CN110347937A - A kind of taxi intelligent seeks objective method - Google Patents

A kind of taxi intelligent seeks objective method Download PDF

Info

Publication number
CN110347937A
CN110347937A CN201910567942.0A CN201910567942A CN110347937A CN 110347937 A CN110347937 A CN 110347937A CN 201910567942 A CN201910567942 A CN 201910567942A CN 110347937 A CN110347937 A CN 110347937A
Authority
CN
China
Prior art keywords
hot spot
objective
taxi
carrying
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.)
Granted
Application number
CN201910567942.0A
Other languages
Chinese (zh)
Other versions
CN110347937B (en
Inventor
王桐
沈昭晛
张乐君
李升波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin Engineering University
Original Assignee
Harbin Engineering University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Harbin Engineering University filed Critical Harbin Engineering University
Priority to CN201910567942.0A priority Critical patent/CN110347937B/en
Publication of CN110347937A publication Critical patent/CN110347937A/en
Application granted granted Critical
Publication of CN110347937B publication Critical patent/CN110347937B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q50/40—
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/025—Services making use of location information using location based information parameters
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
    • H04W4/48—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for in-vehicle communication

Abstract

The invention discloses a kind of taxi intelligents to seek objective method, include: passenger capacity prediction: being predicted based on passenger capacity of the taxi historical trajectory data to the carrying hot spot region in city, the date similar with same day passenger capacity in historical data is filtered out according to prediction result, the date filtered out is generated into space-time Index；Objective achievement data library is sought in building: establishing the driving time database between objective efficiency data library and hot spot of seeking of carrying hot spot region based on taxi historical trajectory data, described to seek objective efficiency data library include that hot spot seeks objective time, carrying probability and carrying income；The screening of carrying hot spot: the space-time Index generated according to step 1 seeks driving time database between objective efficiency data library and hot spot from filter out the corresponding date in step 2, and go to the objective efficiency of seeking of different hot spots to carry out equilibrium taxi, filter out optimal hot spot region.Method of the invention includes two kinds of online and offline treatment process, can greatly shorten the calculating time of recommendation service.

