CN106710144B - A kind of driving stroke evaluation method and device - Google Patents
A kind of driving stroke evaluation method and device Download PDFInfo
- Publication number
- CN106710144B CN106710144B CN201611192641.7A CN201611192641A CN106710144B CN 106710144 B CN106710144 B CN 106710144B CN 201611192641 A CN201611192641 A CN 201611192641A CN 106710144 B CN106710144 B CN 106710144B
- Authority
- CN
- China
- Prior art keywords
- driving
- stroke
- evaluation
- index
- driving stroke
- 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.)
- Active
Links
Classifications
-
- 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
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/06—Alarms for ensuring the safety of persons indicating a condition of sleep, e.g. anti-dozing alarms
-
- 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
- G07C5/0808—Diagnosing performance data
Abstract
The present invention discloses a kind of driving stroke evaluation method and device, which comprises first according to the driving behavior data of driver, generates the index for evaluate driving stroke.Secondly, calculating any evaluation result for driving stroke under each index.Finally, the evaluation result based on the driving stroke under each index, determines the final appraisal results for driving stroke.The present invention can be pre-generated for evaluating the index for driving stroke, and be evaluated respectively from the angle of each index stroke is driven, and the final appraisal results for driving stroke are finally obtained.Compared with prior art, the present invention is realized to the Assessment for driving stroke, while evaluation result is more accurate.
Description
Technical field
The present invention relates to data processing fields, and in particular to a kind of driving stroke evaluation method and device.
Background technique
Driving stroke evaluation is that the evaluation of the safety of stroke is driven to one section, is normally based on this section and drives in stroke
Travelling data drives the evaluation that stroke carries out safety to this section.
Current driving stroke evaluation method is all the subjective method of comparison, generally by professional according to travelling data pair
Each section of driving stroke marking, to complete the evaluation to stroke is driven.Aforesaid way is largely by the subjective factor of people
It influences, it is not accurate enough for the evaluation for driving stroke.
Summary of the invention
The present invention provides a kind of driving stroke evaluation method and devices, can will drive the impact factor in stroke and divide
It for multiple indexs, and evaluates from each index stroke is driven, is realized to the Assessment for driving stroke respectively.
The present invention provides a kind of driving stroke evaluation methods, which comprises
According to the driving behavior data of driver, generate for evaluating the index for driving stroke；
Calculate evaluation result of any driving stroke under each index；
Evaluation result based on the driving stroke under each index determines the final evaluation knot for driving stroke
Fruit.
Preferably, the index includes following one or more indexs:
Focus evaluation index, handling evaluation index, evaluating pavement condition index, fatigue evaluation index, vigilance evaluation refer to
Mark.
Preferably, any evaluation result for driving stroke under the focus evaluation index is calculated, comprising:
The data that driver in any driving stroke uses mobile phone are collected, the driver includes making using the data of mobile phone
Mode of operation and corresponding mode of operation duration with mobile phone, each mode of operation using mobile phone have preset take
Value；
Based on the focus evaluation model pre-established, the driving stroke is generated under the focus evaluation index
Evaluation result；
Wherein, the focus evaluation model is
R indicates to drive evaluation result of the stroke under the focus evaluation index, λ1... λ n respectively indicates each behaviour using mobile phone
Make the corresponding value of state, t1...tnRespectively indicate λ1... the λ n corresponding mode of operation duration.
Preferably, the mode of operation using mobile phone includes sending short message, checking short message, make a phone call, hand-held phone
And hands-free phone, wherein the corresponding value of each mode of operation using mobile phone is sequentially reduced, and is positive real number.
Preferably, any evaluation result for driving stroke under the handling evaluation index is calculated, comprising:
Identify the bad steering behavior of driver in any driving stroke, and bad steering row described in unit of account mileage
For frequency；
Based on the handling evaluation model pre-established, the driving stroke is generated under the handling evaluation index
Evaluation result；
Wherein, the handling evaluation model isP (u) indicates that driving stroke exists
Evaluation result under the handling evaluation index, u indicate the frequency of bad steering behavior in unit mileage；θ1It is default
Unit mileage in bad steering behavior frequency lower limit value, θ2For bad steering behavior in preset unit mileage
The upper limit value of frequency.
Preferably, the bad steering behavior includes anxious acceleration behavior, anxious deceleration behavior and zig zag behavior.
Preferably, any zig zag behavior for driving driver in stroke is identified, comprising:
Acquire any GPS data for driving stroke, and from obtaining each position in the driving stroke in GPS data
GPS bearing data；
The azimuthal variation amount zero setting to add up from the initial position for driving stroke or the driving stroke is corresponding
Position start, successively calculate the azimuthal variation amount between every adjacent position, and to the azimuthal variation amount being calculated into
Row is cumulative, until cumulative azimuthal variation amount meets following Rule of judgment, it is determined that occurs zig zag behavior, and will be cumulative
Azimuthal variation amount zero setting；
Wherein, the Rule of judgment isΔ bearing expression is driven
The azimuthal variation amount to add up in stroke is sailed, thres indicates the preset bend curvature for being identified as zig zag behavior, and 0 <
Thres≤90.
Preferably, any evaluation result for driving stroke under the evaluating pavement condition index is calculated, comprising:
Identify the jogging period in any driving stroke；
According to the jogging time started and jogging end time of the jogging period, calculating is persistently walked or drive slowly the time；
Accounting of the persistently jogging time in the total duration for driving stroke is calculated, is accounted for as the jogging time
Than；
Based on the road condition evaluation model pre-established, the evaluation for driving stroke under the evaluating pavement condition index is generated
As a result；
Wherein, the road condition evaluation model isP (u) indicates to drive stroke in institute
The evaluation result under evaluating pavement condition index is stated, u indicates the jogging time accounting, θ1For the lower limit of preset jogging time accounting
Value, θ2For the upper limit value of preset jogging time accounting.
Preferably, the identification any jogging period driven in stroke, comprising:
It is determined as the period of walking or drive slowly less than the period of preset first threshold value for average speed is met in any driving stroke.
Preferably, any evaluation result for driving stroke under the fatigue evaluation index is calculated, comprising:
Using preset duration as the duration of unit sub-line journey, any driving stroke is in turn divided into several sub-line journeys；
Determine that default driving period belonging to each sub-line journey, each driving period have preset value；
Based on the fatigue evaluation model pre-established, the evaluation for driving stroke under the fatigue evaluation index is generated
As a result；
Wherein, the fatigue evaluation model is
R indicates to drive evaluation result of the stroke under the fatigue evaluation index, R (tn) indicate that n-th of sub-line journey is commented in the fatigue
Evaluation result under valence index, λ1... λ n respectively indicates the value that the period is driven belonging to each sub-line journey, and λ1< λ2< ... <
λn, t1...tnRespectively indicate λ1... the duration of period, k are driven belonging to the corresponding sub-line journey of λ n1...knRespectively in λ1...λ
When the corresponding sub-line journey of n belongs to the default fatigue driving period driven in the period, value is greater than 1, and otherwise value is equal to 1.
Preferably, the driving period includes fatigue driving period and normal driving period, the fatigue driving period packet
Include morning peak period, evening peak period and night-time hours.
Preferably, any evaluation result for driving stroke under the vigilant evaluation index is calculated, comprising:
Obtain driving over the speed limit the time in any driving stroke；
Drive over the speed limit described in calculating the time it is described drive stroke total duration in accounting, accounted for as overspeed time
Than；
Based on the vigilant evaluation model pre-established, the driving stroke is generated under the vigilant evaluation index
Evaluation result；
Wherein, the vigilant evaluation model isP (u) indicates that driving stroke exists
Evaluation result under the vigilance evaluation index, u indicate the overspeed time accounting, θ1For preset overspeed time accounting
Lower limit value, θ2For the upper limit value of preset overspeed time accounting.
Preferably, before the time of driving over the speed limit obtained in any driving stroke, further includes:
Acquisition drives the GPS data of stroke, and from extracting driving for each road in the driving stroke in the GPS data
Sail data；
Using the driving data of each road in the driving stroke as the input of default classifier, by the classifier
Processing after, determine the road type of each road；
Based on the road type of each road, driving over the speed limit the time on each road is determined.
Preferably, the driving data of each road includes in poor velocity standard on each road, speed mean value and unit
Journey bend number.
Preferably, the evaluation result based on the driving stroke under each index, determines the driving stroke
Final appraisal results, comprising:
Based on the relationship of the evaluation result under accident occurrence probability and each index, the evaluation result of each index is determined
Weight coefficient；
Evaluation result of the driving stroke under each index is weighted and averaged, obtains the driving stroke most
Whole evaluation result.
