CN110254435A - A kind of driving style recognition methods - Google Patents
A kind of driving style recognition methods Download PDFInfo
- Publication number
- CN110254435A CN110254435A CN201910580087.7A CN201910580087A CN110254435A CN 110254435 A CN110254435 A CN 110254435A CN 201910580087 A CN201910580087 A CN 201910580087A CN 110254435 A CN110254435 A CN 110254435A
- Authority
- CN
- China
- Prior art keywords
- driving style
- opposite
- data
- recognition methods
- road type
- 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.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
- B60W40/09—Driving style or behaviour
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2720/00—Output or target parameters relating to overall vehicle dynamics
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention relates to a kind of driving style recognition methods, method includes the following steps: step S1: acquiring wheelpath data using sensor；Step S2: being based on wheelpath data, and combining geographic information data obtain opposite hypervelocity behavior；Step S3: based on opposite hypervelocity behavior, driving style parameter is calculated；Step S4: being based on driving style parameter, identifies driving style using clustering algorithm.Compared with prior art, the data precision requirement of driving style identification is relaxed, low-frequency data is based on, advantageously reduces the data acquisition cost of driving style identification；Extend the application that low frequency wheelpath data probe into aspect in driving style；Traffic safety can be improved.
Description
Technical field
The present invention relates to driving styles to identify field, more particularly, to a kind of driving style recognition methods.
Background technique
Driving style is the whole evaluation index for characterizing the intrinsic drive manner of driver.Driving style is studied mainly from section
Angularly, driver is divided for energy driving, traffic safety.It is existing studies have shown that the style of driver and traffic safety are close
Cut association: radical driving style driver there are frequent lane-change, anxious acceleration and deceleration, closely with the bad steerings behavior such as speed, Yi Zeng
Add contingency occurrence probability.It is studied by driving style, the bad steering style of driver's drive routine is detected or fed back,
The supervision and education to driver's driving behavior can be achieved, and assist with corresponding measure to promote travel safety.
Data towards driving style research mainly include Subjective questionnaire on negative impact and two kinds of objective driving data, and with high-precision
Based on the driving behavior data of degree, such as the high frequency GPS wheelpath data that frequency acquisition is 1Hz, 10Hz, from acceleration plus add
The microcosmic angles such as speed carry out driving style assessment.However, the automobile services such as current timesharing lease provide the driving of enterprise's acquisition
Track data mainly acquires low-frequency wheelpath data (30~120s of GPS gathers interval) for the purpose of vehicle location.Such as
What feature based on low frequency wheelpath data, constructs driving style parameter and carries out driving style identification and still lack reasonable side
Case.
Summary of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide a kind of driving styles to identify
Method.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of driving style recognition methods, method includes the following steps:
Step S1: wheelpath data are acquired using sensor；
Step S2: being based on wheelpath data, and combining geographic information data obtain opposite hypervelocity behavior；
Step S3: based on opposite hypervelocity behavior, driving style parameter is calculated；
Step S4: being based on driving style parameter, identifies driving style using clustering algorithm.
The wheelpath data be GPS wheelpath data, the GPS wheelpath data be acquisition interval 30~
The low frequency wheelpath data of 60s.
The step S2 includes:
Step S21: wheelpath data are matched with geographic information data, obtain the road type of each tracing point；
Step S22: behavior of the speed of service higher than 99 quantiles of the road type operating speed of each tracing point is extracted, is obtained
To opposite hypervelocity behavior.
The step S3 includes:
Step S31: based on opposite hypervelocity behavior, opposite overspeed time ratio is calculated；
Step S32: it based on opposite overspeed time ratio, is calculated averagely with respect to overspeed time ratio；
Step S33: based on average opposite overspeed time ratio, the coefficient of variation is calculated；
Step S34: based on coefficient of variation peace with respect to overspeed time ratio, driving style parameter is calculated.