Description

A kind of taxi intelligent seeks objective method
Technical field
Present invention combination taxi historical trajectory data and real-time traffic information propose a kind of taxi intelligent and seek visitor Method.
Background technique
In recent years, with the fast development of the communication technology and computer technology, intelligent transportation system (Intelligent Transportation System, ITS) favor of many researchers is received, intelligent taxi technology is as intelligent transportation The important component of system receives the extensive concern of researcher.Meanwhile demand of the passenger to taxi is also not only stopped It stays in and waits on taxi, more intelligent matching services are put forward one after another, such as: taxi subscription services, Ride-share service, most Excellent drive route service etc., these services provide the analysis all be unableing to do without to taxi historical trajectory data.Due to daily all A large amount of taxi historical trajectory data can be generated, therefore in order to meet the real-time of each service, quick data are parsed into For the task of top priority.Although big data platform is to increase the important means of analysis efficiency, taxi historical trajectory data is divided Analysis strategy is more crucial, and many researchers use the service recommendation mode in historical track processing mode and line under line to meet The real-time of recommendation.
For taxi and this problem of passenger's intelligent Matching, there are mainly two types of the directions of research, the first is from passenger Angle be that passenger recommends optimal seats reserved for guests or passengers of seeking to set, seek objective efficiency to improve passenger, and provide taxi Expected Arrival Time and It reaches probability.It is for second from the angle of driver is that driver recommends most preferably to seek objective route, in order to which that saves driver seeks objective cost, Carrying hot spot region near taxi is analyzed, analysis taxi goes to the expection of different hot spots to seek the objective time and its seek Objective probability according to above two Factor Selection carrying hot spot and is recommended.All due to the taxi historical data that generates daily Difference leads to that different taxi historical datas is selected to have a great impact the accuracy of recommendation service, at present in the field It is that taxi historical data is divided into working days evidence and nonworkdays data using more mode, however this segmentation side Method is simultaneously not careful enough.There are many kinds of the modes for influencing people's trip, such as the factors such as weather, vacation, time, in order to reasonably select Historical data is taken to realize recommendation service, it should comprehensively consider these factors, in real time, dynamically chooses reasonable historical data, and It is not single consideration a certain kind factor.
Since taxi is during finding passenger, driving experience driver abundant can effectively search out train It stands, the carryings hot spot region such as cinema, and the passengers quantity in these regions is and the general that cannot be single with time change Recommend driver in carrying hot spot region.And the variation of passengers quantity and taxi driver is unconspicuous seeks visitor side in carrying hot spot Case is all hidden in taxi historical trajectory data.Therefore, we integrate treatment process under the line of taxi historical trajectory data With treatment process on Real-time Traffic Information line, proposes TCSFP and seek objective strategy.
Summary of the invention
The purpose of the present invention is to propose to a kind of taxi intelligents to seek objective method, to solve taxi and passenger in city " hardly possible " this problem of matching proposes a kind of based on passenger capacity in conjunction with taxi historical trajectory data and real-time traffic information The taxi of prediction seeks objective strategy (Taxi Cruising Strategy Based on Forecasting Passenger Volume,TCSFP)。
The invention is realized by the following technical scheme: a kind of taxi intelligent seeks objective method, and the taxi intelligent seeks visitor Method the following steps are included:
Step 1: passenger capacity prediction: the carrying based on taxi historical trajectory data to the carrying hot spot region in city Amount is predicted, filters out the date similar with same day passenger capacity in historical data according to prediction result, and the date filtered out is raw At space-time Index；
Step 2: objective achievement data library is sought in building: establishing seeking for carrying hot spot region based on taxi historical trajectory data Driving time database between objective efficiency data library and hot spot, described to seek objective efficiency data library include that hot spot seeks objective time, carrying Probability and carrying income；
Step 3: the corresponding date screening of carrying hot spot: is filtered out from step 2 according to the space-time Index that step 1 generates Seek driving time database between objective efficiency data library and hot spot, and to taxi go to different hot spots seek objective efficiency carry out it is equal Weighing apparatus filters out optimal hot spot region, shown in screening principle such as formula (1):
Wherein i indicates that taxi is currently located zone number, and j indicates j-th of hot spot region number, and m indicates daily m A period,The time required to going to j-th of hot spot for taxi,Visitor is sought in j-th of hot spot by taxi to take Between, pjmFor the carrying probability of j-th of hot spot, rjmFor the average carrying income of j-th of hot spot.
Further, in step 1 the following steps are included:
Step 1 one: all passenger loading points are extracted from historical data, and to the carrying time that different time sections occur Number is counted, as shown in formula (2):
Wherein, hiIndicate i-th of hot spot region, djIndicate the jth day in taxi historical data, one day is divided by L expression L period,Indicate the passengers quantity occurred in k-th of period；
Step 1 two: hotspot is collectediAll passenger loading datas in the historical data, as shown in formula (3):
M(hi)={ P (hi,d1),P(hi,d2),...,P(hi,dn)}
(3)