Preferably, the relationship based on the evaluation result under accident occurrence probability and each index, determines each index
Evaluation result weight coefficient, comprising:
The driving stroke that evaluation is completed is obtained in advance, and according to accident whether occurs in each driving stroke, is determined each
A classification for driving stroke, the classification include first category and second category, and accident occurs for the first category expression, described
Accident does not occur for second category expression；
The driving stroke of the first category and the driving stroke of the second category are obtained respectively under same index
Evaluation result, respectively as first category data and second category data；
Feature extraction is carried out to the first category data and the second category data respectively, and to the feature extracted
Cosine similarity calculating is carried out, the characteristic similarity value of the first category data Yu the second category data is obtained；
According to the characteristic similarity value, the weight coefficient of the index is determined, wherein the weight coefficient and the spy
Levying similarity value has inversely prroportional relationship.
The embodiment of the invention also provides a kind of driving stroke evaluating apparatus, described device includes:
Generation module is generated for the driving behavior data according to driver for evaluating the index for driving stroke；
Computing module, for calculating evaluation result of any driving stroke under each index；
Determining module determines the driving stroke for the evaluation result based on the driving stroke under each index
Final appraisal results.
Preferably, the index includes following one or more indexs:
Focus evaluation index, handling evaluation index, evaluating pavement condition index, fatigue evaluation index, vigilance evaluation refer to
Mark.
Preferably, the computing module, comprising:
First collects submodule, and the data of mobile phone, the driver are used for collecting driver in any driving stroke
Data using mobile phone include the mode of operation and corresponding mode of operation duration using mobile phone, each using mobile phone
Mode of operation has preset value；
First generates submodule, for generating the driving stroke in institute based on the focus evaluation model pre-established
State the evaluation result under focus evaluation index；
Wherein, the focus evaluation model is
R indicates to drive evaluation result of the stroke under the focus evaluation index, λ1... λ n respectively indicates each behaviour using mobile phone
Make the corresponding value of state, t1...tnRespectively indicate λ1... the λ n corresponding mode of operation duration.
Preferably, the mode of operation using mobile phone includes sending short message, checking short message, make a phone call, hand-held phone
And hands-free phone, wherein the corresponding value of each mode of operation using mobile phone is sequentially reduced, and is positive real number.
Preferably, the computing module, comprising:
Identify submodule, any bad steering behavior for driving driver in stroke, and unit of account mileage for identification
Described in bad steering behavior frequency；
Second generates submodule, for generating the driving stroke in institute based on the handling evaluation model pre-established
State the evaluation result under handling evaluation index；
Wherein, the handling evaluation model isP (u) indicates that driving stroke exists
Evaluation result under the handling evaluation index, u indicate the frequency of bad steering behavior in unit mileage；θ1It is default
Unit mileage in bad steering behavior frequency lower limit value, θ2For bad steering behavior in preset unit mileage
The upper limit value of frequency.
Preferably, the bad steering behavior includes anxious acceleration behavior, anxious deceleration behavior and zig zag behavior.
Preferably, the identification submodule, comprising:
First acquisition submodule is driven described in acquisition for acquiring any GPS data for driving stroke, and from GPS data
Sail the GPS bearing data of each position in stroke；
Determine submodule, the azimuth for adding up from the initial position for driving stroke or the driving stroke
The corresponding position of variable quantity zero setting starts, successively calculate the azimuthal variation amount between every adjacent position, and to being calculated
Azimuthal variation amount adds up, until cumulative azimuthal variation amount meets following Rule of judgment, it is determined that take a sudden turn
Behavior, and the azimuthal variation amount zero setting that will be added up；
Wherein, the Rule of judgment isΔ bearing expression is driven
The azimuthal variation amount to add up in stroke is sailed, thres indicates the preset bend curvature for being identified as zig zag behavior, and 0 <
Thres≤90.
Preferably, the computing module, comprising:
Second identifies submodule, for identification any jogging period driven in stroke；
First computational submodule, for the jogging time started and jogging end time according to the jogging period, meter
Calculation is persistently walked or drive slowly the time；
Second computational submodule, for calculating the persistently jogging time in the total duration for driving stroke
Accounting, as jogging time accounting；
Third generates submodule, for generating the driving stroke described based on the road condition evaluation model pre-established
Evaluation result under evaluating pavement condition index；
Wherein, the road condition evaluation model isP (u) indicates to drive stroke in institute
The evaluation result under evaluating pavement condition index is stated, u indicates the jogging time accounting, θ1For the lower limit of preset jogging time accounting
Value, θ2For the upper limit value of preset jogging time accounting.
Preferably, the second identification submodule, comprising:
First determines submodule, for will meet time of the average speed less than preset first threshold value in any driving stroke
Section is determined as the period of walking or drive slowly.
Preferably, the computing module, comprising:
Submodule is divided, for using preset duration as the duration of unit sub-line journey, any driving stroke to be in turn divided into
Several sub-line journeys；
Second determines submodule, for determining that default driving period belonging to each sub-line journey, each driving period have
Preset value；
4th generates submodule, for generating the driving stroke described based on the fatigue evaluation model pre-established
Evaluation result under fatigue evaluation index；
Wherein, the fatigue evaluation model is
R indicates to drive evaluation result of the stroke under the fatigue evaluation index, R (tn) indicate that n-th of sub-line journey is commented in the fatigue
Evaluation result under valence index, λ1... λ n respectively indicates the value that the period is driven belonging to each sub-line journey, and λ1< λ2< ... <
λn, t1...tnRespectively indicate λ1... the duration of period, k are driven belonging to the corresponding sub-line journey of λ n1...knRespectively in λ1...λ
When the corresponding sub-line journey of n belongs to the default fatigue driving period driven in the period, value is greater than 1, and otherwise value is equal to 1.
Preferably, the driving period includes fatigue driving period and normal driving period, the fatigue driving period packet
Include morning peak period, evening peak period and night-time hours.
Preferably, the computing module, comprising:
Second acquisition submodule, for obtaining driving over the speed limit the time in any driving stroke；
Third computational submodule described drives over the speed limit the time in the total duration for driving stroke for calculating
Accounting, as overspeed time accounting；
5th generates submodule, for generating the driving stroke in institute based on the vigilant evaluation model pre-established
State the evaluation result under vigilant evaluation index；
Wherein, the vigilant evaluation model isP (u) indicates that driving stroke exists
Evaluation result under the vigilance evaluation index, u indicate the overspeed time accounting, θ1For preset overspeed time accounting
Lower limit value, θ2For the upper limit value of preset overspeed time accounting.
Preferably, described device, further includes:
Second extracting sub-module is driven described in extraction for acquiring the GPS data for driving stroke, and from the GPS data
Sail the driving data of each road in stroke；
Classification submodule, for using the driving data of each road in the driving stroke as the defeated of default classifier
Enter, after the processing of the classifier, determines the road type of each road；
Third determines submodule, for the road type based on each road, determine on each road when driving over the speed limit
Between.
Preferably, the driving data of each road includes in poor velocity standard on each road, speed mean value and unit
Journey bend number.
Preferably, the determining module, comprising:
4th determines submodule, for the relationship based on the evaluation result under accident occurrence probability and each index, determines
The weight coefficient of the evaluation result of each index；
Submodule is weighted, for being weighted and averaged to evaluation result of the driving stroke under each index, is obtained
The final appraisal results for driving stroke.
Preferably, the described 4th submodule is determined, comprising:
5th determines submodule, for obtaining the driving stroke that evaluation is completed in advance, and according in each driving stroke
Whether accident occurs, determines each classification for driving stroke, the classification includes first category and second category, the first kind
Accident Biao Shi not occur, accident does not occur for the second category expression；
Third acquisition submodule, the driving of driving stroke and the second category for obtaining the first category respectively
Evaluation result of the stroke under same index, respectively as first category data and second category data；
Third extracting sub-module is mentioned for carrying out feature to the first category data and the second category data respectively
It takes, and cosine similarity calculating is carried out to the feature extracted, obtain the first category data and the second category data
Characteristic similarity value；
6th determines submodule, for determining the weight coefficient of the index according to the characteristic similarity value, wherein
The weight coefficient and the characteristic similarity value have inversely prroportional relationship.
The present invention provides a kind of driving stroke evaluation method and devices, first according to the driving behavior data of driver,
It generates for evaluating the index for driving stroke.Secondly, calculating any evaluation result for driving stroke under each index.Finally,
Evaluation result based on the driving stroke under each index, determines the final appraisal results for driving stroke.The present invention
It can pre-generate for evaluating the index for driving stroke, and evaluate respectively from the angle of each index stroke is driven,
Finally obtain the final appraisal results for driving stroke.Compared with prior art, the present invention realizes the automation to stroke is driven
Evaluation, while evaluation result is more accurate.