For once going on a journey, opposite overspeed time ratio of the driver on road type i is with respect to overspeed time ratio
tI_speeding%Are as follows:
Wherein, i is road type (i=1,2 ..., m), ti199 points for being higher than i road type operating speed for the speed of service
The traveling duration of digit, ti2It is higher than the traveling duration of the road type level-one service level minimum speed for the speed of service.
The average opposite overspeed time ratio x of i road type in j driver each time tripji1Are as follows:
Wherein, p is the number of trip road type i, tI_speeding%_lFor the secondary trip road class of l (l=1,2 ..., p)
The opposite overspeed time ratio of type i.
The coefficient of variation x of i road type in j driver each time tripji2Are as follows:
The driving style parameter x of j driverjAre as follows:
xj=(xj11,xj12,…,xji1,xji2,…,xjm1,xjm2)
The clustering algorithm is K-Means clustering algorithm.
Compared with prior art, the invention has the following advantages that
(1) the data precision requirement of driving style identification is relaxed, low-frequency data is based on, advantageously reduces driving style knowledge
Other data acquisition cost.
(2) application that low frequency wheelpath data probe into aspect in driving style is extended.
It (3), can specific aim formulation safety religion for the bad style driver (the radical group of such as driving style) identified
The risk management measures such as project or UBI vehicle insurance scheme are educated, traffic safety can be improved and reduce enterprise operation risk.
(4) only by speed, this vehicle data is calculated, and is reduced the usage amount of sensor and and is reduced
Data traffic expense.
(5) driving style parameter can accurately embody driving style.
Detailed description of the invention
Fig. 1 is flow chart of the invention；
Fig. 2 is K-Means clustering algorithm flow chart；
Fig. 3 is embodiment driving style cluster result.
Specific embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.The present embodiment is with technical solution of the present invention
Premised on implemented, the detailed implementation method and specific operation process are given, but protection scope of the present invention is not limited to
Following embodiments.
Embodiment
The present invention provides a kind of driving style recognition methods based on low frequency wheelpath data, do not depend on height to realize
Accuracy data is the group that can recognize different driving styles, finds potential high risk driver, formulates safety education for specific aim
The risk management measures such as project or UBI vehicle insurance scheme provide foundation.
The present invention provides a kind of driving style recognition methods based on low frequency wheelpath data, comprising the following steps:
Step S1: wheelpath data are acquired using sensor；
Step S2: being based on wheelpath data, and combining geographic information data obtain opposite hypervelocity behavior；
Step S3: based on opposite hypervelocity behavior, driving style parameter is calculated；
Step S4: being based on driving style parameter, identifies driving style using clustering algorithm.
Further, the wheelpath data of acquisition described in step S1 are to acquire vehicle driving by vehicle arrangement sensor
Satellite positioning (Global Positioning System, GPS) track data, wheelpath data be low frequency wheelpath
Data, GPS gathers wheelpath data break are 30~60s.
Further, combining geographic information data described in step S2 obtain opposite hypervelocity behavior, specifically:
(2.1) track data and geography information figure layer Data Matching: based on OpenStreetMap (OSM) and open source route
Planning engine (OSRM, Open Source Routing Machine) technology realizes track data and geography information figure layer data
Matching, obtain the road and traffic management information of each tracing point, including but not limited to road name, road type i (i=1,
2 ..., m) and speed limit；
(2.2) obtain opposite hypervelocity behavior: based on low frequency wheelpath data characteristics, opposite behavior, that is, driver of exceeding the speed limit exists
The speed of service is higher than the behavior of 99 quantiles of the road type operating speed on road type i, based on road locating for each tracing point
Road type i and the opposite hypervelocity behavior of operating speed identification.
Wherein, quantile meaning is as follows: one group of data is sorted from small to large, and calculates accumulative percentile accordingly, certain
The value of data corresponding to one percentile is known as the quantile of this percentile, i.e. the value in the position p% claims p quantile.