Assuming that the passenger capacity of prediction m-th of period of i-th of hot spot, and use the passenger capacity of preceding k period as pre- Feature is surveyed, machine learning feature and label as shown in formula (4) and formula (5) can be constructed,
Step 1 three: according to feature provided by formula (4) and formula (5) and label training machine learning classification model. And the passenger capacity close with prediction result is searched out from formula (5)Choose the historical data point of m-th of period of jth day Analysis hot spot to seek objective index most reasonable.
Further, in step 2, specifically, rightpjk, rjkQuantified, quantized result such as formula (6) shown in:
WhereinWithIt respectively seeks the objective time, seek objective probability and the quantized result of carrying income, WithValue between 0 and k-1,
It usesInstead of the quantized result for seeking the objective timeThen there are formula (7):
Selection target is changed to findWithBiggish hot spot.
Further, in step 3 the following steps are included:
Step 3 one: being arranged weight for each hot spot, the weight of hot spot be carrying hot spot seek the objective time, seek objective probability and The sum of carrying income:
Step 3 two: the weight of carrying hot spot is ranked up according to descending, and the result of sequence is recorded are as follows:
S(hi, m) and=(weight1,weight2,...,weightl) (9)
Edge hot spot is filtered out, and by the edge focus recommendation filtered out to taxi driver.
The beneficial effects of the present invention are: the present invention recommends the best angle for finding passenger's strategy from the taxi for empty driving Start with, in conjunction with the passenger capacity information on taxi historical trajectory data and line, proposes a kind of taxi based on passenger capacity prediction Vehicle seek objective strategy (Taxi Cruising Strategy Based on Forecasting Passenger Volume, TCSFP), which is divided into two stages: passenger capacity forecast period and taxi hot spot screening stage.It is predicted in passenger capacity Stage can generate space-time Index, dynamically choose historical trajectory data, can push away in the carrying focus recommendation stage for taxi It recommends and seeks objective hot spot.The strategy includes two kinds of online and offline treatment process simultaneously, can greatly shorten the calculating of recommendation service Time.
Detailed description of the invention
Fig. 1 is the method flow diagram that a kind of taxi intelligent of the invention seeks objective method；
Fig. 2 is that carrying hot spot screens schematic diagram.
Specific embodiment
Technical solution in the embodiment of the present invention that following will be combined with the drawings in the embodiments of the present invention carries out clear, complete Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based on this Embodiment in invention, every other reality obtained by those of ordinary skill in the art without making creative efforts Example is applied, shall fall within the protection scope of the present invention.
Shown in referring to Fig.1, the invention is realized by the following technical scheme: a kind of taxi intelligent seeks objective method, it is described go out Hire a car intelligence seek objective method the following steps are included:
Step 1: passenger capacity prediction: the carrying based on taxi historical trajectory data to the carrying hot spot region in city Amount is predicted, filters out the date similar with same day passenger capacity in historical data according to prediction result, and the date filtered out is raw At space-time Index；
Step 2: objective achievement data library is sought in building: establishing seeking for carrying hot spot region based on taxi historical trajectory data Driving time database between objective efficiency data library and hot spot, described to seek objective efficiency data library include that hot spot seeks objective time, carrying Probability and carrying income；
Step 3: the corresponding date screening of carrying hot spot: is filtered out from step 2 according to the space-time Index that step 1 generates Seek driving time database between objective efficiency data library and hot spot, and to taxi go to different hot spots seek objective efficiency carry out it is equal Weighing apparatus filters out optimal hot spot region, shown in screening principle such as formula (1):
Wherein i indicates that taxi is currently located zone number, and j indicates j-th of hot spot region number, and m indicates daily m A period,The time required to going to j-th of hot spot for taxi,Visitor is sought in j-th of hot spot by taxi to take Between, pjmFor the carrying probability of j-th of hot spot, rjmFor the average carrying income of j-th of hot spot.
Specifically, being realized by the prediction to passenger capacity in step 1 and dynamically choosing historical data from database Processing result；And in step 2, by constructing database, provide each hot spot seeks objective efficiency and driving time, accelerates step Rapid three recommendation efficiency.Under the line of the comprehensive taxi historical trajectory data of the present invention in treatment process and Real-time Traffic Information line Reason process proposes TCSFP and seeks objective strategy, as shown in Figure 1.TCSFP seeks objective strategy and is divided into two kinds of online and offline treatment process, Treatment process in Fig. 1 more than dotted line is treatment process under taxi historical data line, and dotted line treatment process below is on line Recommendation process.Two aspects are divided into the processing of historical data, left-hand broken line frame represents passenger capacity forecast period, and right side represents The carrying focus recommendation stage.Taxi seek the objective time, carrying probability and carrying income be screen carrying hot spot it is very crucial because Element, it is very huge to the influence for seeking objective performance, and also these factors are real-time changes, it is therefore desirable to the fast of these factors Speed obtains and overall merit.
Referring to Fig.1 shown in, in the preferred embodiment of this part, in step 1 the following steps are included:
Step 1 one: all passenger loading points are extracted from historical data, and to the carrying time that different time sections occur Number is counted, as shown in formula (2):
