CN108241866A - A kind of method, apparatus guided to driving behavior and vehicle - Google Patents
A kind of method, apparatus guided to driving behavior and vehicle Download PDFInfo
- Publication number
- CN108241866A CN108241866A CN201611216065.5A CN201611216065A CN108241866A CN 108241866 A CN108241866 A CN 108241866A CN 201611216065 A CN201611216065 A CN 201611216065A CN 108241866 A CN108241866 A CN 108241866A
- Authority
- CN
- China
- Prior art keywords
- data
- attribute
- data set
- vehicle
- grader
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C5/00—Registering or indicating the working of vehicles
- G07C5/08—Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle or waiting time
- G07C5/0841—Registering performance data
Abstract
The present invention provides a kind of method, apparatus guided to driving behavior and vehicle, method to include：Collect the data that vehicle traveling is related to；The data of collection are carried out with pretreatment operation and obtains data set D, data set D has a corresponding property set A, and each data corresponds to the data involved in a vehicle travel process in data set D；Grader is built based on the data set D after data prediction, each attribute in grader is as the standard classified；The data in data set D are classified to obtain classification results using grader, according to the classification results by vehicle shunting to different tracks.Collect the data involved in vehicle travel process, the particularly data of the driving behavior of driver, data have multiple attributes, and the relative strength index for each attribute sets several classifications, structure grader classifies to vehicle traveling, realizes anti-collision warning and guiding vehicle traveling.
Description
Technical field
The present invention relates to car networking technologies, particularly relate to a kind of method, apparatus guided to driving behavior and vehicle.
Background technology
Vehicle speed bootstrap technique includes two classes, based on the vehicle assisted system of vehicle-mounted hardware, such as vehicle adaptive cruise
System (ACC, Adaptive Cruise Control) and, vehicle based on wireless network auxiliary cooperative system.
Vehicle assisted system uses multiple-sensor integration technology and integration technology, when car body is close to danger or barrier
When or advancing, having pedestrian, animal, locomotive suddenly when mobile objects are in front of car body, rear in reversing process, if
Driver does not have discovery or does not have enough time braking, and vehicle assisted system meeting active intelligently makes vehicle self-actuating brake, so as to avoid out
Now thrust into the safety accident of collision.
Vehicle assists cooperative system, is by GPS, radio frequency identification (RFID, Radio Frequency
Identification), the devices such as sensor and camera image processing, acquire itself environment and status information；Pass through interconnection
Network technology, all vehicles can be by the various information Transmission Convergences of itself to car networking platform；It is analyzed by computer technology
Information with a large amount of vehicles are handled calculates the best route of different vehicle, reports without delay road conditions and arranges signal lamp cycle.It is real
Existing deviation, safe distance between vehicles early warning, pedestrian anti-collision early warning, driver's behavior and fatigue state monitoring, road signs
Identification and intelligent bus or train route collaboration etc..
Vehicle assisted system is limited to the measuring distance of hardware, and therefore, closely using effect is good, when vehicle headway farther out
When, sensor is affected by surrounding enviroment, easily influences using effect, and such as in the environment of blind area, sensor signal is kept off
Using effect can firmly be influenced.
Vehicle auxiliary cooperative system needs vehicle operation data uploading to platform, and platform sends instructions to vehicle down again, this
The limitation of network transmission is limited to, and vehicle collision avoidance system is higher to delay requirement, the technology based on car networking platform cannot
Meet the requirement of real-time.
Existing vehicle assisted system is limited to the size in hardware controls region and inevitable blind zone problem, is only capable of
Solve the problems, such as the anticollision between vehicle and vehicle, there is no the Research on Interactive Problem for solving vehicle and vehicle.
Existing vehicle auxiliary cooperative system is limited to network bandwidth and time delay, can not solve wanting for car networking real-time
It asks.The time delay of vehicle to base station is 20 milliseconds in the lte networks, base station to Internet data center (IDC, Internet Data
Center the time delay of the car networking platform in) be 50 milliseconds, be superimposed service logic processing time will more than 100 milliseconds, and
The delay requirement of car networking application scenarios as defined in standardization body is 100 milliseconds even 20 milliseconds, existing vehicle auxiliary collaboration
System can not be up to standard.