Further, driving style parameter is calculated based on opposite hypervelocity behavior in step S3, specifically:
(3.1) based on opposite hypervelocity behavior, define opposite overspeed time ratio: for once going on a journey, driver is in road
Opposite overspeed time ratio t on type iI_speeding%Are as follows:
Wherein, ti1It is higher than the traveling duration of 99 quantiles of i road type operating speed for the speed of service in this time trip,
ti2It is higher than typical rate (such as level-one service level minimum speed) under the unobstructed state of traffic circulation for the speed of service in this time trip
Travel duration；
(3.2) based on opposite overspeed time ratio, following driving style parameter is calculated:
The average opposite overspeed time ratio x of i road type in j driver each time tripji1Are as follows:
Wherein, p is the number of trip road type i, tI_speeding%_lFor the secondary trip road class of l (l=1,2 ..., p)
The opposite overspeed time ratio of type i.
Coefficient of variation x of the i road type with respect to overspeed time ratio in j driver each time tripji2Are as follows:
Coefficient of variation xji2Behavior stability is characterized, i.e., for specific driver, opposite hypervelocity be it is habitual or
Accidental behavior.
The driving style parameter of driver j is xj=(xj11,xj12,…,xji1,xji2,…,xjm1,xjm2)。
Further, clustering algorithm is K-Means clustering algorithm in step S4, is driven based on K-Means clustering algorithm
The identification and classification of style, specifically:
(4.1) driving style assessment, specific steps are carried out based on K-Means clustering algorithm are as follows:
(4.1a) obtains driver's set X=(x1,x2,…,xj,…,xn)T, and data are normalized, it obtains
Treated matrix X'=(x'1,x'2,…,x'j,…,x'n)T；
(4.1b) initializes cluster centre number k=2；
(4.1c) is using K-Means clustering algorithm to x'1,x'2,…,x'j,…,x'nCarry out driving style classification；
(4.1d) calculates the mean profile coefficient s under k clusterk, calculation method are as follows:
Wherein, a is inter- object distance average value, and b is the average value at a distance from nearest neighbor class；
(4.1e) compares skWith sk-1Size, if sk>sk-1, then enable k=k+1, repeat step (4.1c), (4.1d) and
(4.1e)；If sk<sk-1Then stop recycling；
(4.1f) exports the driving style classification results under preferable clustering number k-1 and cluster numbers k-1；
(4.2) driving style classification results are based on, different driving styles are identified.
It is illustrated below with reference to a specific alternative embodiment:
(1) GPS data of vehicle is acquired by vehicle arrangement sensor, is divided into 30s between GPS gathers, is obtained 235 altogether
The city expressway driving data of driver, accumulative 19348km.
(2) combining geographic information data extract opposite hypervelocity behavior, the specific steps are as follows:
(1) track data and geography information figure layer Data Matching: based on OpenStreetMap (OSM) and open source route rule
It draws engine (OSRM, Open Source Routing Machine) technology and realizes track data and geography information figure layer data
Matching obtains the road and traffic management information of each tracing point, including road name, road type i (i=1,2 ..., m) and limit
Speed only includes road type i and speed limit, city expressway i=2 in this example；
(2) obtain opposite hypervelocity behavior: based on low frequency wheelpath data characteristics, opposite behavior, that is, driver of exceeding the speed limit is in road
The speed of service is higher than the behavior of 99 quantiles of the road type operating speed on the type i of road, based on road locating for each tracing point
Type i and the opposite hypervelocity behavior of operating speed identification, 99 quantile v of operating speed21=90km/h.
