CN105989062A - Defining method based on electric vehicle travelling track characteristics and data mining technology - Google Patents
Defining method based on electric vehicle travelling track characteristics and data mining technology Download PDFInfo
- Publication number
- CN105989062A CN105989062A CN201510066876.0A CN201510066876A CN105989062A CN 105989062 A CN105989062 A CN 105989062A CN 201510066876 A CN201510066876 A CN 201510066876A CN 105989062 A CN105989062 A CN 105989062A
- Authority
- CN
- China
- Prior art keywords
- track
- normal practice
- modeling
- algorithm
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The invention discloses a defining method based on electric vehicle travelling track characteristics and a data mining technology. The defining method comprises a common track modeling method of track starting point and finishing point, a track envelope curve and driving characteristics and an efficient common track mining algorithm in massive electric vehicle travelling track data, wherein the common track modeling method comprises track starting point and finishing point modeling, track envelope curve modeling and driving characteristic modeling; and the common track mining algorithm comprises a track starting point and finishing point space-time clustering algorithm, a common track finding and track envelope curve determining algorithm and a driving characteristic extracting algorithm. The invention provides a modeling method of a parameterized common track and the track characteristics are parameterized and abstracted; and on the basis of guaranteeing the accuracy of describing a common track rule by the common track, the simplicity of track description is further improved and common modeling is carried out on diversified tracks in the aspect of a macroscopic space-time scale so that the extraction of dynamic information characteristics in a travelling process is realized and identification and classification of track data are realized.
Description
Technical field
The present invention relates to motion track research and calculate field, specifically a kind of definition method based on electric motor car trip track characteristic
And data mining technology.
Background technology
A large amount of with electric bicycle popularize, and cycling trip has become a kind of popular important trip mode.To electric motor car
The big data of track are analyzed, and can therefrom excavate high prices such as including potential commercial value, track exception, real-time traffic
The information of value.How in constantly dynamic electric motor car driving trace that generate, magnanimity, excavate have high abstraction level,
Semantic abundant track rule, is to be presently required to solve the technical problem that.
With the fast development of the technology such as positioning, remote sensing, sum up vague generalization track rule for a large amount of mobile concrete tracks of object
This problem has obtained the common concern of industrial quarters and academia.In this field trajectory clustering problem be intended to by a large number scattered,
Concrete tracking clustering becomes at time or spatially similar or close track, to be analyzed in abstract and macroscopical aspect,
Typical method of trajectory clustering potentially includes typical clustering algorithm such as K-Means etc..But, for trajectory clustering problem,
Its key problem is to define locus model and distance function.Typical method of trajectory clustering often uses building based on track example
Mould method, and the distance function based on structural similarity.
In the modeling method based on track example, the track template being positioned at trajectory clustering center has same with common track
Form, in some instances it may even be possible to be equal to the concrete track of a certain bar.By the distance between continuous relatively newer track and track template,
Determine that new track, closest to which track template, thus determines the trajectory clustering of new track institute subordinate.
In the clustering method based on track example, the feature of another core is that its track distance calculates function.And based on knot
It is conventional method that the distance of structure similitude calculates function, as used the track knot including the direction of motion, speed and corner
Structure feature carries out the calculating of track similarity；On the basis of space similarity, propose to use chronotaxis to make further
Basis for estimation for track configuration similitude.
This class models based on track example and utilizes the method for track configuration Similarity measures track distance to exist on this basis
For travelling regular strong mobile target, such as: the train that travels on the automobile that travels on highway, railway and have solid
On the trajectory analysis of the aircraft of tramp-liner and boats and ships, have obvious advantage.
But, in the face of this class speed of electric motor car is low, travel road conditions require low, drive flexibly for traveling instrument, existing
With the presence of the obvious limitation of method: first, electric motor car is a kind of low speed, flexibly vehicles, requires low to road conditions,
There is not fixing navigation channel, in a large amount of tracks that therefore electric motor car produces, it is difficult to sum up one and there is complete representational track
Track template as cluster centre；Meanwhile, under conditions of lacking track template, traditional based on structural similarity away from
Also cannot use from computational methods；Further, widely available due to electric motor car, therefore has substantial amounts of electric motor car trip
Data, and be dynamic at any time generation, modeling and distance calculating method for a large amount of electric motor car track datas must possess
Enhanced scalability and the feature of dynamic increment formula.Therefore, traditional distance calculating side based on track template and structural similarity
There is calculating and the storage overhead relatively shortcomings such as big, part clustering algorithm dynamic increment characteristic difference in method.