Wherein, hiIndicate i-th of hot spot region, djIndicate the jth day in taxi historical data, one day is divided by L expression L period,Indicate the passengers quantity occurred in k-th of period；
Step 1 two: hotspot is collectediAll passenger loading datas in the historical data, as shown in formula (3):
M(hi)={ P (hi,d1),P(hi,d2),...,P(hi,dn)}
(3)
Assuming that the passenger capacity of prediction m-th of period of i-th of hot spot, and use the passenger capacity of preceding k period as pre- Feature is surveyed, machine learning feature and label as shown in formula (4) and formula (5) can be constructed,
Step 1 three: according to feature provided by formula (4) and formula (5) and label training machine learning classification model. And the passenger capacity close with prediction result is searched out from formula (5)So m-th of period of jth day historical data Passenger capacity and prediction result most proximity, therefore the historical data analysis hot spot for choosing m-th of period of jth day seeks objective index most Rationally.
Specifically, the trip due to people is influenced by many factors, such as weather, festivals or holidays etc., therefore to hire out Vehicle searching is most preferably sought during objective hot spot, selects historical data appropriate to be analyzed extremely crucial, the history that the present invention takes Data screening method is the information of daily different time sections passenger capacity to be extracted based on historical trajectory data, and construct prediction model, And the passenger loading data for obtaining real-time traffic information obtains the prediction result of passenger capacity as prediction data.From history number The date similar with the passenger capacity result of prediction is filtered out in, then having reason the passenger's trip for illustrating the date and phase today Seemingly, with generating space-time Index according to the date filtered out, and step 3 is inputed to, the foundation of historical data is chosen as step 3.
Shown in referring to Fig.1, in the preferred embodiment of this part, in step 2, specifically, since taxi is from start bit Set hiGo to carrying hot spot hjSeek the objective time, seek objective probability and carrying income be respectively different type index, it is therefore desirable to it is rightpjk, rjkQuantified, shown in quantized result such as formula (6):
WhereinWithIt respectively seeks the objective time, seek objective probability and the quantized result of carrying income, and Its value between 0 and k-1,
Should be small as far as possible due to seeking the objective time, and seek objective probability and carrying income should height as far as possible, therefore use Can still it guarantee instead of the quantized result for seeking the objective timeBetween 0 and k-1, and selection target is changed to seek It looks forWithBiggish hot spot then has formula (7):
As shown in Fig. 2, should select as far as possible, to find out three reference axis of distance farther away " edge hot spot ", delete apart from far point compared with Close " internal hot spot ", therefore delete hot spot h2。
Specifically, since the recommendation that taxi seeks objective scheme needs to meet real-time, to the place of historical trajectory data It lower online ought to complete.The carrying probability of taxi and carrying income can directly correspond to the carrying probability and load of each hot spot Visitor's income, is relatively easily extracted from historical data, and taxi is sought the time required to the objective time goes to different hot spots by driver Seek objective time composition with being averaged for each hot spot, therefore the attribute of hot spot is divided by the present invention: hot spot to seek objective time, carrying general Rate and carrying income.Time needed for taxi goes to hot spot can obtain from the driving time database between hot spot, taxi The objective time of seeking be that taxi is gone to the time required to hot spot and being averaged for hot spot seeks the sum of objective time.This step, which belongs under line, to be handled Process.
Referring to Fig.1 shown in, in the preferred embodiment of this part, in step 3 the following steps are included:
Step 3 one: being arranged weight for each hot spot, the weight of hot spot be carrying hot spot seek the objective time, seek objective probability and The sum of carrying income:
Step 3 two: the weight of carrying hot spot is ranked up according to descending, and the result of sequence is recorded are as follows:
S(hi, m) and=(weight1,weight2,...,weightl) (9)
Step 3 three: to the result S (h of sequencei, m) in each element judge whether it is edge hot spot:
1. due to weight1For weight limit, it can be deduced that its affiliated hot spot is edge hot spot.It proves as follows:
If hot spot haWith weight limit, and haBelong to " inside " hot spot, then certainly existing hot spot hbSeek the objective time, It seeks objective probability and carrying income is all larger than hot spot ha, therefore hot spot hbWeight weight (hi,hb, k) and it is greater than haWeight weight(hi,ha, k), this and haWith weight limit contradiction, therefore haFor edge hot spot, card is finished.
2. if weight2Place hot spot seeks the objective time, seeks objective probability and carrying income respectively less than weight1Place hot spot, So weight2Place hot spot is internal hot spot；Otherwise weight2Place hot spot is edge hot spot, it was demonstrated that as follows:
If weight2Place hot spot is internal hot spot, then there are hot spot hbSeek the objective time, seek objective probability and carrying receive Enter to be all larger than weight2Place hot spot, therefore hbWeight be greater than weight2.And it is greater than weight2Weight there was only weight1, This and hypothesis test, card are finished.
And so on, to S (hi, k) in each element, be compared with the edge hot spot selected, judge one by one, sieve Edge hot spot is selected, and by the edge focus recommendation filtered out to taxi driver.
Specifically, as can be seen from Figure 1 taxi seeks visitor's request only one step of experience, under other treatment processes are online Or fulfil ahead of schedule, taxi can greatly be shortened and seek visitor's request the time it takes.