(3) it proposes and calculates driving style parameter, specific steps are as follows:
(1) based on opposite hypervelocity behavior, opposite overspeed time ratio is defined are as follows: driver is opposite on road type i
Hypervelocity behavior ratio t2_speeding%Are as follows: the speed of service is higher than 99 quantile v of city expressway operating speed21=90km/h row
Sail duration t21It is higher than level-one service level minimum speed v with the speed of service22=73km/h travels duration t22Ratio, it may be assumed that
(2) it based on opposite overspeed time ratio, proposes and calculates following driving style parameter:
Average opposite overspeed time ratio x of the j driver in city expresswayj21The hypervelocity tendency for characterizing driver, indicates
Are as follows:
Coefficient of variation x of the j driver in city expresswayj22Are as follows:
The driving style parameter for remembering driver j is xj=(xj21,xj22)。
(4) driving style assessment is carried out based on K-Means clustering algorithm, as shown in Fig. 2, specific steps are as follows:
(1) driver's set X=(x is obtained1,x2,…,xj,…,x235)T, and data are normalized, it obtains
Treated matrix X'=(x'1,x'2,…,x'j,…,x'235)T；
(2) cluster centre number k=2 is initialized；
(3) using K-Means clustering algorithm to x'1,x'2,…,x'j,…,x'235Carry out driving style classification；
(4) the mean profile coefficient s under k cluster is calculatedk, calculation method are as follows:
Wherein, a is inter- object distance average value, and b is the average value at a distance from nearest neighbor class；
(5) compare skWith sk-1Size, if sk>sk-1, then k=k+1 is enabled, step (3), (4) and (5) are repeated；If sk<sk-1
Then stop recycling；
(6) the driving style classification results under preferable clustering number k-1 and cluster numbers k-1 are exported, are most preferably gathered in the present embodiment
Class number is 3, and mean profile coefficient is 0.43 under preferable clustering number；
(8) driving style classification results are based on, different driving styles are identified: in the present embodiment, according to relatively super
Driver's driving style is divided into careful, warm by the coefficient of variation of fast time scale and opposite overspeed time ratio of going on a journey for each time
With radical three classes, as shown in figure 3, each style accounts for 54.04%, 36.60%, the 9.36% of sample driver, feature respectively are as follows:
1) with caution: the coefficient of variation of opposite overspeed time ratio minimum and each time opposite overspeed time ratio of going on a journey is higher,
That is hypervelocity tendency is low, and opposite hypervelocity is accidental behavior；
2) mild: the coefficient of variation of opposite overspeed time ratio lower and each time opposite overspeed time ratio of going on a journey is higher,
That is hypervelocity tendency is lower, and opposite hypervelocity is accidental behavior；
3) radical: the coefficient of variation of opposite overspeed time ratio highest and opposite overspeed time ratio of going on a journey for each time is lower,
That is hypervelocity tendency is higher, and opposite hypervelocity occurs relatively stablely in each trip.
Through the above steps, complete based on low-frequency data driving style identification, compared with the prior art in, driving style
Assessment relies on high-precision track data and extracts microcosmic driving behavior, and above-mentioned steps relax the data essence of driving style assessment
Degree requires, and proposes based on low-frequency data driving style parameter and carries out the reasonable plan of driving style identification, be conducive to drop
The data acquisition cost of low driving style identification, and extend low frequency wheelpath data and probe into answering for aspect in driving style
With.For the bad style driver (the radical group of such as driving style) identified, can specific aim formulate safety education project or
The risk management measures such as UBI vehicle insurance scheme can be improved traffic safety and reduce enterprise operation risk.
Claims (9)
1. a kind of driving style recognition methods, which is characterized in that method includes the following steps:
Step S1: wheelpath data are acquired using sensor；
Step S2: being based on wheelpath data, and combining geographic information data obtain opposite hypervelocity behavior；
Step S3: based on opposite hypervelocity behavior, driving style parameter is calculated；
Step S4: being based on driving style parameter, identifies driving style using clustering algorithm.
2. a kind of driving style recognition methods according to claim 1, which is characterized in that the wheelpath data are
GPS wheelpath data, the GPS wheelpath data are the low frequency wheelpath data of 30~60s of acquisition interval.
3. a kind of driving style recognition methods according to claim 1, which is characterized in that the step S2 includes:
Step S21: wheelpath data are matched with geographic information data, obtain the road type of each tracing point；
Step S22: behavior of the speed of service higher than 99 quantiles of the road type operating speed of each tracing point is extracted, phase is obtained
To hypervelocity behavior.
4. a kind of driving style recognition methods according to claim 1, which is characterized in that the step S3 includes:
Step S31: based on opposite hypervelocity behavior, opposite overspeed time ratio is calculated；
Step S32: it based on opposite overspeed time ratio, is calculated averagely with respect to overspeed time ratio；
Step S33: based on average opposite overspeed time ratio, the coefficient of variation is calculated；
Step S34: based on coefficient of variation peace with respect to overspeed time ratio, driving style parameter is calculated.
5. a kind of driving style recognition methods according to claim 4, which is characterized in that for once going on a journey, driver
Opposite overspeed time ratio t on road type iI_speeding%Are as follows:
Wherein, i is road type (i=1,2 ..., m), ti1It is higher than 99 quantiles of i road type operating speed for the speed of service
Traveling duration, ti2It is higher than the traveling duration of the road type setting speed for the speed of service.
6. a kind of driving style recognition methods according to claim 5, which is characterized in that the road i in j driver each time trip
The average opposite overspeed time ratio x of road typeji1Are as follows:
Wherein, p is the number of trip road type i, tI_speeding%_lFor the secondary trip i road type of l (l=1,2 ..., p)
Opposite overspeed time ratio.
7. a kind of driving style recognition methods according to claim 6, which is characterized in that the road i in j driver each time trip
The coefficient of variation x of road typeji2Are as follows:
8. a kind of driving style recognition methods according to claim 7, which is characterized in that the driving style of j driver is joined
Number xjAre as follows:
xj=(xj11,xj12,…,xji1,xji2,…,xjm1,xjm2)。
9. a kind of driving style recognition methods according to claim 1, which is characterized in that the clustering algorithm is K-
Means clustering algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910580087.7A CN110254435B (en) | 2019-06-28 | 2019-06-28 | Driving style recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910580087.7A CN110254435B (en) | 2019-06-28 | 2019-06-28 | Driving style recognition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110254435A true CN110254435A (en) | 2019-09-20 |
| CN110254435B CN110254435B (en) | 2021-06-04 |
Family
ID=67923212
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910580087.7A Active CN110254435B (en) | 2019-06-28 | 2019-06-28 | Driving style recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110254435B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111259767A (en) * | 2020-01-13 | 2020-06-09 | 厦门大学 | Traffic illegal behavior identification method and system based on traffic data and street view data |
| CN112721949A (en) * | 2021-01-12 | 2021-04-30 | 重庆大学 | Method for evaluating longitudinal driving personification degree of automatic driving vehicle |
| CN113096379A (en) * | 2021-03-03 | 2021-07-09 | 东南大学 | Driving style identification method based on traffic conflict |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100019880A1 (en) * | 2008-07-24 | 2010-01-28 | Gm Global Technology Operations, Inc. | Adaptive vehicle control system with driving style recognition based on traffic sensing |
| CN103914985A (en) * | 2014-04-25 | 2014-07-09 | 大连理工大学 | Method for predicting future speed trajectory of hybrid power bus |
| CN107153916A (en) * | 2017-04-30 | 2017-09-12 | 安徽中科美络信息技术有限公司 | A kind of driving behavior evaluation method clustered based on FCM with BP neural network |
| CN108491859A (en) * | 2018-02-11 | 2018-09-04 | 郭静秋 | The recognition methods of driving behavior heterogeneity feature based on automatic coding machine |
| CN109034181A (en) * | 2018-06-05 | 2018-12-18 | 广州杰赛科技股份有限公司 | The classification method and device of motion track, equipment, storage medium |
| CN109523390A (en) * | 2018-09-28 | 2019-03-26 | 深圳市麦谷科技有限公司 | A method of air control model is established according to running car rule |
| CN109559507A (en) * | 2018-11-16 | 2019-04-02 | 西南交通大学 | A kind of about vehicle excess speed event recognition methods of the net based on history GPS track data |
-
2019
- 2019-06-28 CN CN201910580087.7A patent/CN110254435B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100019880A1 (en) * | 2008-07-24 | 2010-01-28 | Gm Global Technology Operations, Inc. | Adaptive vehicle control system with driving style recognition based on traffic sensing |
| CN103914985A (en) * | 2014-04-25 | 2014-07-09 | 大连理工大学 | Method for predicting future speed trajectory of hybrid power bus |
| CN107153916A (en) * | 2017-04-30 | 2017-09-12 | 安徽中科美络信息技术有限公司 | A kind of driving behavior evaluation method clustered based on FCM with BP neural network |
| CN108491859A (en) * | 2018-02-11 | 2018-09-04 | 郭静秋 | The recognition methods of driving behavior heterogeneity feature based on automatic coding machine |
| CN109034181A (en) * | 2018-06-05 | 2018-12-18 | 广州杰赛科技股份有限公司 | The classification method and device of motion track, equipment, storage medium |
| CN109523390A (en) * | 2018-09-28 | 2019-03-26 | 深圳市麦谷科技有限公司 | A method of air control model is established according to running car rule |
| CN109559507A (en) * | 2018-11-16 | 2019-04-02 | 西南交通大学 | A kind of about vehicle excess speed event recognition methods of the net based on history GPS track data |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111259767A (en) * | 2020-01-13 | 2020-06-09 | 厦门大学 | Traffic illegal behavior identification method and system based on traffic data and street view data |
| CN111259767B (en) * | 2020-01-13 | 2023-04-18 | 厦门大学 | Traffic illegal behavior identification method and system based on traffic data and street view data |
| CN112721949A (en) * | 2021-01-12 | 2021-04-30 | 重庆大学 | Method for evaluating longitudinal driving personification degree of automatic driving vehicle |
| CN112721949B (en) * | 2021-01-12 | 2022-07-12 | 重庆大学 | Method for evaluating longitudinal driving personification degree of automatic driving vehicle |
| CN113096379A (en) * | 2021-03-03 | 2021-07-09 | 东南大学 | Driving style identification method based on traffic conflict |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110254435B (en) | 2021-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11836802B2 (en) | Vehicle operation analytics, feedback, and enhancement | |
| CN104008647B (en) | A kind of road traffic energy consumption quantization method based on motor-driven vehicle going pattern | |
| CN110254435A (en) | A kind of driving style recognition methods | |
| US9109914B2 (en) | Energy efficient routing using an impedance factor | |
| EP2619059B1 (en) | Driver profiling system and method | |
| CN106846806A (en) | Urban highway traffic method for detecting abnormality based on Isolation Forest | |
| CN107145989A (en) | Real-road Driving Cycle construction method based on people's car traffic | |
| CN107038860B (en) | A kind of user's driving behavior methods of marking based on ADAS technology and regression model | |
| CN104916133B (en) | Road altitude information extraction method and system based on traffic track data | |
| CN105788250B (en) | Vehicle drive behavior treating method and apparatus | |
| CN104933863A (en) | Method and system for recognizing abnormal segment of traffic road | |
| CN106956680B (en) | Electric automobile driving behavior recognition analysis method | |
| CN110188797A (en) | A kind of intelligent automobile method for rapidly testing based on Bayes's optimization | |
| CN103895649A (en) | Driver safety driving warning method | |
| CN109002981A (en) | Driving behavior methods of marking, system, device and computer readable storage medium | |
| Szénási et al. | Statistical accident analysis supporting the control of autonomous vehicles | |
| CN113095387B (en) | Road risk identification method based on networking vehicle-mounted ADAS | |
| CN114912689A (en) | Map grid index and XGBOST-based over-limit vehicle destination prediction method and system | |
| Cohen et al. | Travel time estimation between loop detectors and FCD: A compatibility study on the Lille network, France | |
| CN116153082B (en) | Expressway road condition acquisition, analysis and processing system based on machine vision | |
| CN116092037B (en) | Vehicle type identification method integrating track space-semantic features | |
| US11113292B2 (en) | Feature data storage apparatus and driving feature and distribution databases | |
| CN108920753B (en) | Curve color road surface design method based on excellent driver driving track | |
| CN113838282A (en) | Vehicle abnormal behavior detection method based on Beidou positioning | |
| Wen et al. | Analysis of vehicle driving styles at freeway merging areas using trajectory data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |