CN109800903A - A kind of profit route planning method based on taxi track data - Google Patents
A kind of profit route planning method based on taxi track data Download PDFInfo
- Publication number
- CN109800903A CN109800903A CN201811541105.2A CN201811541105A CN109800903A CN 109800903 A CN109800903 A CN 109800903A CN 201811541105 A CN201811541105 A CN 201811541105A CN 109800903 A CN109800903 A CN 109800903A
- Authority
- CN
- China
- Prior art keywords
- point
- taxi
- track data
- cluster
- profit
- 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.)
- Withdrawn
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of profit route planning methods based on taxi track data, include the following steps: to pre-process taxi track data, and pretreatment is used for cancelling noise data, improve the quality of data;The excavation of best traveller point, the highest best traveller's point of value density is excavated by clustering algorithm twice from the track data of magnanimity;The planning of profit route converts TSP mathematical model for the route planning problem based on point-to-point based on the set of best traveller's point, carries out route planning using heuritic approach, and get a profit function with the unit time to judge the profitability of route.The present invention is by way of first clustering and planning again, the planning of more scientific and rational optimum profitability route is provided for taxi, not only improve the running income of taxi, realize the benefit of taxi driver, and the attendance of taxi is improved, improve the traffic condition of urban road.
Description
Technical field
The present invention relates to route planning technical fields, and in particular to a kind of profit route rule based on taxi track data
The method of drawing.
Background technique
With wireless communication and the development of mobile calculation technique and the development of global location and navigation system, space number
It can be used for the storage of track data according to increasing, novel computing technique and system, pre-process, retrieval and excavation.Therefore,
Space tracking calculating has become more and more important research topic.The application of vehicle GPS and universal, so that traditional traffic system
Develop to intelligent transportation system.The track data amount of taxi is big, data distribution is wide, easily obtains, and when containing abundant
Between and Spatial Dimension, have very big excavated space.
With urban economy fast development, resident trip demand constantly rises, weight of the taxi as urban public transport
Component part is wanted, is brought great convenience for public trip.The taxi-hailing software released at present, such as drop drop, Uber etc.,
Main is still guiding with passenger, and the mode of service is to be issued to request by passenger, and driver goes to designated place to pick passenger.But
For driver, in the case where no order, passive waiting in situ can only be selected, or independently seek visitor.Such case
It not only consumes a large amount of time and oil is costly, reduce the running income of taxi, moreover, being handed over for entire urban road
For logical, peak time especially on and off duty, a large amount of unloaded taxis be easy to cause the traffic paralysis of entire urban road.
Therefore, have can not for the sound development in rationally effective taxi route planning city even entire for entire taxi trade
The important function ignored.
Current taxi planning scheme main target is to realize that route is most short and the time is most short, is but difficult to reach most Cayee
The purpose of benefit.Such as it using cab driving experience, analyzes and researches to taxi history GPS track, proposes that one kind is based on
The taxi route programme of experience track realizes taxi optimal route, but the optimal route is most fast roadway
Line, rather than optimum profitability route.Based on hiring out in the intelligent travelling crane route planning system of wheel paths, by terrestrial reference graph model,
Personalized optimal driving route is provided the user with using the track data of taxi driver and ordinary user.But the system is only
It is only to aid in user and selects the most fast route for being most suitable for the user in practice, be not based on the route of driver benefit
Planning.The traffic data generated in solution based on cloud by processing taxi, excavates a large amount of taxis in the cluster
Track data collection and travel pattern propose a kind of real-time best route recommender system.But the system more focus on be knot
It closes current traffic condition to be planned in real time, is a kind of public traffic service system, and be not based on being full of for taxi driver
The planning of sharp route.
Summary of the invention
Goal of the invention: in order to overcome the deficiencies in the prior art, one kind is provided towards taxi driver, based on taxi
The profit route planning method of wheel paths data.
