavior of vehicle based on expressway radar data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106777776A (en) | A kind of vehicle lane-changing decision-making technique based on supporting vector machine model | |
US11084494B2 (en) | Method for detecting safety of driving behavior, apparatus, device and storage medium | |
CN106080590B (en) | The acquisition methods and device of control method for vehicle and device and decision model | |
CN105620480B (en) | Intelligent vehicle independence lane-change opportunity decision-making technique | |
CN106525454B (en) | The static-obstacle thing response performance test method and checkout area of automatic driving vehicle | |
US20220126871A1 (en) | Lane Change Monitoring Method and Lane Change Monitoring System for Autonomous Vehicle | |
CN107346612A (en) | A kind of vehicle collision avoidance method and system based on car networking | |
CN112224211A (en) | Driving simulation system based on multi-autonomous-body traffic flow | |
CN108313054A (en) | The autonomous lane-change decision-making technique of automatic Pilot and device and automatic driving vehicle | |
CN107640154A (en) | New hand's driver driving accessory system | |
CN108133644A (en) | A kind of evaluation system and evaluation method of automobile driver examination | |
CN106706334A (en) | Test method and test field for capacity of autonomous driving vehicle in avoiding approaching vehicle | |
CN107919027A (en) | Predict the methods, devices and systems of vehicle lane change | |
CN105679020A (en) | Driving behavior analysis device and method | |
Wang et al. | Evaluation of lane departure correction systems using a regenerative stochastic driver model | |
CN104417550B (en) | The rear side of vehicle is to warning system and its alarm control method | |
CN105513405B (en) | A kind of track of vehicle analysis method and device | |
CN108657066A (en) | Early warning system of jumping a queue and method for early warning of jumping a queue | |
CN107848534A (en) | Controller of vehicle, control method for vehicle and wagon control program | |
CN112373485A (en) | Decision planning method for automatic driving vehicle considering interactive game | |
CN107878450A (en) | A kind of vehicle condition intelligent monitoring method based on deep learning | |
CN109849785B (en) | Lane change judgment system and method based on Internet of vehicles | |
CN106781454B (en) | The appraisal procedure and device of driving behavior | |
Scanlon et al. | Preliminary potential crash prevention estimates for an Intersection Advanced Driver Assistance System in straight crossing path crashes | |
CN207351690U (en) | Automatic driving vehicle avoids the checkout area close to vehicle capability |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170531 |
|
WD01 | Invention patent application deemed withdrawn after publication |