Invention content
Technical problems to be solved of the embodiment of the present invention be to provide a kind of method, apparatus guided to driving behavior and
Vehicle realizes that anticollision and collaboration between vehicle travel, shortens the time delay in car networking application scenarios.
In order to solve the above technical problems, a kind of method guided to driving behavior provided in an embodiment of the present invention, it should
For vehicle, method includes：
Collect the data that vehicle traveling is related to；
Pretreatment operation is carried out to the data of collection and obtains data set D, data set D has corresponding property set A, data set
Each data corresponds to the data involved in a vehicle travel process in D；
Grader is built based on the data set D after data prediction, each attribute in grader is as the standard classified；
The data in data set D are classified to obtain classification results using grader, according to the classification results by vehicle
It is diverted to different tracks.
In a preferred embodiment, it further includes：
Running time is estimated according to what classification results obtained vehicle, running time and actual travel time pair are estimated by described
Than obtaining the time difference；
When the time difference being more than traveling threshold value, corresponding noise data is deleted from data set D.
In a preferred embodiment, data set D is obtained to the data of collection progress pretreatment operation to specifically include：
The corresponding property set A of setting data set D include attribute A1, A2 ..., Ak；
For each attribute setup c1, c2 ..., cn have n classification altogether.
In a preferred embodiment, included based on the data set D structure graders after data prediction：
Using ID3 decision Tree algorithms separate data collection D, the ID3 decision trees built as grader, wherein, ID3 determines
Each diverging paths of plan tree correspond to an attribute in property set A.
In a preferred embodiment, included using ID3 decision Tree algorithms separate data collection D：
Using ID3 decision Tree algorithms, in recurrence each time：
Information gain is obtained based on the entropy in information theory, use information gain as hybrid UV curing function, wherein, entropy is things
Probabilistic measurement, entropy show that more greatly things uncertainty is higher；
Optimal classification attribute is found using hybrid UV curing function；
Optimal classification attribute is selected as the attribute for separating the data set D；
When the obtained each data subset D1 of separate data collection D ..., Dv meet pure requirement condition when, it is described best
Categorical attribute causes information gain maximum, and recurrence terminates；It is described meet pure requirement condition refer to according to calculate from data set D
Enough noise datas are removed in deletion so that data subset feature is apparent.
In a preferred embodiment, information gain is obtained based on the entropy in information theory to include：
Attribute in property set A include A1, A2 ... Ai ..., Ak, and each attribute have c1, c2 ..., cn mono-
Common n classification, then the entropy entropy (D) of data set D is expressed as under reset condition：
If using attribute Ai by data set D be divided into v disjoint data subset D1 ..., Dv, use attribute Ai
After division, the entropy of data set D is：
Then the information gain gain (D, Ai) of attribute Ai is：
In a preferred embodiment, vehicle shunting is included to different tracks according to classification results：
If classified using decision Tree algorithms, if classification results are mean value, mean value represents the attribute of all vehicles
It is identical, then it is arbitrarily dispatched buses nearby to different tracks.
A kind of device guided to driving behavior, applied to vehicle, including：
Data cell, for collecting the data that vehicle traveling is related to；
Pretreatment unit obtains data set D for the data of collection to be carried out with pretreatment operation, and data set D, which has, to be corresponded to
Property set A, each data corresponds to the data involved in a vehicle travel process in data set D；
Grader unit, for building grader based on the data set D after data prediction, each attribute in grader
Standard as classification；
Dividing cell, for being classified to obtain classification results to the data in data set D using grader, according to described
Classification results are by vehicle shunting to different tracks.
In a preferred embodiment, grader unit includes：
ID3 decision tree modules, for using ID3 decision Tree algorithms separate data collection D, the ID3 decision trees built make
For grader, wherein, each diverging paths of ID3 decision trees correspond to an attribute in property set A.
In a preferred embodiment, ID3 decision trees module includes：
Recurrence module all selects optimal classification attribute as the attribute for separating the data set D for recurrence each time；
Hybrid UV curing function module, for obtaining described information gain, use information gain conduct based on the entropy in information theory
The hybrid UV curing function of ID3 decision Tree algorithms finds optimal classification attribute using hybrid UV curing function, wherein, entropy is that things is not known
The measurement of property, entropy show that more greatly things uncertainty is higher；
Separating modules, for work as the obtained each data subset D1 of separate data collection D ..., Dv meets and pure requires item
During part, the optimal classification attribute causes information gain maximum, and recurrence terminates.