Technical solution: to achieve the above object, the present invention provides a kind of profit route rule based on taxi track data
The method of drawing, includes the following steps:
1) taxi track data is pre-processed, pretreatment is used for cancelling noise data, improves the quality of data;
2) value density the excavation of best traveller point: is excavated most from the track data of magnanimity by clustering algorithm twice
High best traveller's point;
3) planning of profit route: the set based on best traveller's point that step 2 is formed advises the route based on point-to-point
The problem of drawing is converted into TSP (Travelling Salesman Problem) mathematical model, carries out route rule using heuritic approach
It draws, and gets a profit function with the unit time to judge the profitability of route.
Further, successively three cleaning including data, sampling and subregion steps are pre-processed in the step 1.It is described
The cleaning of data is to clean original taxi track data, described different for removing the abnormal point in track data
Often point includes default values and erroneous values;Since the sampled data for hiring out wheel paths is excessively intensive, the data meeting of bulk redundancy
Lead to a large amount of consumption of clustering algorithm computing resource, so We conducted at sampling appropriate in the Data processing of early period
Reason, the method that the sampling of the data uses mixed sampling;In order to realize best traveller's point of more accurateization, we are herein
The multidomain treat-ment of data is carried out, the subregion of the data includes being distributed the track data of taxi according to urban core region
Region division is carried out, three-level division is generally divided into, level-one city is the nucleus in entire city, and three-level city is the outer of city
Enclose area and refer to some regions far from city main city zone, taxi track data here is more sparse, based on level-one city and
The region of external zones is exactly second level city.
Further, the method for the mixed sampling: daily data are first waited first equal more last than sampling again than cutting
Sampling results are merged, several days data of processing are then merged into resampling.Using this methods of sampling, we are guaranteeing
Packing density is diluted in the distortionless situation of data distribution.
Further, clustering algorithm respectively is density-based algorithms and is based on drawing twice in the step 2
The clustering algorithm divided, the step 2 specifically: obtained using density-based algorithms in taxi track data first
Cluster, be based on each cluster, the clustering algorithm based on division recycled to obtain the central point of cluster, gained central point is best visitor
Source point.
The purpose of the step 2 is that best traveller's point of most value density is excavated from track data.Best traveller
Point is the central point that point most intensive in taxi historical trajectory data is concentrated, this shows that a large amount of taxi driver accessed this
Region.If unloaded taxi driver can accurately find these places, it necessarily can be connected to maximum probability passenger, improved
Profitability.Carry out best traveller's point in mining track data present invention employs two different clustering algorithms.Using being based on
The clustering algorithm of density can excavate the cluster in track data, the historical track number of taxi while cancelling noise data
According to distribution, there are nucleus to be distributed extensively, and remote districts are distributed sparse feature, and density-based algorithms can be good at
The class in each area is detected, data form data volume cluster of different sizes by cluster for the first time, and the corresponding reality of each cluster is raw
Region not of uniform size namely the region of some hot topics in work.Planning for taxi driver between zones, especially
Be between large area region planning be compared to for the planning of point-to-point precisely quickly.Therefore, it is necessary to central point come
Instead of a panel region, the clustering algorithm based on division can find out the central point of each cluster, namely best traveller point, based on division
Clustering algorithm firstly the need of the number of a given cluster finally clustered, then algorithm adjusts object in cluster by iteration
It divides, generates final cluster result, by the cluster again to each cluster, the route planning problem between region is converted into a little
To the accurate route planning problem of point.
For the step 3 in the route planning based on best traveller's point, the present invention has abandoned traditional Route Planning Algorithm,
But it uses advanced heuritic approach and carries out route planning.In real-time Route Planning Algorithm, heuritic approach is not
Can be as traditional Route Planning Algorithm as problem scale becomes larger, exponentially type increases for computing resource consumption.Heuritic approach energy
Accurate reasonable route planning is provided in acceptable time range.In solution procedure, driver current location is obtained first,
Best traveller's point on periphery is searched for, secondly the input by the set of best traveller's point as heuritic approach, passes through heuristic calculation
Method solves to obtain the traversal order of best traveller's point, and gets a profit function using the unit time to judge the profit energy of the route solution
Power is finally shown the optimum profitability route of selection on map by visualization technique.