Claims (4)

1. a kind of taxi intelligent seeks objective method, which is characterized in that the taxi intelligent seek objective method the following steps are included:
Step 1: passenger capacity prediction: made based on passenger capacity of the taxi historical trajectory data to the carrying hot spot region in city Prediction, filters out the date similar with same day passenger capacity in historical data according to prediction result, when the date filtered out is generated Empty Index；
Step 2: objective achievement data library is sought in building: seeking objective effect based on what taxi historical trajectory data established carrying hot spot region Driving time database between rate database and hot spot, described to seek objective efficiency data library include that hot spot seeks objective time, carrying probability It is taken in carrying；
Step 3: seeking for corresponding date the screening of carrying hot spot: is filtered out from step 2 according to the space-time Index that step 1 generates Driving time database between objective efficiency data library and hot spot, and go to the objective efficiency of seeking of different hot spots to carry out equilibrium taxi, Optimal hot spot region is filtered out, shown in screening principle such as formula (1):
Wherein i indicates that taxi is currently located zone number, and j indicates j-th of hot spot region number, when m indicates daily m-th Between section,The time required to going to j-th of hot spot for taxi,It is taxi the time required to j-th of hot spot seeks visitor, pjmFor the carrying probability of j-th of hot spot, rjmFor the average carrying income of j-th of hot spot.
2. a kind of taxi intelligent according to claim 1 seeks objective method, which is characterized in that include following step in step 1 It is rapid:
Step 1 one: extracting all passenger loading points from historical data, and the carrying times that different time sections are occurred into Row statistics, as shown in formula (2):
Wherein, hiIndicate i-th of hot spot region, djIndicate the jth day in taxi historical data, L indicates to be divided into L for one day Period,Indicate the passengers quantity occurred in k-th of period；
Step 1 two: hotspot is collectediAll passenger loading datas in the historical data, as shown in formula (3):
M(hi)={ P (hi,d1),P(hi,d2),...,P(hi,dn)} (3)
Assuming that the passenger capacity of prediction m-th of period of i-th of hot spot, and use the passenger capacity of preceding k period special as prediction Sign, can construct machine learning feature and label as shown in formula (4) and formula (5),
Step 1 three: according to feature provided by formula (4) and formula (5) and label training machine learning classification model.And from Formula searches out the passenger capacity close with prediction result in (5)Choose the historical data analysis heat of m-th of period of jth day Point to seek objective index most reasonable.
3. a kind of taxi intelligent according to claim 1 seeks objective method, which is characterized in that in step 2, specifically, It is rightpjk, rjkQuantified, shown in quantized result such as formula (6):
WhereinWithIt respectively seeks the objective time, seek objective probability and the quantized result of carrying income, WithValue between 0 and k-1,
It usesInstead of the quantized result for seeking the objective timeThen there are formula (7):
Selection target is changed to findWithBiggish hot spot.
4. a kind of taxi intelligent according to claim 1 seeks objective method, which is characterized in that include following step in step 3 It is rapid:
Step 3 one: weight is set for each hot spot, the weight of hot spot is seeking the objective time, seeking objective probability and carrying for carrying hot spot The sum of income:
Step 3 two: the weight of carrying hot spot is ranked up according to descending, and the result of sequence is recorded are as follows:
S(hi, m) and=(weight1,weight2,...,weightl) (9)
Edge hot spot is filtered out, and by the edge focus recommendation filtered out to taxi driver.
CN201910567942.0A 2019-06-27 2019-06-27 Intelligent passenger searching method for taxi Active CN110347937B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910567942.0A CN110347937B (en) 2019-06-27 2019-06-27 Intelligent passenger searching method for taxi