A kind of vehicle, including the device guided to driving behavior.
Compared with prior art, the embodiment of the present invention at least has the advantages that：It collects and is related in vehicle travel process
And data, the particularly driving behavior of driver data, data have multiple attributes, are that the relative strength index of each attribute is set
Several classifications, structure grader classify to vehicle traveling, realize anti-collision warning and guiding vehicle traveling.
Description of the drawings
Fig. 1 is the carrier grade service environment that mobile edge calculations utilize that Radio Access Network is created；
Fig. 2 is a kind of method flow schematic diagram guided to driving behavior；
Fig. 3 is the flow diagram of ID3 decision Tree algorithms.
Specific embodiment
To make the technical problem to be solved in the present invention, technical solution and advantage clearer, below in conjunction with attached drawing and tool
Body embodiment is described in detail.In the following description, such as specific configuration is provided and the specific detail of component is only
In order to help comprehensive understanding the embodiment of the present invention.It therefore, it will be apparent to those skilled in the art that can be to reality described herein
Example is applied to make various changes and modifications without departing from scope and spirit of the present invention.In addition, for clarity and brevity, it is omitted pair
The description of known function and construction.
As shown in Figure 1, mobile edge calculations (MEC, Mobile Edge Computing) are created using Radio Access Network
As soon as produce the carrier grade service environment for having high-performance, low latency and high bandwidth, proximad telecommunication user provide IT service and
High in the clouds computing function, accelerate every content in network, service and alllication it is quick-downloading, consumer is allowed to enjoy continual high-quality
Measure network experience.
The embodiment of the present invention provides a kind of method guided to driving behavior, as shown in Fig. 2, including：
Step 201, the data that vehicle traveling is related to are collected；
Step 202, the data of collection are carried out with pretreatment operation and obtains data set D, data set D has corresponding property set
Each data corresponds to the data involved by the driving process of a vehicle in A, data set D；
Step 203, grader is built based on the data set D after data prediction, each attribute in grader as point
The standard of class；
Step 204, the data in data set D are classified to obtain classification results using grader, according to the classification
As a result by vehicle shunting to different tracks.
Using the technology provided, the driving behavior of the data, particularly driver involved in vehicle travel process is collected
Data, data have multiple attributes, and the relative strength index for each attribute sets several classifications, and structure grader travels vehicle
Classify, realize anti-collision warning and guiding vehicle traveling.
In a preferred embodiment, running time is estimated according to what classification results obtained vehicle, traveling is estimated by described
Time compares with actual travel time, obtains the time difference；
When the time difference being more than traveling threshold value, corresponding noise data is deleted from data set D.
In a preferred embodiment, data set D is obtained to the data of collection progress pretreatment operation to include：
The corresponding property set A of setting data set D include attribute A1, A2 ..., Ak；
For each attribute setup c1, c2 ..., cn have n classification altogether.
Property set A can include but is not limited to：The speed of vehicle, acceleration, brake and simultaneously linear index etc., are each attribute
A1, A2 ..., Ak is according to the power setting classifications such as high, medium and low.
In a preferred embodiment, included based on the data set D structure graders after data prediction：
Using ID3 decision Tree algorithms separate data collection D, the ID3 decision trees built are as grader.
In a preferred embodiment, included using ID3 decision Tree algorithms separate data collection D：
Using ID3 decision Tree algorithms, each time in recurrence：
Information gain is obtained based on the entropy in information theory, use information gain as hybrid UV curing function, wherein, entropy is things
Probabilistic measurement, entropy show that more greatly things uncertainty is higher；
Optimal classification attribute is found using hybrid UV curing function；
Optimal classification attribute is selected as the attribute for separating the data set D；
The obtained each data subset D1 of separate data collection D ..., Dv should meet pure requirement condition, at this point, institute
Information gain caused by stating optimal classification attribute is maximum, and recurrence terminates；Wherein, it is described to meet pure requirement condition and refer to according to meter
Calculation deletes noise data from data set D so that the feature of data subset is apparent.
In other words, ID3 decision Tree algorithms all select optimal classification category by recursively separate data collection D, each time recurrence
Property as separate current data set attribute, obtain each data subset D1 ..., Dv, wherein, selected using hybrid UV curing function
Select optimal classification attribute；
ID3 decision Tree algorithms use information gains are based on the entropy in information theory, entropy as hybrid UV curing function, information gain
The probabilistic measurement of things, entropy is bigger to illustrate that things uncertainty is higher.