Detailed description of the invention
In order to more clearly explain the technical solutions in the embodiments of the present application, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, the drawings in the following description are only some examples of the present application, for
For those of ordinary skill in the art, without any creative labor, it can also be obtained according to these attached drawings
His attached drawing.
Fig. 1 is a kind of driving stroke evaluation method flow chart provided in an embodiment of the present invention；
Fig. 2 is provided in an embodiment of the present invention a kind of to calculate any driving stroke commenting under the focus evaluation index
The method flow diagram of valence result；
Fig. 3 is provided in an embodiment of the present invention a kind of to calculate any driving stroke commenting under the handling evaluation index
The method flow diagram of valence result；
Fig. 4 is a kind of azimuthal coordinate schematic diagram of sample point-provided in an embodiment of the present invention；
Fig. 5 is a kind of evaluation for calculating any driving stroke under the evaluating pavement condition index provided in an embodiment of the present invention
As a result method flow diagram；
Fig. 6 is a kind of evaluation for calculating any driving stroke under the fatigue evaluation index provided in an embodiment of the present invention
As a result method flow diagram；
Fig. 7 is a kind of method schematic diagram that driving stroke is divided into sub-line journey provided in an embodiment of the present invention；
Fig. 8 is provided in an embodiment of the present invention a kind of to calculate any driving stroke commenting under the vigilant evaluation index
The method flow diagram of valence result；
Fig. 9 is a kind of road type recognition principle schematic diagram provided in an embodiment of the present invention；
Figure 10 is a kind of method flow diagram for the weight coefficient for determining each index provided in an embodiment of the present invention；
Figure 11 is a kind of driving stroke evaluating apparatus structural schematic diagram provided in an embodiment of the present invention.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on
Embodiment in the application, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall in the protection scope of this application.
It is a kind of driving stroke evaluation method flow chart provided in an embodiment of the present invention with reference to Fig. 1, the method specifically may be used
To include:
S101: it according to the driving behavior data of driver, generates for evaluating the index for driving stroke.
In the embodiment of the present invention, the driving behavior data of driver are obtained first, and the driving behavior data may include
Traffic information etc. in the operation behavior data and driving procedure of driver itself.Secondly, to the driving behavior data
After being analyzed, generate for evaluating the index for driving stroke.
It can be made of following one or more indexs in the embodiment of the present invention for evaluating the index of driving stroke, specifically
For focus evaluation index, handling evaluation index, evaluating pavement condition index, fatigue evaluation index, vigilant evaluation index.
In addition, in the embodiment of the present invention with no restrictions to the index for evaluating driving stroke.
S102: any evaluation result for driving stroke under each index is calculated.
S103: based on evaluation result of the driving stroke under each index, the most final review for driving stroke is determined
Valence result.
In the embodiment of the present invention, when evaluating either segment driving stroke, the driving row is calculated separately out first
Evaluation result of the journey under each index.Secondly, the evaluation result based on the driving stroke under each index, final to determine
The final appraisal results for driving stroke out.
In a kind of implementation, the driving stroke is being calculated after the evaluation result under each index, respectively respectively
Corresponding weight coefficient is arranged in a evaluation result, and the most final review for driving stroke is finally determined in the way of average weighted
Valence result.
It is raw first according to the driving behavior data of driver in driving stroke evaluation method provided in an embodiment of the present invention
At for evaluating the index for driving stroke.Secondly, calculating any evaluation result for driving stroke under each index.Finally, base
In evaluation result of the driving stroke under each index, the final appraisal results for driving stroke are determined.The present invention is real
Applying example can pre-generate for evaluating the index for driving stroke, and comment respectively from the angle of each index stroke is driven
Valence finally obtains the final appraisal results for driving stroke.Compared with prior art, the embodiment of the present invention is realized to driving stroke
Assessment, while evaluation result is more accurate.
The index to each for evaluating driving stroke is introduced individually below, provides calculate any driving stroke respectively
The method of evaluation result under each index.
It is a kind of any driving stroke of calculating provided in an embodiment of the present invention in the focus evaluation index with reference to Fig. 2
Under evaluation result method flow diagram.The method specifically includes:
S201: collecting any data for driving driver in stroke and using mobile phone, and the driver uses the data of mobile phone
Mode of operation and corresponding mode of operation duration, each mode of operation using mobile phone including using mobile phone have pre-
If value.
S202: it based on the focus evaluation model pre-established, generates the driving stroke and refers in focus evaluation
Evaluation result under mark；
Wherein, the focus evaluation model is
R indicates to drive evaluation result of the stroke under the focus evaluation index, λ1... λ n respectively indicates each behaviour using mobile phone
Make the corresponding value of state, t1...tnRespectively indicate λ1... the λ n corresponding mode of operation duration.
Since driver will lead to driver distraction using mobile phone in driving procedure, reduces and drive focus, so, this
Inventive embodiments are collected to driver in stroke is driven using the data of mobile phone, including use mobile phone mode of operation and
The corresponding mode of operation duration, and the driver based on collection uses the data of mobile phone to the driving stroke described special
Evaluation result under note degree evaluation index is calculated.
Specifically, to driving determine to include following several after the event that may cause driver distraction in stroke is analyzed
The event of type: the first is screen event (event is actively opened in such as screen passive open)；Second is sound and vibration thing
Part (such as incoming ring tone, incoming call vibration, short message sound, short message vibration, notification voice, notice shock event)；The third is communication
Event (such as receives calls, makes a phone call, check short message, send short message, send IM, check IM, check mail, if using
Bluetooth headset or onboard system such as receive calls at the events).By driver in investigation driving procedure using mobile phone to driving stroke
Safety influence, be concluded that
The first, sequence of the driver using the mode of operation of mobile phone to the safety effects for driving stroke: transmission short message > look into
See that short message > making a phone call > holds phone > hands-free phone.
The second, the influence otherness using hand-held phone and hands-free phone to drive safety is not very big.Although using
The hand free devices such as bluetooth headset can reduce the risk of driving, but effect is not significant, and after prolonged call, different
The risk of talking mode reaches unanimity.
Third, dialog context are bigger compared with influence of the call form to driving behavior, and especially most have can for interference call
Road traffic accident can be caused, even if conversation content is simple, can equally cause the thinking activities of driver, disperse the note of driver
Meaning power, to influence driving behavior.The conversation content high for those mental demands but will directly result in safety accident danger
Increase.
4th, by the analysis to the drive safety evaluation under different riving conditions, mobile phone time is used with driving
Increase, drive safety significantly successively decreases under any type of talking mode and the dialog context of any property, and frequently it is short
When call it is dangerous bigger.
Based on conclusions, the embodiment of the present invention pre-establishes one and comments for evaluating the focus of focus evaluation index
Valence modelMode of operation using mobile phone includes sending
Short message checks short message, makes a phone call, holding phone and hands-free phone, wherein λ1... the value of λ n can satisfy following condition:
It sends short message > checking short message > and makes a phone call > hold phone > hands-free phone, that is to say, that the λ determined according to above-mentioned relation1...
In the value of λ n, the corresponding value of short message is sent greater than the corresponding value of short message is checked, each mode of operation using mobile phone divides
Not corresponding value is sequentially reduced, and is positive real number.
It is a kind of any driving stroke of calculating provided in an embodiment of the present invention in the handling evaluation index with reference to Fig. 3
Under evaluation result method flow diagram.The method specifically includes:
S301: any bad steering behavior for driving driver in stroke of identification, and it is bad described in unit of account mileage
The frequency of driving behavior.
S302: it based on the handling evaluation model pre-established, generates the driving stroke and refers in the handling evaluation
Evaluation result under mark；
Wherein, the handling evaluation model isP (u) indicates that driving stroke exists
Evaluation result under the handling evaluation index, u indicate the frequency of bad behavior in unit mileage；θ1For preset list
The lower limit value of the frequency of bad behavior, θ in the mileage of position2For in preset unit mileage the frequency of bad behavior it is upper
Limit value.
Described in the embodiment of the present invention manipulation sexual valence index under evaluation result be based on driver in driving procedure
What bad steering behavior determined, accelerate, suddenly slow down, zig zag behavior such as anxious.Table is used for the operability evaluation for driving stroke
Driver is levied to the control ability of driving and the stability of driving.
The embodiment of the present invention is before evaluating the handling index for driving stroke, it is necessary first to identify the driving
Bad steering behavior in the process, such as anxious acceleration, anxious deceleration, zig zag behavior.Then the handling evaluation pre-established is utilized
Model realization is to the handling evaluation for driving stroke.
In practical application, anxious acceleration, anxious deceleration behavior are the behavior of throttle, brake acutely to be operated in driving conditions, and know
Not Jia Shi the zig zag behavior in stroke can be based on GPS data.