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910567942.0A CN110347937B (en) 2019-06-27 2019-06-27 Intelligent passenger searching method for taxi

Publications (2)

Publication Number Publication Date
CN110347937A true CN110347937A (en) 2019-10-18
CN110347937B CN110347937B (en) 2022-02-11

Family

ID=68176772

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910567942.0A Active CN110347937B (en) 2019-06-27 2019-06-27 Intelligent passenger searching method for taxi

Country Status (1)

Country Link
CN (1) CN110347937B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112052405A (en) * 2020-08-24 2020-12-08 杭州电子科技大学 Passenger searching area recommendation method based on driver experience
CN112070529A (en) * 2020-08-24 2020-12-11 贵州民族大学 Passenger carrying hotspot parallel prediction method, system, terminal and computer storage medium

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001022777A (en) * 1999-07-09 2001-01-26 Hitachi Ltd Data analyzing method
US6456207B1 (en) * 2001-02-20 2002-09-24 John Yen Intelligent taxi total service system
US20060034201A1 (en) * 2004-07-28 2006-02-16 Nobutoshi Umeda Taxi dispatching system and dispatching method
US20070061301A1 (en) * 2005-09-14 2007-03-15 Jorey Ramer User characteristic influenced search results
US20070061302A1 (en) * 2005-09-14 2007-03-15 Jorey Ramer Location influenced search results
CN103544834A (en) * 2013-11-14 2014-01-29 孙林 Taxi customer seeking strategy selection method based on GPS track
CN103854472A (en) * 2012-12-05 2014-06-11 深圳先进技术研究院 Taxi cloud-intelligent scheduling method and system
US20150012310A1 (en) * 2012-02-26 2015-01-08 Zhe Shen Taxi Calling System and Method Based on Handheld Mobile Terminals
CN107590175A (en) * 2017-07-31 2018-01-16 同济大学 A kind of taxi pickup point based on deep learning recommends method
CN107832882A (en) * 2017-11-03 2018-03-23 上海交通大学 A kind of taxi based on markov decision process seeks objective policy recommendation method

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001022777A (en) * 1999-07-09 2001-01-26 Hitachi Ltd Data analyzing method
US6456207B1 (en) * 2001-02-20 2002-09-24 John Yen Intelligent taxi total service system
US20060034201A1 (en) * 2004-07-28 2006-02-16 Nobutoshi Umeda Taxi dispatching system and dispatching method
US20070061301A1 (en) * 2005-09-14 2007-03-15 Jorey Ramer User characteristic influenced search results
US20070061302A1 (en) * 2005-09-14 2007-03-15 Jorey Ramer Location influenced search results
US20150012310A1 (en) * 2012-02-26 2015-01-08 Zhe Shen Taxi Calling System and Method Based on Handheld Mobile Terminals
CN103854472A (en) * 2012-12-05 2014-06-11 深圳先进技术研究院 Taxi cloud-intelligent scheduling method and system
CN103544834A (en) * 2013-11-14 2014-01-29 孙林 Taxi customer seeking strategy selection method based on GPS track
CN107590175A (en) * 2017-07-31 2018-01-16 同济大学 A kind of taxi pickup point based on deep learning recommends method
CN107832882A (en) * 2017-11-03 2018-03-23 上海交通大学 A kind of taxi based on markov decision process seeks objective policy recommendation method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
刘丽: "基于出租车GPS数据的高效益寻客推荐策略研究", 《中国优秀硕士学位论文全文数据库工程科技||辑》 *
王贝贝: "基于北京市载客热点区的出租车出行需求研究", 《中国优秀硕士学位论文全文数据库工程科技||辑》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112052405A (en) * 2020-08-24 2020-12-08 杭州电子科技大学 Passenger searching area recommendation method based on driver experience
CN112070529A (en) * 2020-08-24 2020-12-11 贵州民族大学 Passenger carrying hotspot parallel prediction method, system, terminal and computer storage medium

Also Published As

Publication number Publication date
CN110347937B (en) 2022-02-11

Similar Documents

Publication Publication Date Title
CN107680401B (en) Parking space satisfaction evaluation method based on interval two-type fuzzy set
CN107423837A (en) The Intelligent planning method and system of tourism route
CN108153867B (en) User trajectory prediction technique and device based on temporal regularity
CN107919014B (en) Taxi running route optimization method for multiple passenger mileage
TWI393378B (en) Hotspot analysis systems and methods, and computer program products thereof
Shen et al. Dynamic ridesharing
Mendes-Moreira et al. Improving the accuracy of long-term travel time prediction using heterogeneous ensembles
CN110347937A (en) A kind of taxi intelligent seeks objective method
CN106991727B (en) Highway intercommunication charging method and system
CN114037175B (en) Large-scale public transportation network hierarchical optimization method based on multi-scale clustering
CN109934401A (en) A kind of customization public transport method based on mobile interchange
CN107992865A (en) A kind of vehicle identification method and system based on video analysis
Popkova et al. Digitalisation in transport and logistics: A roadmap for entrepreneurship in Russia
CN110836675A (en) Decision tree-based automatic driving search decision method
CN110348969A (en) Taxi based on deep learning and big data analysis seeks objective policy recommendation method
CN110796317A (en) Urban taxi scheduling method based on demand prediction
Guan et al. A unified framework for predicting kpis of on-demand transport services
CN109800908A (en) Signalized intersections operating status prediction technique and system based on LSTM model
CN114253975B (en) Load-aware road network shortest path distance calculation method and device
CN112287241B (en) Travel recommendation method and system
CN110084402A (en) A kind of public transport self-adapting dispatching method preferably to be tracked with ant based on website
Faghih et al. Predicting short-term Uber demand using spatio-temporal modeling: A New York City Case Study
Beraldi et al. A Decision support system for trip tourism recommendation
Wang et al. Taxi-cruising recommendation via real-time information and historical trajectory data
CN112989194B (en) Recommendation method and system integrating user request and service preference of Internet of vehicles

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