Assuming that there is a data set D, the attribute of data set D is A1, A2 ... Ai ..., Ak, and each attribute have c1,
C2 ..., the common n classification of cn, then during original state, the entropy of data set D is expressed as：
If data set D can be divided into v disjoint subset Ds 1, D2..., Dv using attribute Ai, attribute Ai is used
The entropy of data set D is after division：
Then the information gain of attribute Ai is：
Information gain caused by using a certain attribute Ai separate data collection is bigger, i.e. decision tree information gain-ratio is higher, then
Illustrate that the effect of attribute Ai progress Interval datas is better, the attribute for causing information gain maximum is optimal classification attribute.
The condition that recurrence terminates is so that the finally obtained each data subset of separation is pure as far as possible, that is, to be looked for
Go out the attribute that all child nodes can be made purer, remove noise data as possible so that data subset feature is apparent.
The grader that ID3 decision Tree algorithms obtain is a tree, and each diverging paths of tree represent (possible) category
Property, each leafy node corresponds to a classification.As shown in figure 3, the step of implementing ID3 decision Tree algorithms includes：
Step 301, parameter initialization sets data set D, property set A, decision tree T { T=Φ }；
Step 302, judge, if so, going to step 311, otherwise to go to step whether only comprising a classification cj in data set D
303；
Step 303, whether property set A is empty, if so, going to step 312, otherwise goes to step 304；
Step 304, calculate the entropy p0 of data set D at this time and, calculate using the entropy pj after attribute Aj partitioned data sets D,
Entropy is the probabilistic measurement of things, and entropy shows that more greatly things uncertainty is higher.
Step 305, due to have in property set A multiple attribute A1, A2 ... Ai ..., Ak, according to ID3 algorithm meters
The method of optimal classification is calculated to calculate the corresponding optimal classification attribute Ag of max { p0-pj }, the corresponding entropys of optimal classification attribute Ag are
pg。
Step 306, whether p0-pg is less than threshold epsilon, if so, going to step 312, otherwise goes to step 307；
Step 307, increase decision node Ag, a j=0 for decision tree T, son is obtained using attribute Ag partitioned data sets D
Collect D1, D2 ... ..., Dm.
Step 308, j=j+1；
Step 309, whether subset D j is empty, if so, going to step 308, otherwise goes to step 301；
Step 310, increase a branch Tj, and D=Dj, A=A- { Ag } for decision tree T；Go to step 302；
Step 311, be decision tree T addition leaf node classification be Cj, go to step 313；
Step 312, be decision tree T addition leaf node classification be the highest category Cj of accounting in D, go to step 313；
Step 313, decision tree T is returned.
In a preferred embodiment, according to the decision tree of structure, the data concentrated to data are classified, and according to point
Class result will be on vehicle shunting to different tracks.
If being mean value according to the classification results that decision tree is classified, mean value represents that all vehicle-states are identical, at this moment
It arbitrarily can nearby dispatch buses in different lanes.
In a preferred embodiment, running time is estimated, the running time of estimating is compared with actual travel time,
Obtain the time difference；When the time difference being more than traveling threshold value, the step of corresponding noise data is deleted from data set D, has
Body includes：
It determines to estimate running time according to the result that ID3 decision Tree algorithms perform, dispatch buses in different track rows
After sailing, actual travel time is obtained, select actual travel time and estimates the larger vehicle institute of difference between running time
Corresponding data are as Sample data.
It obtains estimating running time according to ID3 decision tree implementing results, running time and actual travel time pair will be estimated
Than such as：
| T performs running time-T and estimates running time |>>△ T, △ T are the acceptable traveling threshold values of setting；
If it is more that actual travel time is more than the estimated time, illustrate that this attribute corresponding data in data set D are
Poor sample data removes the corresponding sample data of this attribute from data set D.It is compared through excessively taking turns test, data
Invalid data in collection D can be removed, and realize dynamic adjusting data collection D.
System operation for a period of time after, repeat this process, using ID3 decision tree implementing results update the data collection D,
The time span of middle system operation can adjust as the case may be.