Specifically, firstly, any GPS data for driving stroke of acquisition, and from being obtained in GPS data in the driving stroke
The GPS bearing data of each position.Secondly, successively calculating every adjacent position since the initial position for driving stroke
Between azimuthal variation amount, and add up to the azimuthal variation amount being calculated, until cumulative azimuthal variation amount
Meet following Rule of judgment, it is determined that zig zag behavior occurs；By currently cumulative azimuthal variation amount zero setting, and continue to calculate
Azimuthal variation amount in the driving stroke between subsequent every adjacent position, and to the azimuthal variation amount being calculated
It adds up, until cumulative azimuthal variation amount meets following Rule of judgment, zig zag behavior is had occurred in determination again.According to
Aforesaid way completes the processing for driving the GPS bearing data of all positions in stroke, finally counts the driving stroke
In the number of zig zag behavior occurs in total.
Wherein, the Rule of judgment isΔ bearing expression is driven
Sail the azimuthal variation amount to add up in stroke, thres indicates the preset bend curvature for being identified as zig zag behavior, according to
The size of the demand setting thres at family, the bend curvature to be identified is bigger, then the value of thres is bigger, it should be noted that 0 <
Thres≤90.
As shown in figure 4, being a kind of azimuthal coordinate schematic diagram of sample point-provided in an embodiment of the present invention.Wherein stain
The position that significant changes take place in bearing data is marked, illustrates the azimuthal variation to add up in next location point
Amount can be able to satisfy above-mentioned Rule of judgment, i.e. generation zig zag behavior.
In the embodiment of the present invention, after the frequency of bad steering behavior occurs in identifying the driving stroke, meter
The frequency of bad steering behavior in unit mileage is calculated, and the frequency of bad steering behavior is low in the unit mileage
In lower limit value θ1When, determine that evaluation result of the driving stroke under the handling evaluation index is 1；In the unit
The frequency of bad steering behavior is higher than upper limit value θ in journey2When, determine the driving stroke in the handling evaluation index
Under evaluation result be 0；The frequency of bad steering behavior is not higher than upper limit value θ in the unit mileage2And it is not less than
Lower limit value θ1When, calculating upper limit value θ2After the difference of the frequency of bad steering behavior in the unit mileage, with upper limit value
θ2With lower limit value θ1Difference do ratio, obtain the evaluation result under the handling evaluation index.
It is a kind of any driving stroke of calculating provided in an embodiment of the present invention under the evaluating pavement condition index with reference to Fig. 5
Evaluation result method flow diagram.The method specifically includes:
S501: identification any jogging period driven in stroke.
S502: according to the jogging time started and jogging end time of the jogging period, calculating is persistently walked or drive slowly the time.
S503: accounting of the persistently jogging time in the total duration for driving stroke is calculated, as jogging
Time accounting.
S504: based on the road condition evaluation model pre-established, the driving stroke is generated under the evaluating pavement condition index
Evaluation result.
Wherein, the road condition evaluation model isP (u) indicates to drive stroke in institute
The evaluation result under evaluating pavement condition index is stated, u indicates the jogging time accounting, θ1For the lower limit of preset jogging time accounting
Value, θ2For the upper limit value of preset jogging time accounting.
The friendship that evaluating pavement condition is mainly based upon driver's driving procedure is carried out to the driving stroke in the embodiment of the present invention
Through-current capacity.Since the size of the magnitude of traffic flow directly affects the psychological stress degree of driver and the height of traffic accident rate, hand over
Interpreter's event growth rate is close with the growth rate of average traffic flow, so, evaluating pavement condition is commented the safety that stroke summarizes is driven
Valence is of great significance.In practical application, evaluating pavement condition be based on drive stroke in drive speed information evaluated, if
It persistently walks or drive slowly in driving procedure, then proves current road conditions congestion.
In practical application, identification drives the jogging period in stroke first, and according to the jogging of the jogging period
Time started and jogging end time, calculating are persistently walked or drive slowly the time.It goes secondly, calculating the persistently jogging time in the driving
Accounting in the total duration of journey, as jogging time accounting.It is based ultimately upon road condition evaluation model, calculates the driving stroke
Evaluation result under the evaluating pavement condition index.
In a kind of implementation, the period for meeting average speed and being less than preset first threshold value in stroke any will be driven
It is determined as the period of walking or drive slowly, to complete the identification to the jogging period driven in stroke.
It is a kind of any driving stroke of calculating provided in an embodiment of the present invention under the fatigue evaluation index with reference to Fig. 6
Evaluation result method flow diagram.The method specifically includes:
S601: using preset duration as the duration of unit sub-line journey, any driving stroke is in turn divided into several sub-line journeys.
S602: determine that preset fatigue belonging to each sub-line journey drives the period, each fatigue driving period has preset
Value.
S603: based on the fatigue evaluation model pre-established, the driving stroke is generated under the fatigue evaluation index
Evaluation result.
Wherein, the fatigue evaluation model isR
It indicates to drive evaluation result of the stroke under the fatigue evaluation index, R (tn) indicate n-th of sub-line journey in the fatigue evaluation
Evaluation result under index, λ1...λnRespectively indicate the value of each sub-line journey affiliated fatigue driving period, and λ1< λ2< ...
< λn, t1...tnRespectively indicate λ1... the duration of the corresponding sub-line journey of λ n affiliated fatigue driving period, k1...knExist respectively
λ1... when the corresponding sub-line journey of λ n belongs to the preset fatigue driving period, value is greater than 1, and otherwise value is equal to 1.
Since fatigue driving is to cause the one of the major reasons of traffic fatalities, cause fatigue driving the reason of have very
It is more, wherein the main reason for driving and fatigue driving period (such as driving at night) are initiation driving fatigues for a long time, so, it is right
The fatigue evaluation for driving stroke is based on driving duration and to drive the period.
The embodiment of the present invention can preset the fatigue driving period as morning peak period (such as 7:00-9:30), evening peak
Period (such as 17:30-20:00) and night-time hours (such as 20:00-6:00).Specifically, (such as according to preset duration by driving stroke
30 minutes) several sub-line journeys are successively divided into, it is that a kind of stroke that will drive provided in an embodiment of the present invention is divided into reference to Fig. 7
The method schematic diagram of sub-line journey.Secondly, determining the driving period belonging to each sub-line journey, preset wherein each driving period has
Value, the corresponding value for driving the period is assigned to the λ of the sub-line journey.In addition, obtaining each sub-line journey in affiliated driving
The duration of period, and it is assigned to the t of the sub-line journey.If the sub-line journey belongs to preset fatigue driving period, k value
For the numerical value greater than 1, otherwise k value is 1.In addition, drive safety gradually decreases due to the increase with driving time, institute
λ value with latter cross-talk stroke compared to the last period sub-line journey is gradually increased.
It is a kind of any driving stroke of calculating provided in an embodiment of the present invention in the vigilant evaluation index with reference to Fig. 8
Under evaluation result method flow diagram.The method specifically includes:
S801: driving over the speed limit the time in any driving stroke is obtained.
S802: accounting of the time in the total duration for driving stroke of driving over the speed limit described in calculating, as hypervelocity
Time accounting.
S803: it based on the vigilant evaluation model pre-established, generates the driving stroke and refers in the vigilant evaluation
Evaluation result under mark.
Wherein, the vigilant evaluation model isP (u) indicates that driving stroke exists
Evaluation result under the vigilance evaluation index, u indicate the overspeed time accounting, θ1For preset overspeed time accounting
Lower limit value, θ2For the upper limit value of preset overspeed time accounting.
Drive over the speed limit be cause traffic accident it is most important due to one of, the speed of urban road is higher than 80km/h category
In driving over the speed limit, the speed of super expressway is more than that 120km/h is also belonged to and driven over the speed limit.So the embodiment of the present invention is based on driving
The driving data of each road carries out vigilant evaluation to the driving stroke in stroke.Firstly, the embodiment of the present invention needs reality
Now drive the identification of the road type of each road in stroke.If travelling on urban road, Statistical Speed is more than 80km/
H's drives over the speed limit the time；If travelling on super expressway, Statistical Speed is more than driving over the speed limit the time for 120km/h.Statistics institute
Accounting of the summation for driving the hypervelocity form time in stroke in the total duration for driving stroke is stated, when as hypervelocity
Between accounting.Finally, based on vigilant evaluation model, the evaluation knot for driving stroke under the vigilant evaluation index is generated
Fruit.
It is first before obtaining the time of driving over the speed limit driven in stroke in a kind of embodiment that realizing road type identification
First acquisition drive stroke GPS data, and from extracted in the GPS data it is described driving stroke in each road driving number
According to.Secondly, using the driving data of each road in the driving stroke as the input of default classifier, by the classifier
Processing after, determine the road type of each road.Wherein, the driving data of each road includes the speed cone on each road
Quasi- poor, speed mean value and unit mileage bend number etc..Finally, the road type based on each road, determines on each road
Drive over the speed limit the time.