Content of the invention
It is an object of the invention to provide a kind of by mathematical modeling is carried out to abstract normal practice mark, for electric motor car track characteristic,
With form succinct, accurate, the regularity being shown in the modeling daily track of user, resource consumption is little, parallelization degree
The high definition method based on electric motor car trip track characteristic and data mining technology, to solve proposition in above-mentioned background technology
Problem.
For achieving the above object, the present invention provides following technical scheme:
A kind of electric motor car travels mathematical modeling and the method for digging of normal practice mark, comprises the steps of
(1) the normal practice mark modeling method of track beginning and end, track envelope curve, travelling characteristic；
(2) the efficient normal practice mark mining algorithm in magnanimity electric motor car driving trace data.
As the further scheme of the present invention: described normal practice mark modeling method comprises the following steps:
(1), in track beginning and end modeling method, the beginning and end of track has form
<longitude, latitude, radius, time, deltaT>
Wherein, longitude and latitude is for identifying the longitude and latitude coordinate of the central point of starting point S or terminal F；
Radius features the error radius of permission, and its unit is accurate to rice；Time is described in one day from starting point S or reaches
In the moment of terminal F, its unit is accurate to minute；The error range of deltaT then corresponding time, its unit is accurate to minute；
(2) track envelope curve has form: comprise starting point s and terminal f, and two summit p that distance s is farthest with f linemax
With pmin；Track envelope curve is made up of four line segments, i.e. E={ < s, pmax>、<pmax, f>,<s, pmin>、<pmin, f > }；
(3) travelling characteristic has a form:
D=<velocity, var, turn>
Wherein velocity is track average overall travel speed, and var is path velocity variance, and turn is to turn more than 45 degree in track
Curved number of times.
As the present invention further scheme: described normal practice mark mining algorithm comprises the steps of
(1) track beginning and end clustering algorithm comprises room and time threshold value Ts and Tt, algorithm according to room and time away from
From with threshold ratio relatively realize cluster；
(2) normal practice mark discovery determines, with track envelope curve, the determination method that algorithm comprises two track summits, and based on line segment
Track envelope curve computational methods；
(3) track that driving trace feature calculation algorithm comprises from normal practice mark calculates include average overall travel speed, average
Velocity variance, average more than 45 degree number of turns are at interior travelling characteristic.
Compared with prior art, the invention has the beneficial effects as follows: the invention discloses a kind of modeling in daily electric motor car track
With the method excavating high-level normal trace information, it is proposed that the Mathematical Modeling Methods to abstract normal practice mark, for electric motor car track
Feature, with form succinct, accurate, the regularity being shown in the modeling daily track of user, on the basis of model, knot
Close electric motor car and travel feature, it is proposed that excavating the highly effective algorithm of normal practice mark in magnanimity track data data, this algorithm has money
Source consumes the advantages such as little, parallelization degree is high；The present invention uses the modeling method of parametrization normal practice mark, by track characteristic parameter
Change, abstract, on the basis of ensureing the accuracy that vague generalization track rule is described by normal practice mark, further increase track and retouch
The terseness stated.Track beginning and end and track envelope curve thereof the important content as locus model is proposed, by shape the most succinct
Formula, carries out general modeling for variation track on macroscopic view spatial and temporal scales；By the modeling to travelling characteristic, reach right
The extraction of multidate information feature in driving process, to realize the identification of track data and classification.
Advantages of the present invention is analyzed in terms of two, first in terms of normal practice mark modeling from the point of view of, normal practice mark proposed by the invention
Take out line modeling method for the flexible and changeable feature of electric motor car driving trace, use include track starting point, terminal, track envelope curve,
Travelling characteristic, in interior parameterized model modeling method, instead of the existing instantiation modeling method based on track template,
Simplicity of expression, the particularly pardon aspect of the electric motor car driving trace that details is changeable to macroscopic view is consistent, have prominent
Advantage.
Second, in terms of normal practice mark mining algorithm from the point of view of, starting point proposed by the invention, terminal clustering method, from space and
Time dimension has carried out comprehensive distance and has calculated to track stop；Meanwhile, this clustering method with time and space distance threshold is
Cluster standard, it is to avoid the dependence of the priori for cluster number K such as traditional clustering method such as K-Means；The present invention carries
The track envelope curve computational methods going out, just can be formed by simple numerical computations that detail areas is other for a large amount of macroscopic views are consistent
The modeling of track, computing cost is low, can parallelization degree high；The travelling characteristic computational methods that the present invention proposes, have used for reference existing
There is track configuration to describe the advantage of related work, be capable of to the abstract representations travelling minutia with track.