Clustering algorithm excavates the highest best traveller's point of value density from the data of million ranks twice in the present invention.
Cluster excavates the hot spot areas in track data for the first time, but these hot spot areas may be a piece of very big range, unfavorable
In the accurate planning of taxi driver, second of cluster excavates the best traveller of most value density from each hot spot areas
Point.
The acquisition of best traveller point is accurate and reliable in the present invention.Best traveller point is from true taxi historical track
It excavates, in order to keep best traveller's point of each region more accurate, we have also carried out cleaning, sampling to track data
And multidomain treat-ment, it is rationally excavated according to the data volume in each region.
The utility model has the advantages that compared with prior art, the present invention having following advantage:
1, the present invention is by first clustering the method planned again, can at a reasonable time with improve taxi within the scope of mileage
Attendance, realize the benefit of driver, unloaded taxi driver can only in the case where no order before solving
It selects passive original place to wait or independently seek visitor, causes to consume a large amount of time cost and oily costly problem.
2, heuritic approach rationally can effectively plan travel route, heuristic compared with traditional Route Planning Algorithm
Algorithm has the characteristics that versatility, stability and very fast constringent, can rapidly and accurately give in real-time route planning
Traffic path out gets a profit function using the unit time to select optimal profit for several routes that heuritic approach solves
Route.
3, the raising of taxi attendance reduces the no-load ratio of taxi, so as to improve urban traffic conditions.
Detailed description of the invention
Fig. 1 is the subregion schematic diagram of data;
Fig. 2 is density-based algorithms flow chart;
Fig. 3 is the distribution schematic diagram of cluster result cluster, is made of Fig. 3 a, Fig. 3 b and Fig. 3 c, and wherein Fig. 3 a is level-one city
Butut is distinguished, Fig. 3 b is level-one city distribution map, and Fig. 3 c is peripheral region distribution map;
Fig. 4 is the clustering algorithm flow chart based on division;
Fig. 5 is heuritic approach flow chart;
Fig. 6 is the planning chart of optimum profitability route;
Fig. 7 is the overall flow schematic diagram of profit route planning method in the present invention.
Specific embodiment
In the following with reference to the drawings and specific embodiments, the present invention is furture elucidated.
The present embodiment goes out a most Cayee using one of Chengdu true taxi as object for the taxi planning
Sharp route, first pair current embodiment require that the related definition used is illustrated.
Define 1 (track): one is hired out a series of sequence that wheel paths T is points with longitude and latitude. T:P1→P2→...
→Pn, tracing point P (x, y), wherein x represents longitude, and y represents dimension.
Define 2 (distances): the distance between two coordinate points.Two coordinate points Ps=(xs,ys) and Pt=(xt,yt) between
Distance ρ (Ps,Pt) be defined as
Wherein, a=| ys-yt| it is Ps、PtThe difference of the latitude of two o'clock, b=| xs-xt| it is Ps、PtThe difference of the longitude of two o'clock; Re
=6378.137 indicate the spherical radius of the earth.
Define 3 (unit time profit function M): the total earnings of taxi is that total income I subtracts totle drilling cost C, by consolidated profit & loss
Benefit is exactly unit time profit M divided by duration T is driven, and formula is as follows:
M=(I-C)/T (2)
Wherein
Taxi is connected to passenger, when carrying kilometres S is in base price public affairs S0When within mileage, total income is exactly basic rent rate I0;
When carrying milimeter number S is more than minimum mileage number S0When, overage is according to every kilometer of charge standard COEFFICIENT K0It is calculated.
C=K1*(S+L) (4)
S is carrying kilometres, and L is dead mileage, K1It is oil price consumption.
T=(S+L)/V (5)
V represents the average speed that taxi driver drives in urban road.
Define 4 (best traveller's points): based on profit purpose, taxi driver's maximum probability is connected to passenger's aggregation of passenger
Point.The present embodiment shows as the point set after the cluster twice based on taxi track data.
Data set used in the present embodiment is the GPS record of 1.4 ten thousand, Chengdu taxi, and the time was from 2014 08
03 day to the 30 days 08 month moon, the dimension of data include 5, and (1 indicates carrying, 0 table for taxi ID, latitude, longitude, passenger carrying status
Show no visitor) and time point.
As shown in fig. 7, the planning for carrying out profit route to taxi in the present embodiment successively respectively includes the pre- place of data
Three reason, the excavation of best traveller point and route planning parts, are below described in detail this three parts.
One, the pretreatment of data
The pretreatment of data mainly contains three steps: the cleaning of data, the sampling of data and data subregion.
(1) cleaning of data.Cancelling noise data, including default values and erroneous values.Default values refer to lacking
Latitude and longitude information or the incomplete tracing point of latitude and longitude information.The numerical value of mistake refers to data type mistake or longitude and latitude
Coordinate has exceeded the tracing point of target cities longitude and latitude range.Such as the longitude and latitude in Chengdu is between 102 ° 54 '~104 ° of east longitude
53 ', between 30 ° 05~31 ° 26 ' of north latitude, what it is more than this longitude and latitude range is exactly noise spot.
(2) sampling of data.If sample frequency is excessively intensive, mass of redundancy data will be present in track data.For example,
If sample frequency is 1 second/time, i.e., 13.9 meters are about travelled in 1 second, this can be neglected in the taxi stroke of hundreds of kilometer
Slightly disregard.Meanwhile in next Data processing, such as cluster, mass of redundancy data can also generate huge memory overhead.
In order to avoid caused due to sampling of data track data distribution distortion, we first to daily taxi track data collection into
The data of row equal part are cut into 20 parts, and the data set then cut through to every part carries out etc. than sampling, by the result sampled into
Row merges, and the data after merging are exactly the track data collection of sampling in one day.But the taxi track data collection of some day is
Be difficult to that the distribution of best traveller's point in entire city is fully described, herein we by the sample data set of continuous a period of time into
Row merges samples again.Using this mixed sampling method, we dilute number in the case where guaranteeing the distortionless situation of data distribution
According to density.
(3) subregion of data.In order to excavate best traveller's point of more accurateization, the present embodiment has carried out data herein
Multidomain treat-ment.The present embodiment is with the downtown in Chengdu, during Tianfu Square Development (30.6633976913,104.0723725172) is
The heart is auxiliary with ring road, carries out the subregion in city, obtain subregion schematic diagram shown in FIG. 1.In order to verify the reasonable of subregion
Property, inventor has investigated the actual conditions in Chengdu.In the subregion of Fig. 1, level-one city includes the area of the core in Chengdu, such as
Jinjiang District, Jinniu District etc.;It include Wenjiang District, Longquanyi District etc. in second level city;The packet in peripheral region (three-level city)
Dayi County, Jianyang City etc. are included.
Two, the excavation of best traveller's point
After the pretreatment of pending data is completed, need to excavate best traveller's point from the track data of magnanimity.Data point
After area's processing, the data volume in each region is different, if will lead to the data meeting of some border districts using global clustering
It is rejected as noise data, it is obtaining the result is that best traveller's point focuses primarily upon the more inner city of data volume, still,
For the planning of whole city, the planning of route cannot be limited only to inner city, so the present embodiment is respectively to division
Three regions are clustered twice, are broadly divided into two steps, and first time cluster result goes out the cluster in each region track,
That is hot spot areas.Second of cluster is the central point in excavating hot spot areas, namely defines best traveller's point in 4.
(1) excavation of hot spot areas.Density-based algorithms are applied for each region, in mining track data
Hot spot areas.Since the data volume of each region is different, according to trizonal data volume number debug clustering algorithm
Parameter.In traditional density-based algorithms, usually measured using Euclidean distance between two points away from
From.Since the present embodiment uses true taxi track data, so here using the range formula (1) defined in 2
Calculate the actual distance between two tracing points, the replacement of range formula is so that the result of cluster is more authentic and valid.
As shown in Fig. 2, specific step is as follows for density-based algorithms:
2.1) taxi track data is read;
2.2) parameter is inputted;
2.3) since the arbitrary point in track data, search all density achievable pairs as;
2.4) identify whether the point is core point, if so, going to step 2.5;If it is not, the point is labeled as noise spot, turn step
Rapid 2.3;
2.5) point all in the vertex neighborhood is continued to investigate as seed point, until finding a complete cluster;
2.6) untreated point is judged whether there is, if so, going to step 2.3;If it is not, will not belong to the point label of any cluster
For noise spot;
2.7) cluster result is exported.
Point of the cluster of each region that taxi track data is formed after being clustered by above-mentioned density-based algorithms
Cloth, it is specific as shown in figure 3, wherein by Fig. 3 a it is found that the cluster distribution in level-one city is most intensive;By Fig. 3 b it is found that second level city
Cluster is distributed comparatively dense;By Fig. 3 c it is found that the cluster distribution of peripheral region is sparse.Data are formed by cluster for the first time in each region
Cluster of different sizes, each cluster are the set of a large amount of tracing points, and we term it hot spot areas.Size based on each cluster
Difference, the range that such hot spot areas is geographically distributed vary, for taxi driver between a wide range of
Planning, be compared to for the planning of point-to-point be less precisely quickly.
(2) excavation of best traveller point.Based on the cluster of each region for clustering formation in last step, used in this step
Clustering algorithm based on division excavates best traveller's point, it is necessary first to the number of a given cluster finally clustered, so
Algorithm adjusts the division of object in cluster, generates final cluster result by iteration afterwards.Purpose is excavated in each cluster
Best traveller's point in heart point namely track data.By to second of each cluster cluster, best traveller's point of excavation can be with
Instead of a panel region, the route planning problem between such region is converted into the accurate route planning problem of point-to-point.
As shown in figure 4, specific step is as follows for the clustering algorithm based on division:
3.1) track data and cluster centre in cluster are inputted;
3.2) cluster centre is initialized;
3.3) for each tracing point in cluster, it is calculated to the distance of cluster centre and is assigned to apart from the smallest poly-
In class corresponding to class center;
3.4) it is directed to each classification, recalculates its cluster centre;
3.5) judge whether data restrain, if so, going to step 3.6;If it is not, going to step 3.3
3.6) cluster result is exported.
Three, route planning
After excavation based on best traveller's point, the taxi track data of magnanimity is converted into the density that is possessed of higher values
Best traveller point.Based on the route planning of best traveller's point, its essence is the planning between point-to-point, the present embodiment turns problem
It turns to TSP problem and these problems is solved using heuristic, as shown in figure 5, solving most Cayee using heuritic approach
The detailed process of sharp route are as follows:
4.1) initialization of algorithm;
4.2) driver current location is obtained, best traveller's point on periphery is searched for;
4.3) traversal order of best traveller's point is solved;
4.4) judge whether to meet condition, if so, going to step 4.5;If it is not, the number of iterations adds 1,4.3 are gone to step;
4.5) result is exported.
In order to judge whether required route solution has really achieved the purpose that profit, the present embodiment is using the list defined in 3
The position time gets a profit function to measure, and the present embodiment obtains relevant information from Chengdu taxi net to determine unit time profit letter
Several parameter settings.
Referring to defining 3, according to 8 yuan basic of rent rate namely I0=8, minimum mileage number is 2 kilometers, S0=2.More than 2 public affairs
In valuate according to 1.9 yuan/kilometer, K0=1.9, according to the profit formula for the unit time that we define, gone out using average
The no-load ratio 30% hired a car calculates namely L=(S+L) * 30%, according to No. 92 7.34 yuan/every liter of gas price, in city
It is travelled on road, about 12.5 kilometers/liter, so K1=0.58 yuan/kilometer.
The unit time profit function obtained by above parameter judges the route cooked up, the present embodiment
Optimum profitability route map as shown in FIG. 6 is obtained, the present embodiment is with a taxi driver in Chengdu respectively according to Fig. 6
Route and conventional route run test, test result is to be significantly more than conventional road using the profit of the route of Fig. 6
Line, whole attendance also have been improved than usual.
Claims (8)
1. a kind of profit route planning method based on taxi track data, characterized by the following steps:
1) taxi track data is pre-processed, pretreatment is used for cancelling noise data, improves the quality of data;
2) value density the excavation of best traveller point: is excavated from the taxi track data of magnanimity by clustering algorithm twice
Highest best traveller's point;
3) planning of profit route: the set based on best traveller's point that step 2 is formed asks the route planning based on point-to-point
Topic is converted into TSP mathematical model, carries out route planning using heuritic approach, and get a profit function with the unit time to judge route
Profitability.
2. a kind of profit route planning method based on taxi track data according to claim 1, it is characterised in that:
Successively three cleaning including data, sampling and subregion steps are pre-processed in the step 1.
3. a kind of profit route planning method based on taxi track data according to claim 2, it is characterised in that:
The cleaning of the data is to clean original taxi track data, for removing the abnormal point in track data, institute
Stating abnormal point includes default values and erroneous values;The method that the sampling of the data uses mixed sampling, for guaranteeing data
It is distributed in distortionless situation and dilutes packing density;The subregion of the data includes by the track data of taxi according to city core
Heart area distribution carries out region division.
4. a kind of profit route planning method based on taxi track data according to claim 1, it is characterised in that:
Clustering algorithm respectively is density-based algorithms and the clustering algorithm based on division twice in the step 2, described
Step 2 specifically: be the cluster obtained using density-based algorithms in taxi track data first, be based on each
Cluster recycles the clustering algorithm based on division to obtain the central point of cluster, and gained central point is best traveller's point.
5. a kind of profit route planning method based on taxi track data according to claim 4, it is characterised in that:
The calculation process of the density-based algorithms is as follows:
2.1) taxi track data is read;
2.2) input parameter;
2.3) since the arbitrary point in track data, search all density achievable pairs as;
2.4) identify whether the point is core point, if so, going to step 2.5;If it is not, the point is labeled as noise spot, go to step
2.3;
2.5) point all in the vertex neighborhood is continued to investigate as seed point, until finding a complete cluster;
2.6) untreated point is judged whether there is, if so, going to step 2.3;If it is not, the point that will not belong to any cluster is labeled as making an uproar
Sound point;
2.7) cluster result is exported.
6. a kind of profit route planning method based on taxi track data according to claim 4 or 5, feature exist
In: the calculation process of the clustering algorithm based on division is as follows:
3.1) track data and cluster centre in cluster are inputted;
3.2) cluster centre is initialized;
3.3) for each tracing point in cluster, it is calculated to the distance of cluster centre and is assigned in the smallest cluster
In class corresponding to the heart;
3.4) it is directed to each classification, recalculates its cluster centre;
3.5) judge whether data restrain, if so, going to step 3.6;If it is not, going to step 3.3;
3.6) cluster result is exported.
7. a kind of profit route planning method based on taxi track data according to claim 1, it is characterised in that:
The detailed process of optimum profitability route is solved in the step 3 using heuritic approach are as follows:
4.1) initialization of algorithm;
4.2) driver current location is obtained, best traveller's point on periphery is searched for;
4.3) traversal order of best traveller's point is solved;
4.4) judge whether to meet condition, if so, going to step 4.5;If it is not, the number of iterations adds 1,4.3 are gone to step;
4.5) result is exported.
8. a kind of profit route planning method based on taxi track data according to claim 1, it is characterised in that:
Unit time profit function M in the step 3 are as follows:
M=(I-C)/T (1)
Wherein I is total income, and C is totle drilling cost, and T is driving duration, wherein
Taxi is connected to passenger, when carrying kilometres S is in minimum mileage number S0Within when, total income is exactly basic rent rate I0;Work as load
Seat-kilometer number S is more than minimum mileage number S0When, overage is according to every kilometer of charge standard COEFFICIENT K0It is calculated,
C=K1*(S+L) (3)
S is carrying kilometres, and L is dead mileage, K1It is oil price consumption,
T=(S+L)/V (4)
V represents the average speed that taxi driver drives in urban road.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811541105.2A CN109800903A (en) | 2018-12-17 | 2018-12-17 | A kind of profit route planning method based on taxi track data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811541105.2A CN109800903A (en) | 2018-12-17 | 2018-12-17 | A kind of profit route planning method based on taxi track data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109800903A true CN109800903A (en) | 2019-05-24 |
Family
ID=66556828
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811541105.2A Withdrawn CN109800903A (en) | 2018-12-17 | 2018-12-17 | A kind of profit route planning method based on taxi track data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109800903A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110490393A (en) * | 2019-09-24 | 2019-11-22 | 湖南科技大学 | Objective route planning method, system and medium are sought in conjunction with the taxi of experience and direction |
CN112348265A (en) * | 2020-11-10 | 2021-02-09 | 交控科技股份有限公司 | Feasible path mining method and device under monitoring scene |
WO2021077300A1 (en) * | 2019-10-22 | 2021-04-29 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for improving an online to offline platform |
-
2018
- 2018-12-17 CN CN201811541105.2A patent/CN109800903A/en not_active Withdrawn
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110490393A (en) * | 2019-09-24 | 2019-11-22 | 湖南科技大学 | Objective route planning method, system and medium are sought in conjunction with the taxi of experience and direction |
CN110490393B (en) * | 2019-09-24 | 2022-05-31 | 湖南科技大学 | Taxi passenger-searching route planning method, system and medium combining experience and direction |
WO2021077300A1 (en) * | 2019-10-22 | 2021-04-29 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for improving an online to offline platform |
CN112348265A (en) * | 2020-11-10 | 2021-02-09 | 交控科技股份有限公司 | Feasible path mining method and device under monitoring scene |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110298500B (en) | Urban traffic track data set generation method based on taxi data and urban road network | |
CN102799897B (en) | Computer recognition method of GPS (Global Positioning System) positioning-based transportation mode combined travelling | |
CN102607553B (en) | Travel track data-based stroke identification method | |
CN108920481B (en) | Road network reconstruction method and system based on mobile phone positioning data | |
CN102332210B (en) | Method for extracting real-time urban road traffic flow data based on mobile phone positioning data | |
CN103177575B (en) | System and method for dynamically optimizing online dispatching of urban taxies | |
CN104330089B (en) | A kind of method that map match is carried out using history gps data | |
CN112133090A (en) | Multi-mode traffic distribution model construction method based on mobile phone signaling data | |
CN105206048A (en) | Urban resident traffic transfer mode discovery system and method based on urban traffic OD data | |
CN110796337B (en) | System for evaluating service accessibility of urban bus stop | |
CN105809962A (en) | Traffic trip mode splitting method based on mobile phone data | |
CN110413855B (en) | Region entrance and exit dynamic extraction method based on taxi boarding point | |
CN109800903A (en) | A kind of profit route planning method based on taxi track data | |
CN112734176B (en) | Charging station building method and device, terminal equipment and readable storage medium | |
CN104731963A (en) | Grid path recommending method and system based on internet of vehicle | |
CN110837973B (en) | Human trip selection information mining method based on traffic trip data | |
CN110414795B (en) | Newly-increased high-speed rail junction accessibility influence method based on improved two-step mobile search method | |
CN115412857B (en) | Resident trip information prediction method | |
CN106547862A (en) | Traffic big data dimension-reduction treatment method based on manifold learning | |
CN115100012A (en) | Method for calculating walking accessibility of rail transit station | |
CN104463442A (en) | Detection method of town and country construction clustering | |
CN115860520B (en) | Method for evaluating dynamic and static space-time reachability of high-speed rail hub and urban public transportation | |
Li et al. | Mapping highway mobile carbon source emissions using traffic flow big data: a case study of Guangdong Province, China | |
Elleuch et al. | Collection and exploration of GPS based vehicle traces database | |
CN103761881B (en) | A kind of telecommunication flow information analytical approach based on smart mobile phone |
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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190524 |
|
WW01 | Invention patent application withdrawn after publication |