For example, the classifier can be the knowledge that a time window carries out road type with 5min in the embodiment of the present invention
Not, the feature of input is poor velocity standard on each road, speed mean value and unit mileage bend number.The classifier is built
Vertical disaggregated model, classification is road type, and 1 represents urban road, and 2 represent super expressway, is the embodiment of the present invention with reference to Fig. 9
A kind of schematic illustration that road type identification is carried out using KNN classifier provided.Wherein, by from drive stroke GPS number
The velocity standard on each road extracted in is poor, speed mean value and unit mileage bend number are classified as the KNN
The input of device, exports the road type of each road after the classification processing of the KNN, that is, belongs to super expressway or city road
Road.
The embodiment of the present invention is obtaining one section of driving stroke after the evaluation result under each index, is gone based on the driving
Evaluation result of the journey under each index determines the final appraisal results for driving stroke.In a kind of implementation, it is based on thing
Therefore the relationship of the evaluation result under occurrence probability and each index, determine the weight coefficient of the evaluation result of each index.To institute
It states evaluation result of the driving stroke under each index to be weighted and averaged, obtains the final appraisal results for driving stroke.
With reference to Figure 10, for the embodiment of the invention also provides a kind of method flow diagram of weight coefficient for determining each index,
The described method includes:
S1001: obtaining the driving stroke of evaluation is completed in advance, and according to whether accident occurring in each driving stroke,
Determine that each classification for driving stroke, the classification include first category and second category, the first category indicates that thing occurs
Therefore accident does not occur for the second category expression.
In the embodiment of the present invention, determined under accident occurrence probability and each index using the driving stroke for having completed evaluation
Evaluation result relationship, and then determine the weight coefficient of the evaluation result of each index.It is commented specifically, obtaining to be completed in advance
The driving stroke of valence, and classify to each driving stroke, the driving stroke that accident occurs specifically is determined as the first kind
Not, the driving stroke that accident does not occur is determined as second category.
S1002: the driving stroke for driving stroke and the second category of the first category is obtained respectively in same finger
Evaluation result under mark, respectively as first category data and second category data.
In the embodiment of the present invention, after completing to each classification for driving stroke, an index is determined, and obtain institute respectively
Evaluation result of the driving stroke of the driving stroke and the second category of stating first category under the index, as the first kind
Other data and second category data.
S1003: carrying out feature extraction to the first category data and the second category data respectively, and to extracting
Feature carry out cosine similarity calculating, obtain the characteristic similarity of the first category data Yu the second category data
Value.
In the embodiment of the present invention, the first category data got and the second category data are carried out respectively special
Sign is extracted, specifically, the feature extracted may include mean value, mode evidence, standard deviation, four/tertile, a quarter point
It is any one or more in digit.Secondly, carrying out cosine similarity calculating to the feature extracted, the first category is obtained
The characteristic similarity value of data and the second category data.It is worth noting that, the first category data and described second
The characteristic similarity value of categorical data is smaller, i.e., the otherness of first category data and second category data under the described index is got over
Greatly, that is to say, that the evaluation result under the index is bigger to the safety effects for driving stroke.
S1004: according to the characteristic similarity value, determine the weight coefficient of the index, wherein the weight coefficient with
The characteristic similarity value has inversely prroportional relationship.
Under characteristic similarity value and the index in order to embody the first category data and the second category data
Evaluation result influence inversely proportional relationship to driving safety travel, in a kind of implementation, according to the characteristic similarity
The inverse of value determines the weight coefficient of the index.
Specifically, calculating the corresponding feature phase of each index by the method for the weight coefficient of each index of above-mentioned determination
Like angle value, and the inverse of the characteristic similarity value is calculated, then, calculates falling for the corresponding characteristic similarity value of each index
Count the accounting in the inverse sum of the characteristic similarity value of all indexs, the weight coefficient as each index.Finally, to each finger
Target evaluation result is weighted and averaged, and obtains the final appraisal results for driving stroke.
Driving stroke evaluation method provided in an embodiment of the present invention, can be based under accident occurrence probability and each index
The incidence relation of evaluation result is evaluated stroke is driven, and obtained final appraisal results are more acurrate.
It is provided in an embodiment of the present invention that the embodiment of the invention provides a kind of driving stroke evaluating apparatus with reference to Figure 11
A kind of structural schematic diagram driving stroke evaluating apparatus, described device include:
Generation module 1101 is generated for the driving behavior data according to driver for evaluating the index for driving stroke；
Computing module 1102, for calculating evaluation result of any driving stroke under each index；
Determining module 1103 determines the driving for the evaluation result based on the driving stroke under each index
The final appraisal results of stroke.
Wherein, the index includes following one or more indexs:
Focus evaluation index, handling evaluation index, evaluating pavement condition index, fatigue evaluation index, vigilance evaluation refer to
Mark.
Specifically, the computing module, comprising:
First collects submodule, and the data of mobile phone, the driver are used for collecting driver in any driving stroke
Data using mobile phone include the mode of operation and corresponding mode of operation duration using mobile phone, each using mobile phone
Mode of operation has preset value；
First generates submodule, for generating the driving stroke in institute based on the focus evaluation model pre-established
State the evaluation result under focus evaluation index；
Wherein, the focus evaluation model is
R indicates to drive evaluation result of the stroke under the focus evaluation index, λ1... λ n respectively indicates each behaviour using mobile phone
Make the corresponding value of state, t1...tnRespectively indicate λ1... the λ n corresponding mode of operation duration.
Specifically, the mode of operation using mobile phone includes sending short message, checking short message, make a phone call, hand-held phone
And hands-free phone, wherein the corresponding value of each mode of operation using mobile phone is sequentially reduced, and is positive real number.
Specifically, the computing module, comprising:
Identify submodule, any bad steering behavior for driving driver in stroke, and unit of account mileage for identification
Described in bad steering behavior frequency；
Second generates submodule, for generating the driving stroke in institute based on the handling evaluation model pre-established
State the evaluation result under handling evaluation index；
Wherein, the handling evaluation model isP (u) indicates that driving stroke exists
Evaluation result under the handling evaluation index, u indicate the frequency of bad steering behavior in unit mileage；θ1It is default
Unit mileage in bad steering behavior frequency lower limit value, θ2For bad steering behavior in preset unit mileage
The upper limit value of frequency.
The bad steering behavior includes anxious acceleration behavior, anxious deceleration behavior and zig zag behavior.
Wherein, the identification submodule, comprising:
First acquisition submodule is driven described in acquisition for acquiring any GPS data for driving stroke, and from GPS data
Sail the GPS bearing data of each position in stroke；
Determine submodule, the azimuth for adding up from the initial position for driving stroke or the driving stroke
The corresponding position of variable quantity zero setting starts, successively calculate the azimuthal variation amount between every adjacent position, and to being calculated
Azimuthal variation amount adds up, until cumulative azimuthal variation amount meets following Rule of judgment, it is determined that take a sudden turn
Behavior, and the azimuthal variation amount zero setting that will be added up；
Wherein, the Rule of judgment isΔ bearing expression is driven
The azimuthal variation amount to add up in stroke is sailed, thres indicates the preset bend curvature for being identified as zig zag behavior, and 0 <
Thres≤90.
Specifically, the computing module, comprising:
Second identifies submodule, for identification any jogging period driven in stroke；
First computational submodule, for the jogging time started and jogging end time according to the jogging period, meter
Calculation is persistently walked or drive slowly the time；
Second computational submodule, for calculating the persistently jogging time in the total duration for driving stroke
Accounting, as jogging time accounting；
Third generates submodule, for generating the driving stroke described based on the road condition evaluation model pre-established
Evaluation result under evaluating pavement condition index；
Wherein, the road condition evaluation model isP (u) indicates to drive stroke in institute
The evaluation result under evaluating pavement condition index is stated, u indicates the jogging time accounting, θ1For the lower limit of preset jogging time accounting
Value, θ2For the upper limit value of preset jogging time accounting.
Wherein, the second identification submodule, comprising:
First determines submodule, for will meet time of the average speed less than preset first threshold value in any driving stroke
Section is determined as the period of walking or drive slowly.