The embodiment of the present invention provides a kind of device guided to driving behavior, applied to vehicle, including：
Data cell, for collecting the data that vehicle traveling is related to；
Pretreatment unit obtains data set D for the data of collection to be carried out with pretreatment operation, and data set D, which has, to be corresponded to
Property set A, each data corresponds to the data involved in a vehicle travel process in data set D；
Grader unit, for building grader based on the data set D after data prediction, each attribute in grader
Standard as classification；
Dividing cell, for being classified to obtain classification results to the data in data set D using grader, according to described
Classification results are by vehicle shunting to different tracks.
In a preferred embodiment, grader unit includes：
ID3 decision tree modules, for using ID3 decision Tree algorithms separate data collection D, the ID3 decision trees built make
For grader, wherein, each diverging paths of ID3 decision trees correspond to an attribute in property set A.
In a preferred embodiment, ID3 decision trees module includes：
Recurrence module all selects optimal classification attribute as the attribute for separating the data set D for recurrence each time；
Hybrid UV curing function module, for obtaining described information gain, use information gain conduct based on the entropy in information theory
The hybrid UV curing function of ID3 decision Tree algorithms finds optimal classification attribute using hybrid UV curing function, wherein, entropy is that things is not known
The measurement of property, entropy show that more greatly things uncertainty is higher；
Separating modules, for work as the obtained each data subset D1 of separate data collection D ..., Dv meets and pure requires item
During part, the optimal classification attribute causes information gain maximum, and recurrence terminates.
The embodiment of the present invention provides a kind of vehicle, including the device guided to driving behavior.
The scheme of speed guiding of the tradition based on vehicle-mounted hardware is limited to the size in hardware controls region and inevitable
Blind zone problem, and be only capable of solving the problems, such as the anticollision between vehicle and vehicle, there is no the Research on Interactive Problem for solving vehicle vehicle, efficiency
It is relatively low.
Using the technology of the application, the otherness of driving behavior has been fully considered, and based on the movement close to base station side
Edge calculations facility effectively solves blind zone problem, Research on Interactive Problem and latency issue.
It should be understood that " one embodiment " or " embodiment " that specification is mentioned in the whole text mean it is related with embodiment
A particular feature, structure, or characteristic is included at least one embodiment of the present invention.Therefore, occur everywhere in the whole instruction
" in one embodiment " or " in one embodiment " not necessarily refer to identical embodiment.In addition, these specific feature, knots
Structure or characteristic can in any suitable manner combine in one or more embodiments.
In various embodiments of the present invention, it should be appreciated that the size of the serial number of following each processes is not meant to perform suitable
The priority of sequence, the execution sequence of each process should be determined with its function and internal logic, without the implementation of the reply embodiment of the present invention
Process forms any restriction.
In addition, the terms " system " and " network " are often used interchangeably herein.
It should be understood that the terms "and/or", only a kind of incidence relation for describing affiliated partner, expression can deposit
In three kinds of relationships, for example, A and/or B, can represent：Individualism A exists simultaneously A and B, these three situations of individualism B.
In addition, character "/" herein, it is a kind of relationship of "or" to typically represent forward-backward correlation object.
In embodiment provided herein, it should be appreciated that " B corresponding with A " represents that B is associated with A, can be with according to A
Determine B.It is also to be understood that determine that B is not meant to determine B only according to A according to A, it can also be according to A and/or other information
Determine B.
The above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, without departing from the principles of the present invention, several improvements and modifications can also be made, these improvements and modifications
It should be regarded as protection scope of the present invention.
Claims (11)
- A kind of 1. method guided to driving behavior, which is characterized in that applied to vehicle, method includes：Collect the data that vehicle traveling is related to；The data of collection are carried out with pretreatment operation and obtains data set D, data set D has a corresponding property set A, in data set D Each data corresponds to the data involved in a vehicle travel process；Grader is built based on the data set D after data prediction, each attribute in grader is as the standard classified；The data in data set D are classified to obtain classification results using grader, are divided vehicle according to the classification results Flow to different tracks.
- 2. the method as described in claim 1, which is characterized in that further include：Running time is estimated according to what classification results obtained vehicle, the running time of estimating is compared with actual travel time, Obtain the time difference；When the time difference being more than traveling threshold value, corresponding noise data is deleted from data set D.