In general, in conjunction with characteristics such as electric motor car trip track regions property are strong, flexible and changeable, proposed by the invention
Method, at the accuracy expressed and terseness, the extensibility of algorithm and parallelization aspect, has prominent advantage.The present invention
The discovery of normal practice mark and the mining algorithm being proposed is a kind of increasable algorithm, i.e. after completing the excavation of normal practice mark, in the face of newly obtaining
The trace information taking, can update existing normal trace information by simple numerical computations, it is to avoid to all trace informations
Again brought computing cost is excavated.
Detailed description of the invention
It is described in more detail below in conjunction with the technical scheme to this patent for the detailed description of the invention.
Embodiment 1
A kind of definition method based on electric motor car trip track characteristic and data mining technology, comprise the following steps:
1st, travelling normal practice mark for electric motor car user and carrying out Mathematical Modeling, Mathematical Modeling is as follows:.
Normal practice mark=<S, F, E, D>
Wherein S and F represents the Origin And Destination of normal practice mark respectively, and E portrays track envelope curve, and D represents driving characteristics；
1.1 track starting points S and terminal F:
Due to electric motor car trip purpose often from A to B, the key character therefore describing track is exactly that it goes out
Send out place and place of arrival；And when describing track beginning and end, in addition to the information of locus, it is necessary in the time
On depict track corresponding departure time and arrival time；To sum up, the beginning and end of normal practice mark is at room and time subscript
Know and electric motor car user and go on a journey the general rule of track；Owing to normal practice mark is the abstractdesription to a large amount of tracks, therefore right
When normal practice mark is modeled, in addition to the central point of room and time, also comprise certain error range；Say in form,
Identify starting point S and the terminating point F of mark of overstepping the limit in two-dimensional map space coordinates, be all hexa-atomic group of a following form:
<longitude, latitude, radius, time, deltaT>
Wherein, longitude and latitude is for identifying the longitude and latitude coordinate of the central point of starting point S or terminal F；
Radius features the error radius of permission, and its unit is accurate to rice；Time is described in one day from starting point S or reaches
In the moment of terminal F, its unit is accurate to minute；The error range of deltaT then corresponding time, its unit is accurate to minute；
1.2 track envelope curve E:
Electric motor car is a kind of flexible, traveling instrument of low speed, compares the track of the vehicles such as automobile, the traveling rail of electric motor car
On the premise of keeping general orientation to stablize, there is bigger randomness in mark.Therefore, carry out at the normal practice mark that electric motor car is travelled
During modeling, it is difficult to use the model of line style to describe.Therefore, when being modeled electric motor car normal practice mark, the present invention proposes
On the basis of initiateing and terminate end points, the mode of track envelope curve is used to summarize the scope of driving trace；Specifically, rail
Mark envelope curve includes two locus of points summits, beginning and end and the envelope curve line segment connecting four summits；
1.3 travelling characteristic D:
When describing the track of electric motor car, owing to it drives flexible, random characteristic, the travelling characteristic of user also must comprise
Among the model of normal practice mark, travelling characteristic comprises to travel average speed, velocity variance, and two-dimensional projection's coordinate and is more than 45
The direction transformation number of times of degree.
2nd, carrying out normal practice mark mining algorithm, described normal practice mark mining algorithm includes three below step:
2.1 track beginning and end space-time clustering algorithms
Space-time clustering algorithm is from time and two, space dimension, from parking spot data discrete in a large number, and rising of discovery track
Point and terminal；Traditional Data Clustering Algorithm, such as K-Means, K-neighbour's scheduling algorithm, there is stronger depending on for cluster number K
Lai Xing；But for a large amount of discrete parking spot point in the presence of electric motor car track, it is difficult to determine the concrete number of its K in advance
Value, i.e. cannot know in advance in all track datas, the number of the beginning and end being comprised；Therefore, the present invention is made
Space-time clustering algorithm when initializing each parking spot point of fair play, by constantly judging sky between parking spot point
Between and temporal distance, and compare with time threshold with space, constantly parking spot point close on space-time carried out
Merge, thus reduce cluster number, ultimately form several inside close to and the parking spot of external discrete point, and by stopping
Sequential between point, determines the beginning and end of track；Specifically, track beginning and end space-time disclosed in this invention
Clustering algorithm, its execution process comprises the steps of
A. all parking spots point set P={p of some electric motor car user is given1, p2..., pn, wherein
pi={ longitude, latitude, time_arrive, time_leave, N} are i-th parking spot points, wherein
Longitude record longitude, latitude record dimension, time_arrive and time_leave record reach down time with
And time departure, N record the record strip number that this parking spot point is comprised, for independent the stopping of each in initial data
Truck position point, N takes 1；
B. an a capacity-threshold Ts and time threshold Tt is given, depending on the final value of Ts is according to application scenarios, for
The traveling feature of electric motor car user, Ts value is Ts≤200 meter, depending on the final value of Tt is according to final application scenarios, pin
Track feature to electric motor car, Tt value is Tt≤30 minute；
C. for any pair stop piWith pj, calculate its space length
Ds (i, j)=ED (longitude_i, latitude_i, longitude_j, latitude_j)
Wherein function ED represents Euclidean distance, the arrival time distance of point-to-point transmission being calculated by the longitude of two points and dimension at 2
Dt_arrive (i, j)=| time_arrive_i-time_arrive_j |
And the time departure distance of point-to-point transmission
Dt_leave (i, j)=| time_leave_i-time_leave_j |
If d. piWith pj((i, j)≤Tt, then by p for i, j)≤Ts and Dt_arrive to meet DsiWith pjMerge into a new arrival
Point pa, the computational methods of the longitude of the new point of arrival, dimension, arrival time, time departure and N are as follows:
Longitude_a=(longitude_i*N_i+longitude_j*N_j)/(N_i+N_j)
Latitude_a=(latitude_i*N_i+latitude_j*N_j)/(N_i+N_j)
Time_arrive_a=(time_arrive_i*N_i+time_arrive_j*N_j)/(N_i+N_j)
Time_leave_a=NaN
N_a=N_i+N_j
Wherein, time_leave_a=NaN is a special marking, is used for representing that this point only comprises arrival information, and does not wraps
Containing leave message, i.e. time departure is a non-number；Similar with above procedure, if piWith pjMeet Ds (i, j)≤Ts and
(i, j)≤Tt, then by p for Dt_leaveiWith pjMerge into one and new leave a p1, newly leave longitude a little, dimension, arrival
The computational methods of time, time departure and N are similar with the algorithm of the point of arrival, only difference is that new arrival time composes
Value is NaN, and new time departure is piWith pjThe weighted sum of time departure；
E. when track does not exist any two points piWith pjMeet Ds (i, j)≤Ts and Dt_arrive (i, j)≤Tt, or
Ds (i, j)≤Ts and Dt_leave (i, j)≤Tt when, algorithm terminates；After final track data will comprise two class clusters
Parking spot point Pa={ paAnd P1={ p1, represent the point of arrival and leave set a little respectively, correspond in track data, be i.e.
The terminal of track and the set of starting point；
F. for P1={ p1Each in } leaves a p1, build a track starting point
S=<longitude_l, latitude_l, Ts, time_leave_l, Tt>
Similarly, for Pa={ paEach point of arrival p in }a, build a trail termination point
F=<longitude_a, latitude_a, Ts, time_arrive_a, Tt>
2.2 normal practice mark discoveries determine algorithm with track envelope curve
In the track beginning and end collection basis being obtained by the cluster of step 2.1, given any one track, if should
The beginning and end of track is respectively s and f, then execution following steps:
If a. there is normal practice mark c=<S, F, E, D>, wherein S=s and F=f, then integrate with c by this track；If not existing
Some normal practice marks can comprise this new track, then newly-built normal practice mark c=<s, f, E, D>, wherein E and D is track to be determined
Envelope curve and travelling characteristic；
B. for certain normal practice mark c, its starting point s and terminal f are connected with straight line, in all tracks being included in this normal practice mark,
In calculating track, each puts distance d with the straight line being connected s and f, if tracing point is in the northeastward of line, then distance is
Just；Otherwise, if tracing point is at the southwestward of line, then distance is negative；
C. in all of tracing point, the minimum and maximum some p of chosen distancemaxAnd pmin, with these 2 additional track starting points and
Terminal s and f is summit, connects < s, pmax>、<pmax, f>,<s, pmin>、<pmin, four line segments of f > gained, form normal practice
The envelope curve of mark c, i.e. E={ < s, pmax>、<pmax, f>,<s, pmin>、<pmin, f > }
2.3 travelling characteristic extraction algorithms
Obtain normal practice mark c and the track that comprised thereof on the basis of a step in 2.2 after, the extraction algorithm meter of travelling characteristic
That calculates all tracks of comprising of c comprises average overall travel speed velocity, average speed variance var, average more than 45 degree turnings
Number of times turn is at interior travelling characteristic, i.e. D=<velocity, var, turn>.
Sum up three above step, normal practice mark Origin And Destination clustering algorithm disclosed in this invention, the discovery of normal practice mark and track
Envelope curve determines that algorithm, travelling characteristic extraction algorithm can dig on the basis of a large amount of electric motor car user's go off daily track record
Excavate with c=< S, F, E, D be the normal practice mark of form.
Although moreover, it will be appreciated that this specification is been described by according to embodiment, but not each embodiment only comprises
One independent technical scheme, this narrating mode of specification is only for clarity sake, and those skilled in the art should be by
Specification is as an entirety, and the technical scheme in each embodiment also can be through appropriately combined, and forming those skilled in the art can
With other embodiments understanding.
Claims (3)
1. an electric motor car travels mathematical modeling and the method for digging of normal practice mark, it is characterised in that comprise the steps of
(1) the normal practice mark modeling method of track beginning and end, track envelope curve, travelling characteristic；
(2) the efficient normal practice mark mining algorithm in magnanimity electric motor car driving trace data.
2. electric motor car according to claim 1 travels mathematical modeling and the method for digging of normal practice mark, it is characterised in that: institute
State normal practice mark modeling method to comprise the following steps:
(1), in track beginning and end modeling method, the beginning and end of track has form
<longitude, latitude, radius, time, deltaT>
Wherein, longitude and latitude is for identifying the longitude and latitude coordinate of the central point of starting point S or terminal F；
Radius features the error radius of permission, and its unit is accurate to rice；Time is described in one day from starting point S or reaches
In the moment of terminal F, its unit is accurate to minute；The error range of deltaT then corresponding time, its unit is accurate to minute；
(2) track envelope curve has form: comprise starting point s and terminal f, and two summit p that distance s is farthest with f linemax
With pmin；Track envelope curve is made up of four line segments, i.e. E={ < s, pmax>、<pmax, f>,<s, pmin>、<pmin, f > }；
(3) travelling characteristic has a form:
D=<velocity, var, turn>
Wherein velocity is track average overall travel speed, and var is path velocity variance, and turn is to turn more than 45 degree in track
Curved number of times.
3. electric motor car according to claim 1 travels mathematical modeling and the method for digging of normal practice mark, it is characterised in that: institute
State normal practice mark mining algorithm to comprise the steps of
(1) track beginning and end clustering algorithm comprises room and time threshold value Ts and Tt, algorithm according to room and time away from
From with threshold ratio relatively realize cluster；
(2) normal practice mark discovery determines, with track envelope curve, the determination method that algorithm comprises two track summits, and based on line segment
Track envelope curve computational methods；
(3) track that driving trace feature calculation algorithm comprises from normal practice mark calculates include average overall travel speed, average
Velocity variance, average more than 45 degree number of turns are at interior travelling characteristic.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510066876.0A CN105989062A (en) | 2015-02-04 | 2015-02-04 | Defining method based on electric vehicle travelling track characteristics and data mining technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510066876.0A CN105989062A (en) | 2015-02-04 | 2015-02-04 | Defining method based on electric vehicle travelling track characteristics and data mining technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105989062A true CN105989062A (en) | 2016-10-05 |
Family
ID=57037585
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510066876.0A Pending CN105989062A (en) | 2015-02-04 | 2015-02-04 | Defining method based on electric vehicle travelling track characteristics and data mining technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105989062A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107122461A (en) * | 2017-04-27 | 2017-09-01 | 东软集团股份有限公司 | One kind trip method of trajectory clustering, device and equipment |
| CN107784597A (en) * | 2017-09-19 | 2018-03-09 | 平安科技（深圳）有限公司 | Trip mode recognition methods, device, terminal device and storage medium |
| CN108109369A (en) * | 2018-02-06 | 2018-06-01 | 深圳市物语智联科技有限公司 | A kind of vehicle in use based on driving trace and non-vehicle in use identification measure of supervision |
| CN109357682A (en) * | 2018-09-19 | 2019-02-19 | 潍坊工程职业学院 | A kind of road navigation method |
| CN110222131A (en) * | 2019-05-21 | 2019-09-10 | 北京交通大学 | The beginning and the end information extracting method and device |
| CN111599165A (en) * | 2020-01-21 | 2020-08-28 | 南京中新赛克科技有限责任公司 | Multi-source big data-based electric vehicle robbery real-time warning method and system |
| CN114428807A (en) * | 2022-01-24 | 2022-05-03 | 中国电子科技集团公司第五十四研究所 | Ground maneuvering target motion trajectory semantic system construction and cognitive optimization method |
-
2015
- 2015-02-04 CN CN201510066876.0A patent/CN105989062A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107122461A (en) * | 2017-04-27 | 2017-09-01 | 东软集团股份有限公司 | One kind trip method of trajectory clustering, device and equipment |
| CN107122461B (en) * | 2017-04-27 | 2019-08-13 | 东软集团股份有限公司 | A kind of trip method of trajectory clustering, device and equipment |
| CN107784597A (en) * | 2017-09-19 | 2018-03-09 | 平安科技（深圳）有限公司 | Trip mode recognition methods, device, terminal device and storage medium |
| CN107784597B (en) * | 2017-09-19 | 2021-09-28 | 平安科技（深圳）有限公司 | Travel mode identification method and device, terminal equipment and storage medium |
| CN108109369A (en) * | 2018-02-06 | 2018-06-01 | 深圳市物语智联科技有限公司 | A kind of vehicle in use based on driving trace and non-vehicle in use identification measure of supervision |
| CN109357682A (en) * | 2018-09-19 | 2019-02-19 | 潍坊工程职业学院 | A kind of road navigation method |
| CN110222131A (en) * | 2019-05-21 | 2019-09-10 | 北京交通大学 | The beginning and the end information extracting method and device |
| CN111599165A (en) * | 2020-01-21 | 2020-08-28 | 南京中新赛克科技有限责任公司 | Multi-source big data-based electric vehicle robbery real-time warning method and system |
| CN114428807A (en) * | 2022-01-24 | 2022-05-03 | 中国电子科技集团公司第五十四研究所 | Ground maneuvering target motion trajectory semantic system construction and cognitive optimization method |
| CN114428807B (en) * | 2022-01-24 | 2023-11-03 | 中国电子科技集团公司第五十四研究所 | Method for constructing semantic system and cognition optimization of ground maneuvering target motion trail |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105989062A (en) | Defining method based on electric vehicle travelling track characteristics and data mining technology | |
| Yu et al. | Deep learning-based traffic safety solution for a mixture of autonomous and manual vehicles in a 5G-enabled intelligent transportation system | |
| CN111492202B (en) | Vehicle operation location determination | |
| CN103914985B (en) | A kind of hybrid power passenger car following speed of a motor vehicle trajectory predictions method | |
| CN105718750A (en) | Prediction method and system for vehicle travelling track | |
| CN103853155B (en) | Intelligent vehicle road junction passing method and system | |
| CN110909788B (en) | Statistical clustering-based road intersection position identification method in track data | |
| CN108068815A (en) | System is improved for the decision-making based on planning feedback of automatic driving vehicle | |
| CN110861650A (en) | Vehicle path planning method and device, vehicle-mounted equipment and storage medium | |
| CN110389995B (en) | Lane information detection method, apparatus, device, and medium | |
| CN107310550A (en) | Road vehicles travel control method and device | |
| CN104608766A (en) | Automatic parking method and system used for intelligent vehicle through parking memory stick | |
| CN107664504A (en) | A kind of path planning apparatus | |
| Valera et al. | Driving cycle and road grade on-board predictions for the optimal energy management in EV-PHEVs | |
| CN113511204B (en) | Vehicle lane changing behavior identification method and related equipment | |
| CN108470460A (en) | A kind of nearby vehicle Activity recognition method based on smart mobile phone and RNN | |
| Qian et al. | Vehicular networking-enabled vehicle state prediction via two-level quantized adaptive Kalman filtering | |
| CN115523934A (en) | Vehicle track prediction method and system based on deep learning | |
| Meng et al. | Trajectory prediction for automated vehicles on roads with lanes partially covered by ice or snow | |
| Wang et al. | Construction and application of artificial intelligence crowdsourcing map based on multi-track GPS data | |
| CN116182884A (en) | Intelligent vehicle local path planning method based on transverse and longitudinal decoupling of frenet coordinate system | |
| CN102082996A (en) | Self-locating mobile terminal and method thereof | |
| CN116989816B (en) | Yaw identification method and device and electronic equipment | |
| Wang et al. | Segmented trajectory clustering-based destination prediction in IoVs | |
| Xi et al. | Map matching algorithm and its application |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20161005 |