Specifically, the computing module, comprising:
Submodule is divided, for using preset duration as the duration of unit sub-line journey, any driving stroke to be in turn divided into
Several sub-line journeys；
Second determines submodule, for determining that default driving period belonging to each sub-line journey, each driving period have
Preset value；
4th generates submodule, for generating the driving stroke described based on the fatigue evaluation model pre-established
Evaluation result under fatigue evaluation index；
Wherein, the fatigue evaluation model is
R indicates to drive evaluation result of the stroke under the fatigue evaluation index, R (tn) indicate that n-th of sub-line journey is commented in the fatigue
Evaluation result under valence index, λ1... λ n respectively indicates the value that the period is driven belonging to each sub-line journey, and λ1< λ2< ... <
λn, t1...tnRespectively indicate λ1... the duration of period, k are driven belonging to the corresponding sub-line journey of λ n1...knRespectively in λ1...λ
When the corresponding sub-line journey of n belongs to the default fatigue driving period driven in the period, value is greater than 1, and otherwise value is equal to 1.
Wherein, the driving period includes fatigue driving period and normal driving period, and the fatigue driving period includes
Morning peak period, evening peak period and night-time hours.
Specifically, the computing module, comprising:
Second acquisition submodule, for obtaining driving over the speed limit the time in any driving stroke；
Third computational submodule described drives over the speed limit the time in the total duration for driving stroke for calculating
Accounting, as overspeed time accounting；
5th generates submodule, for generating the driving stroke in institute based on the vigilant evaluation model pre-established
State the evaluation result under vigilant evaluation index；
Wherein, the vigilant evaluation model isP (u) indicates that driving stroke exists
Evaluation result under the vigilance evaluation index, u indicate the overspeed time accounting, θ1For preset overspeed time accounting
Lower limit value, θ2For the upper limit value of preset overspeed time accounting.
Specifically, described device, further includes:
Second extracting sub-module is driven described in extraction for acquiring the GPS data for driving stroke, and from the GPS data
Sail the driving data of each road in stroke；
Classification submodule, for using the driving data of each road in the driving stroke as the defeated of default classifier
Enter, after the processing of the classifier, determines the road type of each road；
Third determines submodule, for the road type based on each road, determine on each road when driving over the speed limit
Between.
Wherein, the driving data of each road includes poor velocity standard on each road, speed mean value and unit mileage
Bend number.
Wherein, the determining module, comprising:
4th determines submodule, for the relationship based on the evaluation result under accident occurrence probability and each index, determines
The weight coefficient of the evaluation result of each index；
Submodule is weighted, for being weighted and averaged to evaluation result of the driving stroke under each index, is obtained
The final appraisal results for driving stroke.
Specifically, the described 4th determines submodule, comprising:
5th determines submodule, for obtaining the driving stroke that evaluation is completed in advance, and according in each driving stroke
Whether accident occurs, determines each classification for driving stroke, the classification includes first category and second category, the first kind
Accident Biao Shi not occur, accident does not occur for the second category expression；
Third acquisition submodule, the driving of driving stroke and the second category for obtaining the first category respectively
Evaluation result of the stroke under same index, respectively as first category data and second category data；
Third extracting sub-module is mentioned for carrying out feature to the first category data and the second category data respectively
It takes, and cosine similarity calculating is carried out to the feature extracted, obtain the first category data and the second category data
Characteristic similarity value；
6th determines submodule, for determining the weight coefficient of the index according to the characteristic similarity value, wherein
The weight coefficient and the characteristic similarity value have inversely prroportional relationship.
Driving stroke evaluating apparatus provided by the invention can be realized following functions: according to the driving behavior number of driver
According to generation drives the index of stroke for evaluating.Calculate evaluation result of any driving stroke under each index.Based on described
Evaluation result of the stroke under each index is driven, determines the final appraisal results for driving stroke.The present invention can be preparatory
It generates for evaluating the index for driving stroke, and evaluates respectively from the angle of each index stroke is driven, finally obtain
Drive the final appraisal results of stroke.Compared with prior art, the present invention is realized to the Assessment for driving stroke, simultaneously
Evaluation result is more accurate.
For device embodiment, since it corresponds essentially to embodiment of the method, so related place is referring to method reality
Apply the part explanation of example.The apparatus embodiments described above are merely exemplary, wherein described be used as separation unit
The unit of explanation may or may not be physically separated, and component shown as a unit can be or can also be with
It is not physical unit, it can it is in one place, or may be distributed over multiple network units.It can be according to actual
It needs that some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not
In the case where making the creative labor, it can understand and implement.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
It is provided for the embodiments of the invention a kind of driving stroke evaluation method above and device is described in detail, this
Apply that a specific example illustrates the principle and implementation of the invention in text, the explanation of above example is only intended to
It facilitates the understanding of the method and its core concept of the invention；At the same time, for those skilled in the art, think of according to the present invention
Think, there will be changes in the specific implementation manner and application range, in conclusion the content of the present specification should not be construed as pair
Limitation of the invention.
Claims (20)
1. a kind of driving stroke evaluation method, which is characterized in that the described method includes:
According to the driving behavior data of driver, generate for evaluating the index for driving stroke；The index includes with next
Or multiple indexs: focus evaluation index, handling evaluation index, evaluating pavement condition index, fatigue evaluation index, vigilance evaluation
Index；
Calculate evaluation result of any driving stroke under each index；Evaluation based on the driving stroke under each index
As a result, determining the final appraisal results for driving stroke；
Calculate evaluation result of any driving stroke under the focus evaluation index, comprising:
The data that driver in any driving stroke uses mobile phone are collected, the driver includes using hand using the data of mobile phone
The mode of operation of machine and corresponding mode of operation duration, each mode of operation using mobile phone have preset value；
Based on the focus evaluation model pre-established, the evaluation knot for driving stroke under the focus evaluation index is generated
Fruit；The focus evaluation model isR indicates to drive row
Evaluation result of the journey under the focus evaluation index, λ1... it is corresponding that λ n respectively indicates each mode of operation using mobile phone
Value, t1...tnRespectively indicate λ1... the λ n corresponding mode of operation duration；
Calculate evaluation result of any driving stroke under the handling evaluation index, comprising:
Identify the bad steering behavior of driver in any driving stroke, and bad steering behavior described in unit of account mileage
Frequency；Based on the handling evaluation model pre-established, the driving stroke is generated under the handling evaluation index
Evaluation result；The handling evaluation model isP (u) indicates to drive stroke in institute
The evaluation result under handling evaluation index is stated, u indicates the frequency of bad steering behavior in unit mileage；θ1It is preset
The lower limit value of the frequency of bad steering behavior, θ in unit mileage2For the hair of bad steering behavior in preset unit mileage
The upper limit value of raw number；
Calculate evaluation result of any driving stroke under the evaluating pavement condition index, comprising:
Identify the jogging period in any driving stroke；Terminated according to the jogging time started of the jogging period and jogging
Time, calculating are persistently walked or drive slowly the time；Accounting of the persistently jogging time in the total duration for driving stroke is calculated,
As jogging time accounting；Based on the road condition evaluation model pre-established, generates the driving stroke and refer in the evaluating pavement condition
Evaluation result under mark；The road condition evaluation model isP (u) indicates that driving stroke exists
Evaluation result under the evaluating pavement condition index, u indicate the jogging time accounting, θ1For under preset jogging time accounting
Limit value, θ2For the upper limit value of preset jogging time accounting；
Calculate evaluation result of any driving stroke under the fatigue evaluation index, comprising:
Using preset duration as the duration of unit sub-line journey, any driving stroke is in turn divided into several sub-line journeys；It determines each
Default driving period belonging to sub-line journey, each driving period have preset value；Based on the fatigue evaluation mould pre-established
Type generates the evaluation result for driving stroke under the fatigue evaluation index；The fatigue evaluation model isR indicates that drive stroke refers in the fatigue evaluation
Evaluation result under mark, R (tn) indicate evaluation result of n-th of sub-line journey under the fatigue evaluation index, λ1... λ n difference
Indicate the value that the period is driven belonging to each sub-line journey, and λ1< λ2< ... < λn, t1...tnRespectively indicate λ1... λ n is corresponding
The duration of period, k are driven belonging to sub-line journey1...knRespectively in λ1... the corresponding sub-line journey of λ n belongs to the default driving period
In the fatigue driving period when, value be greater than 1, otherwise value be equal to 1；
Calculate evaluation result of any driving stroke under the vigilant evaluation index, comprising:
Obtain driving over the speed limit the time in any driving stroke；Time of driving over the speed limit described in calculating always holding in the stroke that drives
Accounting in the continuous time, as overspeed time accounting；Based on the vigilant evaluation model pre-established, the driving stroke is generated
Evaluation result under the vigilant evaluation index；It is described vigilance evaluation model be
P (u) indicates to drive evaluation result of the stroke under the vigilant evaluation index, the u expression overspeed time accounting, θ1It is pre-
If overspeed time accounting lower limit value, θ2For the upper limit value of preset overspeed time accounting.