- 3. the method as described in claim 1, which is characterized in that pretreatment operation is carried out to the data of collection and obtains data set D It specifically includes：The corresponding property set A of setting data set D include attribute A1, A2 ..., Ak；For each attribute setup c1, c2 ..., cn have n classification altogether.
- 4. the method as described in claim 1, which is characterized in that grader packet is built based on the data set D after data prediction It includes：Using ID3 decision Tree algorithms separate data collection D, the ID3 decision trees built as grader, wherein, ID3 decision trees Each diverging paths correspond to an attribute in property set A.
- 5. method as claimed in claim 4, which is characterized in that included using ID3 decision Tree algorithms separate data collection D：Using ID3 decision Tree algorithms, in recurrence each time：Information gain is obtained based on the entropy in information theory, use information gain as hybrid UV curing function, wherein, entropy is that things is not true It qualitatively weighs, entropy shows that more greatly things uncertainty is higher；Optimal classification attribute is found using hybrid UV curing function；Optimal classification attribute is selected as the attribute for separating the data set D；When the obtained each data subset D1 of separate data collection D ..., Dv meet pure requirement condition when, the optimal classification Attribute causes information gain maximum, and recurrence terminates；It is described to meet pure requirement condition and refer to delete from data set D according to calculating Remove enough noise datas so that data subset feature is apparent.
- 6. method as claimed in claim 5, which is characterized in that information gain is obtained based on the entropy in information theory and is included：Attribute in property set A include A1, A2 ... Ai ..., Ak, and each attribute have c1, c2 ..., cn have n altogether A classification, then the entropy entropy (D) of data set D is expressed as under reset condition：If using attribute Ai by data set D be divided into v disjoint data subset D1 ..., Dv, divided using attribute Ai Afterwards, the entropy of data set D is：Then the information gain gain (D, Ai) of attribute Ai is：
- 7. the method as described in claim 1, which is characterized in that wrapped vehicle shunting to different tracks according to classification results It includes：If classified using decision Tree algorithms, if classification results are mean value, mean value represents that the attribute of all vehicles is identical, Then arbitrarily dispatched buses nearby to different tracks.
- 8. a kind of device guided to driving behavior, which is characterized in that applied to vehicle, including：Data cell, for collecting the data that vehicle traveling is related to；Pretreatment unit obtains data set D for the data of collection to be carried out with pretreatment operation, and data set D has corresponding category Each data corresponds to the data involved in a vehicle travel process in property collection A, data set D；Grader unit, for building grader based on the data set D after data prediction, each attribute conduct in grader The standard of classification；Dividing cell, for being classified to obtain classification results to the data in data set D using grader, according to the classification As a result by vehicle shunting to different tracks.
- 9. device as claimed in claim 8, which is characterized in that grader unit includes：ID3 decision tree modules, for using ID3 decision Tree algorithms separate data collection D, the ID3 decision trees built, which are used as, divides Class device, wherein, each diverging paths of ID3 decision trees correspond to an attribute in property set A.
- 10. device as claimed in claim 9, which is characterized in that ID3 decision tree modules include：Recurrence module all selects optimal classification attribute as the attribute for separating the data set D for recurrence each time；Hybrid UV curing function module, for obtaining described information gain based on the entropy in information theory, use information gain is determined as ID3 The hybrid UV curing function of plan tree algorithm finds optimal classification attribute using hybrid UV curing function, wherein, entropy is the probabilistic weighing apparatus of things Amount, entropy show that more greatly things uncertainty is higher；Separating modules, for work as the obtained each data subset D1 of separate data collection D ..., Dv meet pure requirement condition When, the optimal classification attribute causes information gain maximum, and recurrence terminates.
- 11. a kind of vehicle, which is characterized in that including the dress guided to driving behavior in claim 8-10 any one It puts.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611216065.5A CN108241866A (en) | 2016-12-26 | 2016-12-26 | A kind of method, apparatus guided to driving behavior and vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611216065.5A CN108241866A (en) | 2016-12-26 | 2016-12-26 | A kind of method, apparatus guided to driving behavior and vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108241866A true CN108241866A (en) | 2018-07-03 |
Family
ID=62704801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611216065.5A Pending CN108241866A (en) | 2016-12-26 | 2016-12-26 | A kind of method, apparatus guided to driving behavior and vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108241866A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109683613A (en) * | 2018-12-24 | 2019-04-26 | 驭势（上海）汽车科技有限公司 | It is a kind of for determining the method and apparatus of the ancillary control information of vehicle |
| CN110020748A (en) * | 2019-03-18 | 2019-07-16 | 杭州飞步科技有限公司 | Trajectory predictions method, apparatus, equipment and storage medium |
| WO2021056327A1 (en) * | 2019-09-26 | 2021-04-01 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for analyzing human driving behavior |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104537010A (en) * | 2014-12-17 | 2015-04-22 | 温州大学 | Component classifying method based on net establishing software of decision tree |
| CN105228215A (en) * | 2015-09-01 | 2016-01-06 | 重庆邮电大学 | Based on many copies method for routing of decision tree mechanism in vehicular ad hoc network |
-
2016
- 2016-12-26 CN CN201611216065.5A patent/CN108241866A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104537010A (en) * | 2014-12-17 | 2015-04-22 | 温州大学 | Component classifying method based on net establishing software of decision tree |
| CN105228215A (en) * | 2015-09-01 | 2016-01-06 | 重庆邮电大学 | Based on many copies method for routing of decision tree mechanism in vehicular ad hoc network |
Non-Patent Citations (1)
| Title |
|---|
| 刘志强等: "基于决策树的高速公路入口合流区域换道研究", 《重庆理工大学学报（自然科学）》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109683613A (en) * | 2018-12-24 | 2019-04-26 | 驭势（上海）汽车科技有限公司 | It is a kind of for determining the method and apparatus of the ancillary control information of vehicle |
| CN109683613B (en) * | 2018-12-24 | 2022-04-29 | 驭势（上海）汽车科技有限公司 | Method and device for determining auxiliary control information of vehicle |
| CN110020748A (en) * | 2019-03-18 | 2019-07-16 | 杭州飞步科技有限公司 | Trajectory predictions method, apparatus, equipment and storage medium |
| CN110020748B (en) * | 2019-03-18 | 2022-02-15 | 杭州飞步科技有限公司 | Trajectory prediction method, apparatus, device and storage medium |
| WO2021056327A1 (en) * | 2019-09-26 | 2021-04-01 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for analyzing human driving behavior |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105930625B (en) | Q study combines the design method of the intelligent driving behaviour decision making system of neural network | |
| US20210125076A1 (en) | System for predicting aggressive driving | |
| CN111489588B (en) | Vehicle driving risk early warning method and device, equipment and storage medium | |
| CN112487617B (en) | Collision model-based risk prevention method, device, equipment and storage medium | |
| CN105745131B (en) | Travel controlling system, server, car-mounted device | |
| US10755565B2 (en) | Prioritized vehicle messaging | |
| CN107886718B (en) | Road condition analysis method, device and network system | |
| CN111695418A (en) | Method and system for safe driving based on road condition detection | |
| CN111291697A (en) | Method and device for recognizing obstacle | |
| CN108241866A (en) | A kind of method, apparatus guided to driving behavior and vehicle | |
| CN105632203B (en) | A kind of traffic security early warning method of traffic control and system | |
| CN112925309A (en) | Intelligent networking automobile data interaction method and system | |
| CN110599791A (en) | Information monitoring method, device and equipment | |
| CN112116809A (en) | Non-line-of-sight vehicle anti-collision method and device based on V2X technology | |
| US20230339502A1 (en) | Safety measurement of autonomous vehicle driving | |
| US20220343637A1 (en) | Traffic flow machine-learning modeling system and method applied to vehicles | |
| Valiente et al. | Connected and autonomous vehicles in the deep learning era: A case study on computer-guided steering | |
| Mohammadi et al. | Signalised intersection control in a connected vehicle environment: User throughput maximisation strategy | |
| CN114787010A (en) | Driving safety system | |
| Wu et al. | Lane-GNN: Integrating GNN for Predicting Drivers' Lane Change Intention | |
| KR20160071164A (en) | Apparatus and Method for Drive Controlling of Vehicle Considering Cut in | |
| Martinelli et al. | Context-awareness mobile devices for traffic incident prevention | |
| US11374667B2 (en) | Localizing communications interference node | |
| CN113627332A (en) | Gradient control federal learning-based distraction driving behavior identification method | |
| CN113905941A (en) | Vehicle control in a geographic control zone |
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 |
Application publication date: 20180703 |
|
| RJ01 | Rejection of invention patent application after publication |