2. driving stroke evaluation method according to claim 1, which is characterized in that the mode of operation packet using mobile phone
It includes and sends short message, check short message, make a phone call, holding phone and hands-free phone, wherein each mode of operation using mobile phone point
Not corresponding value is sequentially reduced, and is positive real number.
3. driving stroke evaluation method according to claim 1, which is characterized in that the bad steering behavior includes anxious adds
Fast behavior, anxious deceleration behavior and zig zag behavior.
4. driving stroke evaluation method according to claim 3, which is characterized in that driver in any driving stroke of identification
Zig zag behavior, comprising:
Acquire it is any drive stroke GPS data, and from obtained in GPS data it is described driving stroke in each position the side GPS
Azimuth data；
The corresponding position of azimuthal variation amount zero setting added up from the initial position for driving stroke or the driving stroke
Beginning is set, the azimuthal variation amount between every adjacent position is successively calculated, and the azimuthal variation amount being calculated is carried out tired
Add, until cumulative azimuthal variation amount meets following Rule of judgment, it is determined that zig zag behavior, and the orientation that will be added up occurs
Angle variable quantity zero setting；
Wherein, the Rule of judgment isΔ bearing indicates to drive stroke
In the azimuthal variation amount that adds up, thres indicates the preset bend curvature for being identified as zig zag behavior, and 0 < thres≤
90。
5. driving stroke evaluation method according to claim 1, which is characterized in that in any driving stroke of identification
It walks or drive slowly the period, comprising:
It is determined as the period of walking or drive slowly less than the period of preset first threshold value for average speed is met in any driving stroke.
6. driving stroke evaluation method according to claim 1, which is characterized in that the driving period includes fatigue driving
Period and normal driving period, the fatigue driving period include morning peak period, evening peak period and night-time hours.
7. driving stroke evaluation method according to claim 1, which is characterized in that described to obtain in any driving stroke
It drives over the speed limit before the time, further includes:
Acquisition drive stroke GPS data, and from extracted in the GPS data it is described driving stroke in each road driving number
According to；
Using the driving data of each road in the driving stroke as the input of default classifier, by the place of the classifier
After reason, the road type of each road is determined；
Based on the road type of each road, driving over the speed limit the time on each road is determined.
8. driving stroke evaluation method according to claim 7, which is characterized in that the driving data of each road includes each
Velocity standard on road is poor, speed mean value and unit mileage bend number.
9. driving stroke evaluation method according to claim 1, which is characterized in that described to be based on the driving stroke each
Evaluation result under a index determines the final appraisal results for driving stroke, comprising:
Based on the relationship of the evaluation result under accident occurrence probability and each index, the weight of the evaluation result of each index is determined
Coefficient；
Evaluation result of the driving stroke under each index is weighted and averaged, the most final review for driving stroke is obtained
Valence result.
10. driving stroke evaluation method according to claim 9, which is characterized in that it is described based on accident occurrence probability with
The relationship of evaluation result under each index determines the weight coefficient of the evaluation result of each index, comprising:
The driving stroke that evaluation is completed is obtained in advance, and according to whether accident occurring in each driving stroke, determines each drive
Sail the classification of stroke, the classification includes first category and second category, and accident occurs for the first category expression, and described second
Accident does not occur for classification expression；
Evaluation of the driving stroke of the driving stroke and the second category that obtain the first category respectively under same index
As a result, respectively as first category data and second category data；
Feature extraction is carried out to the first category data and the second category data respectively, and the feature extracted is carried out
Cosine similarity calculates, and obtains the characteristic similarity value of the first category data Yu the second category data；
According to the characteristic similarity value, the weight coefficient of the index is determined, wherein the weight coefficient and the feature phase
There is inversely prroportional relationship like angle value.
11. a kind of driving stroke evaluating apparatus, which is characterized in that described device includes:
Generation module is generated for the driving behavior data according to driver for evaluating the index for driving stroke；The index
Including following one or more indexs: focus evaluation index, handling evaluation index, evaluating pavement condition index, fatigue evaluation refer to
Mark, vigilant evaluation index；
Computing module, for calculating evaluation result of any driving stroke under each index；
Determining module determines the driving stroke most for the evaluation result based on the driving stroke under each index
Whole evaluation result；
The computing module, comprising: first collects submodule, and the number of mobile phone is used for collecting driver in any driving stroke
According to, the driver using mode of operation and the corresponding mode of operation duration that the data of mobile phone include using mobile phone,
Each mode of operation using mobile phone has preset value；First generates submodule, for based on the focus pre-established
Evaluation model generates the evaluation result for driving stroke under the focus evaluation index；The focus evaluation model
ForR indicates that drive stroke refers in focus evaluation
Evaluation result under mark, λ1... λ n respectively indicates the corresponding value of each mode of operation using mobile phone, t1...tnIt respectively indicates
λ1... the λ n corresponding mode of operation duration；
Alternatively, the computing module, comprising: identify submodule, for identification any bad steering for driving driver in stroke
Behavior, and the frequency of bad steering behavior described in unit of account mileage；Second generates submodule, builds in advance for being based on
Vertical handling evaluation model generates the evaluation result for driving stroke under the handling evaluation index；The manipulation
Property evaluation model isP (u) indicates to drive stroke under the handling evaluation index
Evaluation result, u indicate unit mileage in bad steering behavior frequency；θ1For bad steering in preset unit mileage
The lower limit value of the frequency of behavior, θ2For the upper limit value of the frequency of bad steering behavior in preset unit mileage；
Alternatively, the computing module, comprising: second identifies submodule, for identification any jogging period driven in stroke；
First computational submodule, for the jogging time started and jogging end time according to the jogging period, calculating is persistently delayed
The row time；Second computational submodule, for calculating the persistently jogging time in the total duration for driving stroke
Accounting, as jogging time accounting；Third generates submodule, for based on the road condition evaluation model pre-established, described in generation
Drive evaluation result of the stroke under the evaluating pavement condition index；The road condition evaluation model isP (u) indicates to drive evaluation result of the stroke under the evaluating pavement condition index, u expression
The jogging time accounting, θ1For the lower limit value of preset jogging time accounting, θ2For the upper limit of preset jogging time accounting
Value；
Alternatively, the computing module, comprising: submodule is divided, it, will be any for using preset duration as the duration of unit sub-line journey
It drives stroke and is in turn divided into several sub-line journeys；Second determines submodule, for determining default driving belonging to each sub-line journey
Period, each driving period have preset value；4th generates submodule, for based on the fatigue evaluation mould pre-established
Type generates the evaluation result for driving stroke under the fatigue evaluation index；The fatigue evaluation model isR indicates that drive stroke refers in the fatigue evaluation
Evaluation result under mark, R (tn) indicate evaluation result of n-th of sub-line journey under the fatigue evaluation index, λ1... λ n difference
Indicate the value that the period is driven belonging to each sub-line journey, and λ1< λ2< ... < λn, t1...tnRespectively indicate λ1... λ n is corresponding
The duration of period, k are driven belonging to sub-line journey1...knRespectively in λ1... the corresponding sub-line journey of λ n belongs to the default driving period
In the fatigue driving period when, value be greater than 1, otherwise value be equal to 1；
Alternatively, the computing module, comprising: the second acquisition submodule, for obtain it is any driving stroke in when driving over the speed limit
Between；Third computational submodule, for calculating the accounting of the time in the total duration for driving stroke of driving over the speed limit,
As overspeed time accounting；5th generates submodule, for generating the driving based on the vigilant evaluation model pre-established
Evaluation result of the stroke under the vigilant evaluation index；It is described vigilance evaluation model beP (u) indicates to drive evaluation result of the stroke under the vigilant evaluation index, u
Indicate the overspeed time accounting, θ1For the lower limit value of preset overspeed time accounting, θ2For the upper of preset overspeed time accounting
Limit value.
12. driving stroke evaluating apparatus according to claim 11, which is characterized in that the mode of operation using mobile phone
Including sending short message, checking short message, make a phone call, hand-held phone and hands-free phone, wherein each mode of operation using mobile phone
Corresponding value is sequentially reduced, and is positive real number.
13. driving stroke evaluating apparatus according to claim 11, which is characterized in that the bad steering behavior includes anxious
Acceleration behavior, anxious deceleration behavior and zig zag behavior.
14. driving stroke evaluating apparatus according to claim 13, which is characterized in that the identification submodule, comprising:
First acquisition submodule for acquiring any GPS data for driving stroke, and obtains the driving row from GPS data
The GPS bearing data of each position in journey；
Determine submodule, the azimuthal variation for adding up from the initial position for driving stroke or the driving stroke
The corresponding position of amount zero setting starts, and successively calculates the azimuthal variation amount between every adjacent position, and to the orientation being calculated
Angle variable quantity adds up, until cumulative azimuthal variation amount meets following Rule of judgment, it is determined that zig zag behavior occurs,
And the azimuthal variation amount zero setting that will be added up；
Wherein, the Rule of judgment isΔ bearing indicates to drive stroke
In the azimuthal variation amount that adds up, thres indicates the preset bend curvature for being identified as zig zag behavior, and 0 < thres≤
90。
15. driving stroke evaluating apparatus according to claim 11, which is characterized in that the second identification submodule, packet
It includes:
First determines submodule, true less than the period of preset first threshold value for will meet average speed in any driving stroke
It is set to the jogging period.
16. driving stroke evaluating apparatus according to claim 11, which is characterized in that the driving period drives including fatigue
It sails period and normal driving period, the fatigue driving period includes morning peak period, evening peak period and night-time hours.
17. driving stroke evaluating apparatus according to claim 11, which is characterized in that described device, further includes:
Second extracting sub-module for acquiring the GPS data for driving stroke, and extracts the driving row from the GPS data
The driving data of each road in journey；
Classification submodule, for using the driving data of each road in the driving stroke as the input of default classifier, warp
After crossing the processing of the classifier, the road type of each road is determined；
Third determines submodule, for the road type based on each road, determines driving over the speed limit the time on each road.
18. driving stroke evaluating apparatus according to claim 17, which is characterized in that the driving data of each road includes
Velocity standard on each road is poor, speed mean value and unit mileage bend number.
19. driving stroke evaluating apparatus according to claim 11, which is characterized in that the determining module, comprising:
4th determines submodule, for the relationship based on the evaluation result under accident occurrence probability and each index, determines each
The weight coefficient of the evaluation result of index；
Submodule is weighted, for being weighted and averaged to evaluation result of the driving stroke under each index, is obtained described
Drive the final appraisal results of stroke.
20. driving stroke evaluating apparatus according to claim 19, which is characterized in that the described 4th determines submodule, packet
It includes:
5th determines submodule, for obtaining the driving stroke that evaluation is completed in advance, and according in each driving stroke whether
Generation accident determines each classification for driving stroke, and the classification includes first category and second category, the first category table
Show and accident occurs, accident does not occur for the second category expression；
Third acquisition submodule, the driving stroke of driving stroke and the second category for obtaining the first category respectively
Evaluation result under same index, respectively as first category data and second category data；
Third extracting sub-module, for carrying out feature extraction to the first category data and the second category data respectively,
And cosine similarity calculating is carried out to the feature extracted, obtain the spy of the first category data Yu the second category data
Levy similarity value；
6th determines submodule, for determining the weight coefficient of the index, wherein described according to the characteristic similarity value
Weight coefficient and the characteristic similarity value have inversely prroportional relationship.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611192641.7A CN106710144B (en) | 2016-12-21 | 2016-12-21 | A kind of driving stroke evaluation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611192641.7A CN106710144B (en) | 2016-12-21 | 2016-12-21 | A kind of driving stroke evaluation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106710144A CN106710144A (en) | 2017-05-24 |
| CN106710144B true CN106710144B (en) | 2019-10-18 |
Family
ID=58938558
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611192641.7A Active CN106710144B (en) | 2016-12-21 | 2016-12-21 | A kind of driving stroke evaluation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106710144B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6977490B2 (en) * | 2017-11-06 | 2021-12-08 | トヨタ自動車株式会社 | Information processing equipment, information processing system and information processing method |
| CN107918826B (en) * | 2017-11-13 | 2021-09-28 | 北京航空航天大学 | Driver evaluation and scheduling method for driving environment perception |
| CN110389577B (en) * | 2018-04-17 | 2022-04-01 | 北京三快在线科技有限公司 | Method and device for determining driving style |
| CN108694619A (en) * | 2018-06-20 | 2018-10-23 | 新华网股份有限公司 | Box office receipts prediction technique and system |
| CN108694814A (en) * | 2018-07-11 | 2018-10-23 | 中国医学科学院生物医学工程研究所 | Fatigue driving method for early warning, device, equipment and storage medium |
| CN109034589A (en) * | 2018-07-18 | 2018-12-18 | 南斗六星系统集成有限公司 | A kind of driving behavior evaluation method and system based on big data analysis |
| CN108961680B (en) * | 2018-07-20 | 2020-09-22 | 武汉理工大学 | Performance detection system and method of drunk driving and fatigue driving discrimination system |
| CN109448377B (en) * | 2018-11-29 | 2020-07-28 | 交通运输部公路科学研究所 | Method for evaluating vehicle driving safety by using satellite positioning data |
| CN109800984B (en) * | 2019-01-16 | 2024-03-01 | 平安科技（深圳）有限公司 | Driving level evaluation method, driving level evaluation device, computer device, and storage medium |
| CN110533909B (en) * | 2019-09-10 | 2020-11-06 | 重庆大学 | Driving behavior analysis method and system based on traffic environment |
| CN110615001B (en) * | 2019-09-27 | 2021-04-27 | 汉纳森（厦门）数据股份有限公司 | Driving safety reminding method, device and medium based on CAN data |
| JP2021077128A (en) * | 2019-11-11 | 2021-05-20 | 日本電気株式会社 | Abnormality detection device, abnormality detection method, and program |
| CN116542830B (en) * | 2023-07-06 | 2024-03-15 | 广州市德赛西威智慧交通技术有限公司 | Intelligent judging method and device based on multiple parameters |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103871122A (en) * | 2014-03-11 | 2014-06-18 | 深圳市朗仁科技有限公司 | Driving behavior analysis method and driving behavior analysis system |
| CN104599346A (en) * | 2013-12-11 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Driving behavior evaluation method and driving behavior evaluation apparatus |
| CN105261225A (en) * | 2015-09-30 | 2016-01-20 | 肖建辉 | Monitoring system for improving driving behavioral habits |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5573577B2 (en) * | 2010-10-12 | 2014-08-20 | トヨタ自動車株式会社 | Driving evaluation system and in-vehicle device |
-
2016
- 2016-12-21 CN CN201611192641.7A patent/CN106710144B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104599346A (en) * | 2013-12-11 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Driving behavior evaluation method and driving behavior evaluation apparatus |
| CN103871122A (en) * | 2014-03-11 | 2014-06-18 | 深圳市朗仁科技有限公司 | Driving behavior analysis method and driving behavior analysis system |
| CN105261225A (en) * | 2015-09-30 | 2016-01-20 | 肖建辉 | Monitoring system for improving driving behavioral habits |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106710144A (en) | 2017-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106710144B (en) | A kind of driving stroke evaluation method and device | |
| CN107161153B (en) | A kind of driving behavior methods of marking and device | |
| CN103838772B (en) | A kind of multi-source traffic data fusion method | |
| Ito et al. | Low cost carrier growth in the US airline industry: past, present, and future | |
| CN105809976A (en) | Vehicle running state monitoring method and mobile terminal | |
| DE102010062252A1 (en) | System for calculating the fuel quantity of a travel route and method therefor | |
| Elvik | Effects of mobile phone use on accident risk: Problems of meta-analysis when studies are few and bad | |
| CN106056162A (en) | A traffic safety credit scoring method based on GPS track and traffic law-violation records | |
| DE112009004311T5 (en) | Method and apparatus for providing a navigation summary | |
| CN107257419B (en) | One kind quantifying estimation method based on Bayesian analysis interpersonal relationships | |
| CN107230350A (en) | A kind of urban transportation amount acquisition methods based on bayonet socket Yu mobile phone flow call bill data | |
| CN110378397A (en) | A kind of driving style recognition methods and device | |
| CN107918762A (en) | A kind of highway drops thing rapid detection system and method | |
| DE112018003180T5 (en) | RISK INFORMATION COLLECTOR | |
| CN108898831A (en) | Section condition evaluation method and system based on road high definition bayonet data | |
| CN112434260A (en) | Road traffic state detection method and device, storage medium and terminal | |
| CN103761870A (en) | System for counting bus loads | |
| CN109902712A (en) | Transmission line of electricity bird repellent method based on unmanned plane inspection | |
| CN108665383A (en) | Big data management method of travelling and system | |
| CN110389577A (en) | A kind of method and device of determining driving style | |
| CN109166336B (en) | Real-time road condition information acquisition and pushing method based on block chain technology | |
| CN108010357A (en) | Speed-limiting messages verification/statistical method, apparatus and system | |
| DE112019004005T5 (en) | DRIVE EVALUATION DEVICE | |
| CN103400159A (en) | Target classification identifying method in quick mobile context and classifier obtaining method for target classification and identification in quick mobile context | |
| CN110867075A (en) | Method for evaluating influence of road speed meter on reaction behavior of driver under rainy condition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |